Dual-Licensing Linux Kernel with GPL V2 and GPL V3
Quoted 11 lines · On Thu, 14 Jun 2007 19:45:08 -0400 Bill Nottingham <[email protected]> wrote:
Quoted 3 lines · Alexandre Oliva
And since the specific implementation involves creating a derived work of the GPLed kernel (the signature, or the signed image, or what have you)
Wait, a signed filesystem image that happens to contain GPL code is now a derived work? Under what sort of interpretation does *that* occur?
(This pretty much throws the 'aggregation' premise in GPLv2 completely out.)
Perhaps the FSF will in future remember to pack a copy of the GPL in each of its md5sum files on the mirror if this is a derivative work, and modify the bittorrent protocol to include a copy of the GPL in the seed files 8)
Alan
Quoted 7 lines
Perhaps the FSF will in future remember to pack a copy of the GPL in each of its md5sum files on the mirror if this is a derivative work, and modify the bittorrent protocol to include a copy of the GPL in the seed files 8)
Alan
I realize you're joking, but for the benefit of anyone who might not understand how this works:
A derivative work must, first and foremost, be a work. If it's not a work, it's not a derivative work because a derivative work is a type of work. Aggregations of multiple works, when that aggregation is performed in an automated way, are not works. Even if I compile and link a bunch of source code, provided there is no creative input in the compile and link process, the result is not a work for copyright purposes. It is simply an aggregate of the bits of source code. The gist of a compilation or derivative work is the creative selection and modification process.
If someone argues that a program is a derivative work of a header file it was compiled with, he is probably just being sloppy. The resulting executable contains the header file combines with other works. Of course, a source code file that is designed based on the contents of a header file may be a derivative work of that header file, but that would be because the human who wrote the source code file used bits of the header file in the source code itself. It would not be because the compiler combined them -- such an automated combination has no creative input and so cannot produce a work, and so cannot produce a derivative work.
This is grossly oversimplified, but should give you the idea.
DS
Quoted 8 lines · Bill Nottingham
Quoted 3 lines · Alexandre Oliva
And since the specific implementation involves creating a derived work of the GPLed kernel (the signature, or the signed image, or what have you)
Wait, a signed filesystem image that happens to contain GPL code is now a derived work? Under what sort of interpretation does *that* occur?
Is the signature not derived from the bits in the GPLed component, as much as it is derived from the key?
Isn't the signature is a functional portion of the image, i.e., if I take it out from the system, it won't work any more?
Quoted 2 lines
(This pretty much throws the 'aggregation' premise in GPLv2 completely out.)
Not really. It could take some explicit distinguishing between functional and non-functional signatures, but that's about it.
GPLv3 chose a different path to make this clarification.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · Alexandre Oliva
Quoted 3 lines
Wait, a signed filesystem image that happens to contain GPL code is now a derived work? Under what sort of interpretation does *that* occur?
Is the signature not derived from the bits in the GPLed component, as much as it is derived from the key?
Quoted 8 lines
Isn't the signature is a functional portion of the image, i.e., if I take it out from the system, it won't work any more?
Quoted 2 lines
(This pretty much throws the 'aggregation' premise in GPLv2 completely out.)
Not really. It could take some explicit distinguishing between functional and non-functional signatures, but that's about it.
OK. Let's take this to the simple and logical conclusion. A signed filesystem image containing both GPL and non-GPL code. From your point A, this is a derived work.
Let's read the license...
2. b) You must cause any work that ... is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
...
But when you
distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.
and yet later:
In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License.
Pick one. They can't both be valid.
Moreover, this interpretation means that Red Hat (and pretty much any other Linux distributor) should close up shop, as that's what we've been doing for years.
Bill
Quoted 3 lines · Bill Nottingham
OK. Let's take this to the simple and logical conclusion. A signed filesystem image containing both GPL and non-GPL code. From your point A, this is a derived work.
I claim the signature is derived from the GPLed bits, yes. Whether that's a derived work, in the legal sense, I'm not qualified to say.
And I claim that, in the case of TiVO, it is not only a functional piece of the system that's derived from GPLed code and missing the corresponding sources, but also it's being used to impose restrictions on the exercise of the freedoms that the GPL is designed to protect. And these conditions are what make it a bad thing, and that deviate, if not from the legal conditions, at least from the spirit of the license.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 10 lines · Alexandre Oliva
Subject: Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
Quoted 3 lines · Bill Nottingham
OK. Let's take this to the simple and logical conclusion. A signed filesystem image containing both GPL and non-GPL code. From your point A, this is a derived work.
I claim the signature is derived from the GPLed bits, yes. Whether that's a derived work, in the legal sense, I'm not qualified to say.
it's also derived from the non-GPLed bits as well.
so if it were a derived work in a legal sense (nessasary for your argument to have any legal meaning) then it's now illegal to make and distribute a checksum of a CD that contains software with incompatible licenses.
Quoted 7 lines
And I claim that, in the case of TiVO, it is not only a functional piece of the system that's derived from GPLed code and missing the corresponding sources, but also it's being used to impose restrictions on the exercise of the freedoms that the GPL is designed to protect. And these conditions are what make it a bad thing, and that deviate, if not from the legal conditions, at least from the spirit of the license.
you keep claiming this, but other people claim you are wrong. what good do your claims do (why are your claims about what's in the spirit of the license and what's not any more valid than anyone else's?)
David Lang
Quoted 13 lines · [email protected] writes:
Quoted 10 lines · Alexandre Oliva
Subject: Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
Quoted 3 lines · Bill Nottingham
OK. Let's take this to the simple and logical conclusion. A signed filesystem image containing both GPL and non-GPL code. From your point A, this is a derived work.
I claim the signature is derived from the GPLed bits, yes. Whether that's a derived work, in the legal sense, I'm not qualified to say.
Quoted 6 lines
it's also derived from the non-GPLed bits as well.
so if it were a derived work in a legal sense (nessasary for your argument to have any legal meaning) then it's now illegal to make and distribute a checksum of a CD that contains software with incompatible licenses.
It is not necessary for the end item to be a derived work in order for the GPL to apply. A literal copy is not a derived work; a translation is not a derived work; an executable version of a program is not a derived work of its source code; and so forth.
What is necessary is that the "work based on the [GPLed] Program" be more than a mere aggregation of the GPLed component(s) with non-GPLed components. The fact that part of the work-as-a-whole is a descriptor of the GPLed part does not mean all descriptions the GPLed part are governed by the GPL. The critical factor is that the GPLed part will not function properly without the DRM signature.
Michael Poole
Quoted 6 lines · On Thursday 14 June 2007 22:25:57 Alexandre Oliva wrote:
Quoted 5 lines · Bill Nottingham
Quoted 3 lines · Alexandre Oliva
And since the specific implementation involves creating a derived work of the GPLed kernel (the signature, or the signed image, or what have you)
Quoted 6 lines
Quoted 3 lines
Wait, a signed filesystem image that happens to contain GPL code is now a derived work? Under what sort of interpretation does *that* occur?
Is the signature not derived from the bits in the GPLed component, as much as it is derived from the key?
Actually, you can't copyright, trademark, or patent a number. In order to copyright something it has to have some creative element. You also can't copyright (or trademark) book titles. So no, last I checked you can't copyright an MD5sum or SHA1sum.
I vaguely recall somebody dredging around for the smallest thing there was a legal precedent explicitly affirming you could copyright it, and it was a haiku. So they put an uncompressed ascii haiku in their protocol...
Now if you sign the executable binary, then the binary (as a whole) is a derivative work of your copyrighted code etc. ad nauseum pluribus unum and so on. And THAT is due to Apple vs Franklin in 1983:
http://www.internetlegal.com/impactof.htm
Before which copyright was only guaranteed to apply to source code, not to binaries (which are basically big numbers). That's why everybody distributed source code before then: it was the only thing they knew you could enforce a copyright on...
IBM's "Object code only" initiative happened around the same time...
http://landley.net/history/mirror/ibm/oco.html
Along with the AT&T breakup commercializing Unix, the launch of the GNU project, and the general rise of "shrinkwrap" software.
(There's this marvelous book called "Legal battles that shaped the computer industry" by Lawrence D. Graham, devotes a few pages to Apple vs Franklin. Franklin honestly didn't think Apple's binary ROMs were copyrightable. Just as in a 1980 interview with Bill Gates, he couldn't stop somebody from printing a book with an annotated printout of the TRS-80 ROMs Microsoft had a copyright to. He sounded kind of pissed about it, actually. Also young and whiny:
Transcript:
http://slashdot.org/features/00/01/20/1316236.shtml
Audio:
http://landley.net/history/mirror/ms/gates.mp3
)
Rob
Signature · 2 lines
"One of my most productive days was throwing away 1000 lines of code." - Ken Thompson.
Quoted 7 lines · Rob Landley
Quoted 6 lines · On Thursday 14 June 2007 22:25:57 Alexandre Oliva wrote:
Quoted 5 lines · Bill Nottingham
Quoted 3 lines · Alexandre Oliva
And since the specific implementation involves creating a derived work of the GPLed kernel (the signature, or the signed image, or what have you)
Quoted 8 lines
Quoted 6 lines
Quoted 3 lines
Wait, a signed filesystem image that happens to contain GPL code is now a derived work? Under what sort of interpretation does *that* occur?
Is the signature not derived from the bits in the GPLed component, as much as it is derived from the key?
Actually, you can't copyright, trademark, or patent a number.
Agreed. And this counter-argument of yours is a distraction.
I was careful to not talk about "derived work". Please read it again under this clarification (that I'm pretty sure I'd already made before, but it's getting hard to keep track of everything in this thread ;-)
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 10 lines · On Friday 15 June 2007 15:28:29 Alexandre Oliva wrote:
Quoted 5 lines · Rob Landley
Quoted 2 lines · On Thursday 14 June 2007 22:25:57 Alexandre Oliva wrote:
Is the signature not derived from the bits in the GPLed component, as much as it is derived from the key?
Actually, you can't copyright, trademark, or patent a number.
Agreed. And this counter-argument of yours is a distraction.
I was careful to not talk about "derived work".
"Is the signature not derived from X as much as it is derived from Y."
"I was careful to not talk about "derived work"."
Which personality of yours am I currently addressing?
Quoted 4 lines
Please read it again
under this clarification (that I'm pretty sure I'd already made before, but it's getting hard to keep track of everything in this thread ;-)
I'm going to stop feeding the troll now...
Rob
Signature · 2 lines
"One of my most productive days was throwing away 1000 lines of code." - Ken Thompson.
Quoted 10 lines · Rob Landley
Quoted 8 lines · On Friday 15 June 2007 15:28:29 Alexandre Oliva wrote:
Quoted 5 lines · Rob Landley
Quoted 2 lines · On Thursday 14 June 2007 22:25:57 Alexandre Oliva wrote:
Is the signature not derived from the bits in the GPLed component, as much as it is derived from the key?
Actually, you can't copyright, trademark, or patent a number.
Agreed. And this counter-argument of yours is a distraction.
Quoted 7 lines
Quoted 1 line
I was careful to not talk about "derived work".
"Is the signature not derived from X as much as it is derived from Y."
"I was careful to not talk about "derived work"."
Which personality of yours am I currently addressing?
The one that speaks English, not Legalese. IANAL.
Last I looked it up, "derived" was a plain-English word.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}On Thursday 14 June 2007 17:39:32 Alexandre Oliva wrote:
<snip>
Quoted 7 lines
And since the specific implementation involves creating a derived work of the GPLed kernel (the signature, or the signed image, or what have you) and refraining from providing the corresponding sources to that derived work (the key and the signature "build scripts"), I still think this specific case is a violation of the letter of the GPLv2, even if the FSF doesn't take this position.
Not entirely correct. If TiVO is making a change to the binary to include the signature, then it *could* be considered a derivative work. If the signature is stored in another place - say a bit of Flash or a separate file on the disc - then there is no way for it to be considered a derivative work. (Under US law, IIRC and I I've interpreted it (and the related cases) properly then the change would have to be to the source of the program for it to be considered a "derivative". But, as you say often and I should make clear myself, IANAL)
Quoted 6 lines
Quoted 3 lines
It seems pretty obvious that the only right Tivo is withholding is the right to install new versions on the device
Actually, no. They withhold the right to run versions that they don't authorize themselves.
And this is relevant to a software license in which way? In particular how is this relevant to the GPL, which has always *only* guaranteed access to the source if you have access to the binary, the right to distribute your own versions and the right to modify the code.
Since the "right to run code" was never guaranteed it *cannot* be a violation. It might be in conflict with what RMS intended when he wrote the first version of the GPL and in conflict with the intent of the people that contributed to GPLv2 but that doesn't matter. However, I will not use (or recommend) the GPLv3 in its current form because I feel it makes unnecessary restrictions. The fact that you have to "allow additional rights" to make it equal to the GPLv2 makes a functional (and spiritual) difference to me.
(Why? Because I'm opposed to "In order to protect freedom X we have to restrict freedom Y. Its happening in the US *RIGHT* *NOW* and I have been doing what I can to fight that. Now the same faulty logic is being applied by the FSF with GPLv3)
Quoted 5 lines
Back when GPLv2 was written, the right to run was never considered an issue. It was taken for granted, because copyright didn't control that in the US (it does in Brazil), and nobody had thought of technical measures to stop people from running modified copies of software. At least nobody involved in GPLv2, AFAIK.
Why isn't it in the US? Because the binary form of a program does not and cannot have a separate copyright than the source code. Since it is the *SOURCE* that is actually copyright (mechanical translation cannot create a new work, only a new form of an already copyrighted work) guaranteeing the "right to run" is pointless.
And you are wrong about that "Nobody thought of it" thing - what you mean is that "Nobody that had a hand in drafting and ratifying the GPLv2 thought of it". The "NSA Guidelines for Trusted Systems" (aka "The Orange Book") talks about methods of preventing the execution of code.
What you and the rest of the FSF is doing in response to "tivoization" is saying "we don't care if it wasn't designed to do X, we want to be able to make it do that anyway *AND* the manufacturer has to help us do it". There is no legal way for you to make that demand of a hardware manufacturer. Instead you've gone with the only legal recourse - saying "If you want to use my copyrighted work under license X, you have to do Y". I have no problem with that, and if the FSF wants that, it's fine by me. But, as I said, I could care less where and/or if something I release under the GPL is used. This makes the GPLv2 perfect for me.
Quoted 9 lines
The landscape has changed, and GPLv3 is meant to defend this freedom that was taken for granted.
Quoted 2 lines
they never do (and really never could) "modify" the physical copy on your device (which is your main argument).
Qualifying it as the main argument is a bit of an exaggeration. I have a number of different arguments. The one about incomplete sources is the most solid IMHO.
Yes, it is. But your argument about the TiVO is "they can modify the copy on it but I can't". Hence it is your main argument. And remember, replace != modify.
Quoted 9 lines
Quoted 5 lines
Quoted 2 lines
What do you think you do when you save a modified source file in your editor?
Don't skip the part where the in-memory version started as an exact copy of the original being replaced. Notice the difference? ;)
Sorry, I really don't follow. Both versions of the kernel binary also started from a common source ancestor. Were you trying to make a distinction on these grounds?
No. He was making a distinction that I have seen made a number of times. That is, a copy of a copyright work in a computers RAM is a *distinct* copy, separate from the file on disc it started as. (And that has actually been argued in a court case. I don't recall the specifics, but I do recall that the argument was held as valid)
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 6 lines · Alexandre Oliva
Quoted 3 lines · Florin Malita
No, it's not: replacing does not create derivative work. Modification does.
Quoted 6 lines
Thanks. Good point. This convinces me that this doesn't work as a legal argument under copyright.
I still stand by my understanding that this restriction violates the spirit of the license.
But since this elusive "spirit" is subject to everybody's interpretation of the preamble, you must surely admit that it remains just a matter of opinion ;)
Quoted 7 lines
Quoted 3 lines
It seems pretty obvious that the only right Tivo is withholding is the right to install new versions on the device
Actually, no. They withhold the right to run versions that they don't authorize themselves.
On that particular piece of hw, yes. But who's granted you the right to *run* your modified copy *there* in the first place? GPLv2 explicitly steers clear of anything "other than copying, distribution and modification".
Quoted 9 lines
Back when GPLv2 was written, the right to run was never considered an issue. It was taken for granted, because copyright didn't control that in the US (it does in Brazil), and nobody had thought of technical measures to stop people from running modified copies of software. At least nobody involved in GPLv2, AFAIK.
The landscape has changed, and GPLv3 is meant to defend this freedom that was taken for granted.
Then you agree that GPLv2 does not protect your freedom (taken for granted) to run a modified copy on any particular device, or am I misreading?
Quoted 5 lines
Quoted 3 lines
Quoted 3 lines
What do you think you do when you save a modified source file in your editor?
Quoted 8 lines
Quoted 3 lines
Don't skip the part where the in-memory version started as an exact copy of the original being replaced. Notice the difference? ;)
Sorry, I really don't follow. Both versions of the kernel binary also started from a common source ancestor. Were you trying to make a distinction on these grounds?
Exactly: they have a common ancestor, they are both derived from it. But there's no ancestry relationship *between* them (unlike your edited file example) so you cannot argue that one is a modification of the other. Hence, Tivo is not really *modifying* the copies it distributes with the device - they're *installing* brand new copies instead. They also choose not to offer everybody the same privilege :-|
Does this go against the intent of the GPLv2 authors? Probably. Does it go against the letter of GPLv2? Apparently not. Does it go against your/some people's interpretation of the GPL "spirit"? Obviously. Does it go against everybody's interpretation? Obviously not.
---
fm
Quoted 7 lines · Florin Malita
Quoted 5 lines · Alexandre Oliva
Back when GPLv2 was written, the right to run was never considered an issue. It was taken for granted, because copyright didn't control that in the US (it does in Brazil), and nobody had thought of technical measures to stop people from running modified copies of software. At least nobody involved in GPLv2, AFAIK.
Quoted 6 lines
Quoted 2 lines
The landscape has changed, and GPLv3 is meant to defend this freedom that was taken for granted.
Then you agree that GPLv2 does not protect your freedom (taken for granted) to run a modified copy on any particular device, or am I misreading?
IANAL, but AFAICT it doesn't. Still, encoded in the spirit (that refers to free software, bringing in the free software definition), is the notion of protecting users' freedoms, among them the freeom #0, to run the software for any purpose.
That's why I believe it's in the spirit of the license to defend this freedom.
And that's why lawyers in Brazil believe that, even though the GPL does not affirm the right to run the software, it fits the bill, because, under the light of the preamble, the free software definition, and the US copyright law, it should be interpreted as an intent to grant permission to run the software.
Quoted 3 lines
Hence, Tivo is not really *modifying* the copies it distributes with the device - they're *installing* brand new copies instead. They also choose not to offer everybody the same privilege :-|
Got it. That's bad. :-(
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 11 lines · On Thursday 14 June 2007 23:19:24 Alexandre Oliva wrote:
Quoted 10 lines · Florin Malita
Quoted 8 lines · Alexandre Oliva
Back when GPLv2 was written, the right to run was never considered an issue. It was taken for granted, because copyright didn't control that in the US (it does in Brazil), and nobody had thought of technical measures to stop people from running modified copies of software. At least nobody involved in GPLv2, AFAIK.
The landscape has changed, and GPLv3 is meant to defend this freedom that was taken for granted.
Quoted 8 lines
Quoted 3 lines
Then you agree that GPLv2 does not protect your freedom (taken for granted) to run a modified copy on any particular device, or am I misreading?
IANAL, but AFAICT it doesn't. Still, encoded in the spirit (that refers to free software, bringing in the free software definition), is the notion of protecting users' freedoms, among them the freeom #0, to run the software for any purpose.
And where in GPLv2 is "Freedom #0"?
As a simple matter of fact, the *only* activities covered by the GPLv2 are "copying, distributing and modifying". It says so in the license itself.
Quoted 8 lines
That's why I believe it's in the spirit of the license to defend this freedom.
And that's why lawyers in Brazil believe that, even though the GPL does not affirm the right to run the software, it fits the bill, because, under the light of the preamble, the free software definition, and the US copyright law, it should be interpreted as an intent to grant permission to run the software.
Then they have made a bad decision. While it can be argued that "the right to run the software" is guaranteed, the truth is that the license is very clear about what it covers. That's *DIRECTLY* in section 0 of the license. If someone has interpreted it to cover something besides what it explicitly states then it has been badly interpreted.
In case you don't remember, GPLv2, section 2, paragraph 2:
"Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does."
In other words, the license cannot be sanely interpreted to cover *execution* of the program. Yes, it says that the *license* doesn't restrict you from running the program, but that *DOESN'T* matter, because the opening sentence
says: "Activities other than copying, distribution and modification are not covered by this License; they are outside its scope." QED: The intent of the license is clear and it is to guarantee those three stated rights.
DRH
Quoted 5 lines
Quoted 3 lines
Hence, Tivo is not really *modifying* the copies it distributes with the device - they're *installing* brand new copies instead. They also choose not to offer everybody the same privilege :-|
Got it. That's bad. :-(
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 2 lines
As a simple matter of fact, the *only* activities covered by the GPLv2 are "copying, distributing and modifying". It says so in the license itself.
Unless I have explicitly installed linux myself in the box, I have received the binary from them, so it can fall in the distribution case.
Signature · 5 lines
Glauber de Oliveira Costa. "Free as in Freedom" http://glommer.net "The less confident you are, the more serious you have to act."
Quoted 7 lines · On Friday 15 June 2007 02:29:32 Glauber de Oliveira Costa wrote:
Quoted 3 lines
As a simple matter of fact, the *only* activities covered by the GPLv2 are "copying, distributing and modifying". It says so in the license itself.
Unless I have explicitly installed linux myself in the box, I have received the binary from them, so it can fall in the distribution case.
Sorry if you missed the rest of the discussion, but the above statement was a rebuttal of the "The GPLv2 intended to guarantee me the right to run the software on any given piece of hardware" argument that has been used as the justification for the addition of the "tivoization" language to the GPLv3. As I stated, I fail to see how "running" the program is, in any way, intended by the license, since it *explicitly* states that it only covers "copying, distribution and modification". The exact place where it does that is "Section 0, paragraph 2, first sentence". I'll quote it here again:
"Activities other than copying, distribution and modification are not covered by this License; they are outside its scope."
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 8 lines · Daniel Hazelton
Quoted 4 lines · On Thursday 14 June 2007 23:19:24 Alexandre Oliva wrote:
IANAL, but AFAICT it doesn't. Still, encoded in the spirit (that refers to free software, bringing in the free software definition), is the notion of protecting users' freedoms, among them the freeom #0, to run the software for any purpose.
And where in GPLv2 is "Freedom #0"?
It may sound like thin evidence for someone arriving from Venus today, but the preamble talks about "free software", some passages clearly imply that software under this license is "free software", the license is published by the Free Software Foundation, and the Free Software Foundation has a published definition of Free Software that establishes the 4 freedoms.
The freedoms defined there resonate very strongly with the freedoms/rights that the license talks about. I hope this is enough evidence to convince you that this is the intent.
The only of the freedoms that's not explicitly mentioned in the preamble, the freedom to run the software for any purpose, is mentioned in the legal terms as unrestricted, which is very much in line with freedom #0, but is outside the scope of a copyright license because running the program does not require copyright permission.
I'll give you that the preamble doesn't make it clear that the license is purported to defend freedom #0 too.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · On Friday 15 June 2007 15:49:00 Alexandre Oliva wrote:
Quoted 6 lines · Daniel Hazelton
Quoted 4 lines · On Thursday 14 June 2007 23:19:24 Alexandre Oliva wrote:
IANAL, but AFAICT it doesn't. Still, encoded in the spirit (that refers to free software, bringing in the free software definition), is the notion of protecting users' freedoms, among them the freeom #0, to run the software for any purpose.
Quoted 8 lines
Quoted 1 line
And where in GPLv2 is "Freedom #0"?
It may sound like thin evidence for someone arriving from Venus today, but the preamble talks about "free software", some passages clearly imply that software under this license is "free software", the license is published by the Free Software Foundation, and the Free Software Foundation has a published definition of Free Software that establishes the 4 freedoms.
And that doesn't matter. In the context of the GPLv2 the only legally active parts *ARE* in the GPLv2, under this heading:
GNU GENERAL PUBLIC LICENSE
TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
No other text published by the FSF has any legal bearing on the GPLv2 *except* when it is the FSF that holds copyright to the work and has placed it under the GPLv2. When I place a program under the GPL, it becomes *my* interpretation of the license and those "published texts" I might have (that are relevant to the situation) which have bearing on the license.
Quoted 3 lines
The freedoms defined there resonate very strongly with the freedoms/rights that the license talks about. I hope this is enough evidence to convince you that this is the intent.
Nope. Because the intent of the author of the license is worth nothing. The intent of the person who has placed the code under the GPL is, however, worth quite a bit.
Quoted 5 lines
The only of the freedoms that's not explicitly mentioned in the preamble, the freedom to run the software for any purpose, is mentioned in the legal terms as unrestricted, which is very much in line with freedom #0, but is outside the scope of a copyright license because running the program does not require copyright permission.
In the context of the license it means that the copyright holder is placing no restrictions. And that is because they legally *cannot* - as you agreed. But the copyright holder *can*, OTOH, make it clear that they wish for there to be no limits, artificial or otherwise, placed on the code.
Quoted 2 lines
I'll give you that the preamble doesn't make it clear that the license is purported to defend freedom #0 too.
And the preamble carries no legal weight.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 10 lines · Daniel Hazelton
Quoted 9 lines · On Friday 15 June 2007 15:49:00 Alexandre Oliva wrote:
Quoted 7 lines · Daniel Hazelton
Quoted 4 lines · On Thursday 14 June 2007 23:19:24 Alexandre Oliva wrote:
IANAL, but AFAICT it doesn't. Still, encoded in the spirit (that refers to free software, bringing in the free software definition), is the notion of protecting users' freedoms, among them the freeom #0, to run the software for any purpose.
And where in GPLv2 is "Freedom #0"?
Quoted 8 lines
Quoted 6 lines
It may sound like thin evidence for someone arriving from Venus today, but the preamble talks about "free software", some passages clearly imply that software under this license is "free software", the license is published by the Free Software Foundation, and the Free Software Foundation has a published definition of Free Software that establishes the 4 freedoms.
And that doesn't matter.
Doens't matter for what?
To indicate what the Linux copyright holders meant? Sure it doesn't. I never claimed it did.
To indicate what the authors of the GPL meant? To indicate the spirit of the license they wrote? Yes, it matters a lot.
And the latter is what my participation here is all about: to show that the spirit didn't change at all.
Until you acknowledge and understand this, I should refrain from answering your other postings.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · Alexandre Oliva
Quoted 2 lines
What the GPL *does* say is that you can't "add additional restrictions to the license"
Not quite. It's more general than that:
You may not impose any further restrictions on the recipients' exercise of the rights granted herein.
GPLv2 section 0 says:
Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, ...
The license does not cover running of the program. It doesn't restrict it, but it doesn't cover it. Claiming otherwise is turning the GPL into yet another dreaded EULA.
Nowhere does the GPLv2 define modification as "modify and run in place". The Preamble emphasizes sharing; hardware is a fixed object and can't be shared in the same fashion as software.
Also, GPLv2 section 2 includes:
In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License.
TiVo's firmware (and any restrictions it may carry) is not affected by the GPLv2.
Quoted 7 lines
Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, ...
The license does not cover running of the program. It doesn't restrict it, but it doesn't cover it. Claiming otherwise is turning the GPL into yet another dreaded EULA.
For many juridisctions loading from disk into memory is copying and in some from memory to CPU cache a second copy. This is one reason as I understand it GPLv3 talks about "conveying" - to avoid that mess and confusion.
Quoted 7 lines
In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License.
TiVo's firmware (and any restrictions it may carry) is not affected by the GPLv2.
Really irrelevant to the discussion. Tivo's firmware is up to them. Whether the resulting system permits them to include GPLv2 software with it is what matters.
Alan
Quoted 4 lines · Alan Cox
Quoted 4 lines
Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, ...
Quoted 8 lines
Quoted 3 lines
The license does not cover running of the program. It doesn't restrict it, but it doesn't cover it. Claiming otherwise is turning the GPL into yet another dreaded EULA.
For many juridisctions loading from disk into memory is copying and in some from memory to CPU cache a second copy. This is one reason as I understand it GPLv3 talks about "conveying" - to avoid that mess and confusion.
Hmm... This is interesting. Let me sidetrack a little bit.
Who would be held liable should the copy not be authorized by the copyright holder? The designer of the hardware? The seller? The person who powerer the computer on? The author of the boot loader (if I'm talking about the kernel about to be loaded).
Quoted 3 lines
Really irrelevant to the discussion. Tivo's firmware is up to them. Whether the resulting system permits them to include GPLv2 software with it is what matters.
+1
It all boils down to whether they're keeping their promise to not impose further restrictions on the exercise of the rights granted in the license.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 1 line · Alexandre Oliva
On Jun 14, 2007, Alan Cox <[email protected]> wrote:
[...]
Quoted 10 lines
Quoted 4 lines
For many juridisctions loading from disk into memory is copying and in some from memory to CPU cache a second copy. This is one reason as I understand it GPLv3 talks about "conveying" - to avoid that mess and confusion.
Hmm... This is interesting. Let me sidetrack a little bit.
Who would be held liable should the copy not be authorized by the copyright holder? The designer of the hardware? The seller? The person who powerer the computer on? The author of the boot loader (if
The person copying the software (read: the poor user) - of course with no doubt.
Is the manufacturer of a knife or rifle or car or brick responsible if I kill someone with it?
Quoted 1 line
I'm talking about the kernel about to be loaded).
And to solve the above "legal" problem, (at least in .at) it is explicitly legal (and stated in the local law) to "copy" software for personal use as long as you don't pass it on to others. And this includes (of course) e.g. backup copies on DVDs (and remember, you are not allowed to give them away, just store them in your desk).
Bernd
Signature · 3 lines
Firmix Software GmbH http://www.firmix.at/
mobil: +43 664 4416156 fax: +43 1 7890849-55
Embedded Linux Development and ServicesQuoted 6 lines · Alexandre Oliva
People don't get your copy, so they're not entitled to anything about it.
When they download the software, they get another copy, and they have a right to modify that copy.
Umm. I notice how you must have known how *idiotic* your response was, because you snipped away the part where I talked about Red Haty distributing CD-ROM's.
In other words, Red Hat distributes copies (and yes, you *get* that copy), and you cannot modify that copy that you got.
So the "right to make changes" _must_ be separate from the actual copy of the image.
And don't get fooled by the "all the rights that you have". That _obviously_ and clearly talks about "the program", which in turn equally obviously and clearly has to be about something bigger than the "one copy", since the GPLv2 requires you have the right to change it.
So you edited out the part where I talked about CD's. That's the proof that your reading is untenable, because obviously you cannot change the program on the CD: you got a copy, but the right to make modifications wasn't ON THAT HARDWARE.
Quoted 5 lines
Quoted 3 lines
And here's a big clue for people: anybody who thinks that I'm violating the GPLv2 by not giving out my private SSH key to master.kernel.org is a f*cking moron!
Agreed, except I'd probably use a lighter term.
Hey, I'm not exactly known for being polite. I tell it how I see it, and I tend to be pretty damn blunt about it.
Quoted 5 lines
Quoted 2 lines
See any parallels here? Any parallel to a CD-ROM distribution, or a Tivo distribution?
Yes. You see how TiVO is different? It is modifyable, and I actually receive the copy that TiVO can still modify, but I can't.
You keep on harping on that "modifyable", but no-where in the GPLv2 is that an issue. I claim that it *cannot* be an issue, since CD's are obviously ok.
So the "modifyable" part is a totally new thing to the GPLv3.
You cannot use that as an argument that the GPLv3 didn't change things, that's a circular agument: "the GPLv3 says so, so thus the GPLv3 is in the same spirit as the GPLv2". Doesn't make sense.
The fact is, the GPLv3 does fundamentally new things. Things I didn't sign up for (and things that nobody _else_ signed up for either) when I chose the GPLv2 for the kernel.
The fact that some people would like to change the kernel license to GPLv3 is no different from the fact that some other people would like to cgange the kernel license to the BSD license.
Those people who have argued for using the BSD license, btw, argued so in the name of "freedom". No different from you. Do you think they were right? If so, why the hell do you think _you_ are right?
So here's what it fundamentally boils down to:
- do you admit that the GPLv3 is a new license that does new and different things?
- do you admit that I chose the GPLv2, and have argued that I chose it because I understood what it said?
- do you admit that authors have the right to choose their own licenses?
And here's the fundamental answer:
- if you answered "no" to any of the above questions, you're either stupid (the first two questions) or a douche-bag (the third one)
- if you answered "yes" to all the above questions, HOW THE HELL can you call me confused, and argue against me when I say that the GPLv2 is a better license? It wasn't your choice.
It really is that easy.
Linus
Quoted 8 lines · Linus Torvalds
Quoted 6 lines · Alexandre Oliva
People don't get your copy, so they're not entitled to anything about it.
When they download the software, they get another copy, and they have a right to modify that copy.
Quoted 6 lines
Umm. I notice how you must have known how *idiotic* your response was, because you snipped away the part where I talked about Red Haty distributing CD-ROM's.
In other words, Red Hat distributes copies (and yes, you *get* that copy), and you cannot modify that copy that you got.
And Red Hat can't either. I thought that was quite obvious.
Quoted 9 lines
Quoted 5 lines
Quoted 2 lines
See any parallels here? Any parallel to a CD-ROM distribution, or a Tivo distribution?
Yes. You see how TiVO is different? It is modifyable, and I actually receive the copy that TiVO can still modify, but I can't.
You keep on harping on that "modifyable", but no-where in the GPLv2 is that an issue. I claim that it *cannot* be an issue, since CD's are obviously ok.
The 'passing on the rights you have' makes it an issue.
Quoted 1 line
You cannot use that as an argument that the GPLv3 didn't change things,
Compare the preambles of v2 and v3 and you'll understand why the argument is sound, and not circular.
Quoted 2 lines
Those people who have argued for using the BSD license, btw, argued so in the name of "freedom".
But individual freedom, rather than community freedom.
Think local vs global optimization.
Quoted 1 line
If so, why the hell do you think _you_ are right?
Because, like you, I'm always right, even though not everyone agrees with that assessment? ;-P :-D
Quoted 2 lines
- do you admit that the GPLv3 is a new license that does new and different things?
Yes, of course. The new legal terms are answers to new threats to the freedoms depicted in the preamble, that didn't exist or hadn't been thought of by the time GPLv2 was published.
Quoted 2 lines
- do you admit that I chose the GPLv2, and have argued that I chose it because I understood what it said?
Yes.
Quoted 1 line
- do you admit that authors have the right to choose their own licenses?
Within the boundaries of ethics and morals, yes.
Quoted 3 lines
- if you answered "yes" to all the above questions, HOW THE HELL can you call me confused, and argue against me when I say that the GPLv2 is a better license? It wasn't your choice.
The thing is I'm not arguing that point.
I'm disputing that there was a change in the spirit of the license between v2 and v3. Heck, a mere 48 hours ago you couldn't even tell the spirit from the legal terms.
I still think v3 will serve better any Free Software community, because it will push away the abusers that contribute little, or turn them into cooperative or at least harmless participants, that further enable the active participation of their downstream users. This would enable wider participation under the same 'tit-for-tat' conditions that you attribute to GPLv2.
It appears to me that the only significant point of contention remaining is the issue of Tivoization. If you feel so strongly about permitting Tivoization, even though it denies the freedoms that the original spirit of the license you chose says they are entitled to have, you can make this provision by means of an additional permission for that, on top of GPLv3, and be done with it.
Quoted 1 line
It really is that easy.
Yes, indeed.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · Alexandre Oliva
Quoted 3 lines
In other words, Red Hat distributes copies (and yes, you *get* that copy), and you cannot modify that copy that you got.
And Red Hat can't either. I thought that was quite obvious.
That's TOTALLY IRRELEVANT!
There is no language in the GPLv2 (only in the GPLv3 drafts) about "same upgradability as third parties".
You're arguing a point that DOES NOT EXIST in the GPLv2.
The GPLv2 talks about specific rights, like the ability to make changes and distribute things, and says that you have to give downstream all those same rights.
And I've pointed out to you (now about five times) that those rights CANNOT be able "in-place", since even Red Hat does not actually give you the right to do in-place modification of the software they sell.
Quoted 1 line
The 'passing on the rights you have' makes it an issue.
No. It does not.
I have extra rights as a copyright holder, and that "the rights you have" are as they pertain to the software under the GPLv2, not as it pertains to the physical device, or outside the GPLv2.
For example, for any code that I have full copyright over, I have rights that you DO NOT HAVE! I have the right to re-license it under some other license. The fact that I pass on a copy of the software to you under the GPLv2 does *not* give you those rights, but that's not even what the GPLv2 asks for!
The GPLv2, when it talks about "passing on the rights", talks about the rights you got *per*the*GPLv2*.
Any other reading is nonsensical, since the copyrigth owner *always* has more rights than a licensee! I legally literally *couldn't* pass over all the rights I have to my software! If you read the GPLv2 as meaning that I have to, you are mis-reading it. It's that simple.
Anyway, I'm not interested in continuing this flame war.
The fact is, the license for the kernel is the GPLv2. And I think it's a superior license. As such, I'd be a total moron to relicense the kernel under what I believe is a worse license.
So if you want to argue that I should re-license, you should argue that the GPLv3 is better. And quite frankly, you haven't.
Linus
Quoted 10 lines · Linus Torvalds
Quoted 5 lines · Alexandre Oliva
Quoted 3 lines
In other words, Red Hat distributes copies (and yes, you *get* that copy), and you cannot modify that copy that you got.
And Red Hat can't either. I thought that was quite obvious.
The GPLv2 talks about specific rights, like the ability to make changes and distribute things, and says that you have to give downstream all those same rights.
The spirit gives the intuition of "passing on all the rights". The legal terms have to be more careful about that, to avoid the very situation you're debating, so they state "you can't impose further restrictions on the exercise of the rights".
Do you understand the difference?
Quoted 2 lines
For example, for any code that I have full copyright over, I have rights that you DO NOT HAVE!
No dispute about that, and this is irrelevant to this point. I've already responded and clarified this point 2 or 3 times in this thread. Do you need me to find a URL for you? It was in respose to Dmitri Torokhov.
Quoted 2 lines
So if you want to argue that I should re-license, you should argue that the GPLv3 is better. And quite frankly, you haven't.
In fact, I haven't even tried. So far, I've merely been trying to show that it still follows the same spirit, dispelling the muth that it doesn't, and trying to understand why you think GPLv2 is so much better, which I think is related with tit-for-tat and retribution in kind.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 1 line · Daniel Hazelton
Now stop parroting the FSF's worn and tired tripe.
Are you playing Linus' sheeple and parroting his lines just to make a point, or are you like that all the time? ;-)
Quoted 2 lines
PS: Looking at your .sig I guess maybe you can't do that without getting kicked out of the FSF-LA
Don't worry, I'm not speaking even for FSFLA, and I'm entitled to my own opinion.
I haven't consulted other FSFLA members about this. This is all my own personal opinion.
It just so happens that I'm very closely involved in the process, I've spent a lot of time thinking about it, and I happen to share a similar moral and ethical background with others involved in the process, so I arrive at similar conclusions.
And then, I influence the process myself, so it's not like some of the arguments I brought up here weren't taken into account while creating the GPLv3, and adopted by its other proponents.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · On Wednesday 13 June 2007 21:04:42 Alexandre Oliva wrote:
Quoted 1 line · Daniel Hazelton
Now stop parroting the FSF's worn and tired tripe.
Are you playing Linus' sheeple and parroting his lines just to make a point, or are you like that all the time? ;-)
Nope. I'm just tired of giving proof after proof that you're wrong and having you restate the same tripe.
Quoted 5 lines
Quoted 2 lines
PS: Looking at your .sig I guess maybe you can't do that without getting kicked out of the FSF-LA
Don't worry, I'm not speaking even for FSFLA, and I'm entitled to my own opinion.
Certainly. I never said otherwise. What I stated and then *implied* was that you are repeating the same false logic over and over again trying to make people believe that it isn't borked and that that false logic is exactly the same crap I've seen from the FSF numerous times.
Quoted 2 lines
I haven't consulted other FSFLA members about this. This is all my own personal opinion.
Where I am examining the facts and drawing a logical conclusion. That it happens to form an opinion is secondary to the truth.
Quoted 4 lines
It just so happens that I'm very closely involved in the process, I've spent a lot of time thinking about it, and I happen to share a similar moral and ethical background with others involved in the process, so I arrive at similar conclusions.
Okay. Still doesn't explain why you have argued that the GPLv3 doesn't attempt to cover hardware and then provide proof that it does.
Quoted 3 lines
And then, I influence the process myself, so it's not like some of the arguments I brought up here weren't taken into account while creating the GPLv3, and adopted by its other proponents.
This is no surprise - I had a feeling this was the truth. Not that it changes my opinion at all. As I've said, I have never liked the GPL at all, but v2 is the best that exists - even though I've put together custom licenses myself, none of them have had the number of lawyers look at them that the GPLv2 has had.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 2 lines · Daniel Hazelton
Still doesn't explain why you have argued that the GPLv3 doesn't attempt to cover hardware and then provide proof that it does.
It doesn't cover hardware, in the same way that it doesn't cover patents, and it doesn't cover pro-DRM laws. It merely arranges, as best as we've managed a copyright license to do, that they can't be used as excuses (or tools) to disrespect the freedoms that the GPL demands all licensees to respect for other users.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 9 lines · On Wednesday 13 June 2007 22:04:04 Alexandre Oliva wrote:
Quoted 2 lines · Daniel Hazelton
Still doesn't explain why you have argued that the GPLv3 doesn't attempt to cover hardware and then provide proof that it does.
It doesn't cover hardware, in the same way that it doesn't cover patents, and it doesn't cover pro-DRM laws. It merely arranges, as best as we've managed a copyright license to do, that they can't be used as excuses (or tools) to disrespect the freedoms that the GPL demands all licensees to respect for other users.
Consider this scenario:
Small company A is manufacturing a new WiFi router.
They decide to have it run HURD as the OS.
In complying with the GPLv3 they supply the signing keys and everything else needed to install a new kernel on the hardware.
User B buys the router and modifies the kernel so it drives the WiFi to an output power twice that which it is licensed to carry.
FCC finds out and prosecutes User B for violating the regulations. FCC then pulls the small companies license until they change their hardware so the driver can't push it to transmit at a higher power level and levies a fine.
Small company A loses the money paid on the fine, has to recall all the devices that can be modified (through software) to break the law at a massive cost *AND* has to redesign their hardware. The total cost drives the company into bankruptcy.
Small companies C,D and E, in order to avoid the fate of small company A, purchases a license for proprietary OS "F" to drive their new hardware.
Net loss: A lot of the users and publicity that "Free Software" used to get, because GPLv3 contains language that opens the companies to lawsuits that they wouldn't otherwise face.
Which is better: Growing the base of installed GPL covered software, or "ethics and morals" that demand the language that has been added to the GPLv3 ? Personally I'd like to see proprietary software driven into a very small "niche" market or entirely out of existence. However much I want this to happen, I cannot be anything *BUT* scared of the GPLv3 simply because I see it creating massive problems - and all because of a *small* portion of the new language it contains. It has taken almost 15 years for "Free Software" to make a dent in the market, and, IMHO, a lot of that is both Linux and the "holes" in GPLv2.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 3 lines · Daniel Hazelton
User B buys the router and modifies the kernel so it drives the WiFi to an output power twice that which it is licensed to carry.
FCC finds out and prosecutes User B for violating the regulations.
Ok so far.
Quoted 3 lines
FCC then pulls the small companies license until they change their hardware so the driver can't push it to transmit at a higher power level and levies a fine.
I'd say this is unfair, but if it can happen, then maybe the small company could have been more careful about the regulations. There are various ways to prevent these changes that don't involve imposing restrictions of modification on any software in the device, all the way from hardware-constrained output power to hardware-verified authorized configuration parameters.
Quoted 1 line
Growing the base of installed GPL covered software,
When this doesn't bring freedom to people, when people can't actually enjoy the freedoms that the software is supposed to provide, I don't see why this would be a good thing. What's the merit in being able to claim "vendor X chose my Free Software and locked it down such that users don't get the freedoms I meant for them, and I'm happy about it?"
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}On Thu 14 Jun 2007 01:07, Alexandre Oliva pondered:
Quoted 7 lines · Daniel Hazelton
Quoted 2 lines
User B buys the router and modifies the kernel so it drives the WiFi
to an
Quoted 2 lines
output power twice that which it is licensed to carry. FCC finds out and prosecutes User B for violating the regulations.
Quoted 7 lines
Ok so far.
Quoted 3 lines
FCC then pulls the small companies license until they change their hardware so the driver can't push it to transmit at a higher power level and levies a fine.
I'd say this is unfair, but if it can happen,
it can (at least hypothetically in the US - I do not know of any actual cases - anyone?).
Quoted 6 lines
then maybe the small
company could have been more careful about the regulations. There are various ways to prevent these changes that don't involve imposing restrictions of modification on any software in the device, all the way from hardware-constrained output power to hardware-verified authorized configuration parameters.
As a person pretty familiar with the hardware in these types of devices - this just isn't practical.
The power output goes to the air, has some to do with any external power amp, the antenna design/connectors, the layout of the printed circuit board (PCB), the materials of the PCB, and numerous other factors that are outside the chip where the register sits that controls the "output power". This is why the register is there - so many things exist in the real world that require it to be changed to actually get desired power output in the air.
What you are asking is possible - that people who make the standard CMOS radios add EEPROM or FLASH to their processes - or they could add an interface to a serial EEPROM where these hardware limits could sit - and during manufacturing could program these to add a hardware limit.
But this seems - just silly - adding extra cost/complexity to the hardware - because you limit things in the zero-cost space?
Or run the driver in user space, and forget about it. (Like everyone is starting to do on the desktop).
Quoted 5 lines
When this doesn't bring freedom to people, when people can't actually enjoy the freedoms that the software is supposed to provide, I don't see why this would be a good thing. What's the merit in being able to claim "vendor X chose my Free Software and locked it down such that users don't get the freedoms I meant for them, and I'm happy about it?"
Sometimes joy comes from knowing that your contribution was valuable.
I would ask about two other examples:
- medical devices - which must have their software certified by the FDA (at
least in the US) - and the manufacture can not allow non-certified software to be loaded on it - and would be classified "designed or sold for incorporation into a dwelling" (per the GPL3 dd1v3). Think devices like :
http://www.powerheart.com/products/ [1]
Are Linux developers "happy" knowing their contributions will never help save a life? And this joy goes to Microsoft or VxWorks? Today it can go to Linux developers, since the GPL2 license doesn't restrict this like the GPL3 does.
- gambling devices - which must have their software certified by various
government agencies - to make sure that the odds are known, and there are no backdoors, and consumers don't get screwed - the manufacture can not allow non-certified software to be loaded on it. If these are in a hotel - where various people live - is that considered "incorporation into a dwelling"?
Not wanting to start a debate about the morality of being involved in the gambling industry - (if the statically challenged are giving the government money to keep my taxes down, I am mostly OK with it) - but I'm not "happy" thinking that someone can ledgistate restrictions on embedded OS choice, just because it must be verified by a third party.
Hypothetically - tragedy strikes - use of a modified Nao robot [2] contributes to someone's death (remember - modifications of the OS removed the three laws). Someone (maybe Microsoft or VxWorks, or GreenHills) steps up, and lobbies the government - that embedded OSes need tighter control to ensure safety - and need to be validated by the government. (like the FCC and FDA is already doing in the US).
Since the GPL3 states" -
"If you cannot convey the Program, or other covered work, so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not convey it at all."
Forget DRM - forget Tivo - anyone who develops a product which must be certified can not use anything that is under the GPL3.
On Sun, 10 Jun 2007 10:29:04 Linus Torvalds pondered:
Quoted 5 lines
I still think GPLv2 is simply the better license.
I consider dual-licensing unlikely (and technically quite hard), but at least _possible_ in theory. I have yet to see any actual *reasons* for licensing under the GPLv3, though.
I think there are actual reasons _not_ to.
What I have told people who have asked me (who are thinking about putting Linux in a medical device) is kernel version 2.6.xx (current version at the time), is released under GPL2. No one can go back and change the license on a previously released package.
It might be easier to say that the kernel licence will only be reviewed during a major bump in the kernel version number (3.x.x) and that all 2.x kernels will be under the same license as what exists today.
-Robin
[1] - I don't know if this device has Linux in it or not - but this just represents classes of devices that may never be able to use Linux.
Quoted 11 lines · Robin Getz
On Thu 14 Jun 2007 01:07, Alexandre Oliva pondered:
Quoted 6 lines
then maybe the small
company could have been more careful about the regulations. There are various ways to prevent these changes that don't involve imposing restrictions of modification on any software in the device, all the way from hardware-constrained output power to hardware-verified authorized configuration parameters.As a person pretty familiar with the hardware in these types of devices - this just isn't practical.
I actually left out the most obvious one: store the program in ROM. Is that not practical?
You're claiming that adding hardware locks and chains and bolts, implemented with help from the loader software, is simpler than just using ROM?
Well, then, ok: do all that loader and hardware signature-checking dancing, sign the image, store it in the machine, and throw the signing key away. This should be good for the highly-regulated areas you're talking about. And then, since you can no longer modify the program, you don't have to let the user do that any more. Problem solved.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 12 lines · On Thursday 14 June 2007 13:46:40 Alexandre Oliva wrote:
Quoted 10 lines · Robin Getz
On Thu 14 Jun 2007 01:07, Alexandre Oliva pondered:
Quoted 6 lines
then maybe the small
company could have been more careful about the regulations. There are various ways to prevent these changes that don't involve imposing restrictions of modification on any software in the device, all the way from hardware-constrained output power to hardware-verified authorized configuration parameters.As a person pretty familiar with the hardware in these types of devices - this just isn't practical.
Quoted 6 lines
I actually left out the most obvious one: store the program in ROM. Is that not practical?
You're claiming that adding hardware locks and chains and bolts, implemented with help from the loader software, is simpler than just using ROM?
As far as I know, I'm the first one who brought up the "the current GPLv3 draft forbids burning your code into ROM, you idiots" argument back before Bruce Perens cost the BusyBox project my services over this very issue. (Not that I didn't lock the license of that down to v2 and chase him away before I left, I was just too disgusted to ever again contribute to a project he'd named. Yeah, I hold a grudge.)
Although it's kind of amusing to watch you attempt to dictate terms to hardware manufacturers, the answer to your question is "yes". Having flash is sometimes simpler and cheaper than having ROM. It means you don't have to burn a new mask to bump the firmware revision (especially on a low-volume production run, where "low volume" here is tens or hundreds of thousands instead of millions). It makes the thing a lot more field serviceable (you can upgrade the firmware without a chip puller). It means one physical chip
can give you both read-only and persistent writeable storage. And flash
chips produced in high enough unit volumes honestly can be cheaper than a custom ROM, pricing in semiconductors is all about unit volume.
Quoted 6 lines
Well, then, ok: do all that loader and hardware signature-checking dancing, sign the image, store it in the machine, and throw the signing key away. This should be good for the highly-regulated areas you're talking about. And then, since you can no longer modify the program, you don't have to let the user do that any more. Problem solved.
A) Does that actually satisfy the terms of GPLv3? If so, can't they just wait until they get sued and destroy the keys then?
B) There are actually manufacturers who would be happy with your straw man. Lots of companies in the far east produce products that infringe on patents from 30 different competitors, and rather than try to license everything (which isn't even always possible) they spin off a shell company (or nested series thereof), design and manufacture a product, sell a production run of them into the distribution channel, and then dissolve the shell company before the inventory hits retailers. But the time anybody is in a position to take an enforcement action, the company to take the action against is gone. (Who are you going to sue, customers who bought the devices? The distributors who bought the inventory in good faith, and will then refuse to distribute any of YOUR product if you attack 'em?) There's always the possibility that somebody will sit down and follow the paper trail back to the parent company (through the multiple legal jurisdictions where nobody speaks english), but since they're likely as not to destroy this kind of info _anyway_ while burying their trail...
Rob
Quoted 7 lines
B) There are actually manufacturers who would be happy with your straw man. Lots of companies in the far east produce products that infringe on patents from 30 different competitors, and rather than try to license everything (which isn't even always possible) they spin off a shell company (or nested series thereof), design and manufacture a product, sell a production run of them into the distribution channel, and then dissolve the shell company before the inventory hits retailers. But the time anybody is in a position
This isn't just done for IPR, in fact in many fields IPR is a non-issue. The primary reason for this practice is to render US health and safety regulation irrelevant and to prevent class action suits if/when your device kills someone.
Alan
Quoted 10 lines · Rob Landley
Quoted 6 lines · On Thursday 14 June 2007 13:46:40 Alexandre Oliva wrote:
Well, then, ok: do all that loader and hardware signature-checking dancing, sign the image, store it in the machine, and throw the signing key away. This should be good for the highly-regulated areas you're talking about. And then, since you can no longer modify the program, you don't have to let the user do that any more. Problem solved.
A) Does that actually satisfy the terms of GPLv3?
I think so:
this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product
Quoted 2 lines
If so, can't they just wait until they get sued and destroy the keys then?
I don't think this woulnd't satisfy the above.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}On Thu 14 Jun 2007 13:46, Alexandre Oliva pondered:
Quoted 10 lines · Robin Getz
Quoted 2 lines
As a person pretty familiar with the hardware in these types of devices - this just isn't practical.
Well, then, ok: do all that loader and hardware signature-checking dancing, sign the image, store it in the machine, and throw the signing key away. This should be good for the highly-regulated areas you're talking about. And then, since you can no longer modify the program, you don't have to let the user do that any more. Problem solved.
I don't think so - the GPL3 doesn't state that you must convey the same rights to end users that you have, it says you must provide installation information, including your keys, or you can not ship the product.
That is the way I read the following sections (let me know if I mis-read anything):
======================
"Installation Information" for a User Product means any methods, procedures, authorization keys, or other information required to install and execute modified versions of a covered work in that User Product from a modified version of its Corresponding Source.
The information must suffice to ensure that the continued functioning of the modified object code is in no case prevented or interfered with solely because modification has been made.
If conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License.
If you cannot convey the Program, or other covered work, so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not convey it at all.
===================
I read "are imposed on you (or otherwise)" to mean "by you" as well. If so, you throwing away the private keys are not an option.
I need to think a bit more of Rob's opinion of ROM's are illegal - but providing the installation information of "send $1M NRE and object code to xxx ROM vendor, and wait 16 weeks for 500k units, take one to a board shop, pay $1k for them to re-work your BGA - if the xray says it is screwed up, you have 499,999 other units to try." - may meet the language, but doesn't meet the spirit of the GPL either...
-Robin
Quoted 12 lines · Robin Getz
On Thu 14 Jun 2007 13:46, Alexandre Oliva pondered:
Quoted 10 lines · Robin Getz
Quoted 2 lines
As a person pretty familiar with the hardware in these types of devices - this just isn't practical.
Well, then, ok: do all that loader and hardware signature-checking dancing, sign the image, store it in the machine, and throw the signing key away. This should be good for the highly-regulated areas you're talking about. And then, since you can no longer modify the program, you don't have to let the user do that any more. Problem solved.
Quoted 2 lines
I don't think so - the GPL3 doesn't state that you must convey the same rights to end users that you have,
Right, this is only in the preamble.
Quoted 2 lines
it says you must provide installation information, including your keys, or you can not ship the product.
Unless you throw the keys away:
this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM).
Quoted 1 line
I need to think a bit more of Rob's opinion of ROM's are illegal
See above ;-)
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 11 lines · Robin Getz
- gambling devices - which must have their software certified by various
government agencies - to make sure that the odds are known, and there are no backdoors, and consumers don't get screwed - the manufacture can not allow non-certified software to be loaded on it. If these are in a hotel - where various people live - is that considered "incorporation into a dwelling"?Not wanting to start a debate about the morality of being involved in the gambling industry - (if the statically challenged are giving the government money to keep my taxes down, I am mostly OK with it) - but I'm not "happy" thinking that someone can ledgistate restrictions on embedded OS choice, just because it must be verified by a third party.
Why not go really controversial and dive straight in with "voting machines". There's a whole 'nother can of worms.
Bron.
Quoted 3 lines
the new language it contains. It has taken almost 15 years for "Free Software" to make a dent in the market, and, IMHO, a lot of that is both Linux and the "holes" in GPLv2.
You appear terminally confused. The purpose of the GPL as defined by its authors is not commercial success, world domination or making zillions of dollars - it is keeping the software protected by that license "free" in terms of liberty as measured against the set of freedoms to run/modify/etc they discuss in the licence document.
The fact this is a good license for making zillions of dollars, producing good software and the like is either incidental or a logical result of the protection of freedoms depending upon which views you believe.
Quoted 10 lines · Alan Cox
Quoted 3 lines
the new language it contains. It has taken almost 15 years for "Free Software" to make a dent in the market, and, IMHO, a lot of that is both Linux and the "holes" in GPLv2.
You appear terminally confused. The purpose of the GPL as defined by its authors is not commercial success, world domination or making zillions of dollars - it is keeping the software protected by that license "free" in terms of liberty as measured against the set of freedoms to run/modify/etc they discuss in the licence document.
Quoted 4 lines
The fact this is a good license for making zillions of dollars, producing good software and the like is either incidental or a logical result of the protection of freedoms depending upon which views you believe.
that's fine, but the fundamental question is: where is the moral boundary of the power that the copyright license gives? The FSF seems to believe "nowhere, anything that copyright law allows us to achieve our goals is a fair game" - and the GPLv3 shows that belief. I dont subscribe to that view. I think the proper limit is the boundary where the limit of the software is - because that's the only sane and globally workable way to stop the power-hungry. I.e. the information we produce is covered by the rules of the GPL. It might be used in ways inconvenient to us, it might be put on hardware we dont like (be that a Tivo, a landmine or an abortion instrument) but that does not change the fundamental fact: it's outside the _moral scope_ of our power. Whether some jurisdictions allow the control of _other_ information via our information is immaterial. If a jurisdiction allows the control of hardware that is associated with our software, so what? If a jurisdiction allows the controlling of various aspects of movie theaters that happen to play copyrighted movies, does it make it morally right?
Ingo
Quoted 2 lines
that's fine, but the fundamental question is: where is the moral boundary of the power that the copyright license gives? The FSF seems to
Assuming a democratic state then the laws of the land ought to reflect the 'general will' (if you believe Rousseau anyway). They should thus define the boundary ['derivative work' generally ] according to the general good and with the consent of the people.
Quoted 3 lines
hardware that is associated with our software, so what? If a jurisdiction allows the controlling of various aspects of movie theaters that happen to play copyrighted movies, does it make it morally right?
Does that question not suppose some positivist absolute morality ? I suspect many would argue that it is moral to do so if the end goal of the controls is moral. You might also want to apply the tests in Fuller's Internal Morality of Law ?
I'm not sure Philosophy is on topic for l/k however 8)
Alan
Quoted 8 lines · Alan Cox
Quoted 3 lines
that's fine, but the fundamental question is: where is the moral boundary of the power that the copyright license gives? The FSF seems to
Assuming a democratic state then the laws of the land ought to reflect the 'general will' (if you believe Rousseau anyway). They should thus define the boundary ['derivative work' generally ] according to the general good and with the consent of the people.
uhm, so if the MPAA and the RIAA pays for another nice piece of legislation that extends the power of copyright owners, do you find it morally justified to use those powers, as long as it's argued to be in favor of some long-term goal that you judge to be moral, even if it results in some "temporary injustice"?
i think that could be the main difference in thinking. I argue that the only way to be moral is to be moral _now_, not "later, once this very important fight for the common good is over". I think the moral approach to this is to say _no_ to attempts to extend the license to beyond the "moral scope" of the software we wrote - regardless of what new powers are legislated into the hands of copyright owners. It's naturally hard to do, because giving up power is always hard to do.
In other words: we need to apply our concepts of freedom and fairness not only to the end result, but to the means and methods of achieving those end results as well. The end goals are often forgotten, it's the process that matters to the end result.
Or in yet another set of words: this concept of morality also happens to be expressed fairly accurately in the thousands of years of 'quid pro quo' concept. (shared amongst many, many cultures on this planet, shared amongst far more cultures than the western 'freedom' concept.) (which concept of quid-pro-quo fairness is likely coded into our brains and into our thinking genetically - because it's a simple and very efficient group survival method.)
Ingo
Quoted 10 lines · On Thursday 14 June 2007 08:25:46 Ingo Molnar wrote:
Quoted 8 lines · Alan Cox
Quoted 3 lines
that's fine, but the fundamental question is: where is the moral boundary of the power that the copyright license gives? The FSF seems to
Assuming a democratic state then the laws of the land ought to reflect the 'general will' (if you believe Rousseau anyway). They should thus define the boundary ['derivative work' generally ] according to the general good and with the consent of the people.
Quoted 5 lines
uhm, so if the MPAA and the RIAA pays for another nice piece of legislation that extends the power of copyright owners, do you find it morally justified to use those powers, as long as it's argued to be in favor of some long-term goal that you judge to be moral, even if it results in some "temporary injustice"?
Turnabout is fair play, and unilateral disarmament is a bad strategy in a mexican standoff?
Finding it morally justified to _have_ powers is not the same as finding it morally justified to _use_ powers you have anyway. Lots of companies (like Red Hat) amass defensive software patent portfolios because the patent system is so screwed up.
But then, I'm a pragmatist, not an idealist. You can be one or the other and make it work. Mixing the two tends to suck. Being ruthlessly pragmatic in the pursuit of an ideal (as the FSF seems to be doing) has often been a recipe for disaster...
Rob
Quoted 6 lines · Rob Landley
Quoted 5 lines
uhm, so if the MPAA and the RIAA pays for another nice piece of legislation that extends the power of copyright owners, do you find it morally justified to use those powers, as long as it's argued to be in favor of some long-term goal that you judge to be moral, even if it results in some "temporary injustice"?
Quoted 7 lines
Turnabout is fair play, and unilateral disarmament is a bad strategy in a mexican standoff?
Finding it morally justified to _have_ powers is not the same as finding it morally justified to _use_ powers you have anyway. Lots of companies (like Red Hat) amass defensive software patent portfolios because the patent system is so screwed up.
but the GPLv3 definitely takes action against Tivo. It's not "defensive" in any way. It is outright hostile, it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like. (who, incidentally, is a mathematician who last wrote significant free software perhaps a decade ago, and who thus must have a great and thorough understanding of how hardware and software works today and who must also have a deep knowledge about what makes the free software community tick.)
The GPLv2 never did this kind of restriction _of other works_. Yes, you can use copyright law to control other works and thus (if the affected work is a hardware device for example) to control the _use_ of the free software, but it is _wrong_. The GPLv2 specifically said, in section 0:
Activities other than copying, distribution and modification are not covered by this License; they are outside its scope.
guess why this section has been completely removed from the GPLv3, without a replacement?
Ingo
Quoted 3 lines · Ingo Molnar
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
Quoted 1 line
The GPLv2 never did this kind of restriction
_ofotherworks_.
How about other works in which GPLed software is distributed?
I think your interpretation is mistaken or at least incomplete.
Quoted 5 lines
Activities other than copying, distribution and modification are not covered by this License; they are outside its scope.
guess why this section has been completely removed from the GPLv3, without a replacement?
My guess:
First, because it was redundant, given that the license didn't quite discuss other activities. Unless you count say "imposing restrictions on the exercise of others' freedoms" as other activities, even though these are associated with modification and distribution.
Second, because GPLv3 does indeed talk about other activities, such as starting lawsuits on patent and pro-DRM grounds, or entering agreements for distribution of software along with limited patent licenses. All of these are still associated, at least to some extent, with modification and distribution, but I guess it was worth clarifying that claiming that such harmful activities are outside the scope of the license isn't a valid excuse to escape the conditions determined by the license.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 8 lines · Alexandre Oliva
Quoted 3 lines · Ingo Molnar
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
uhm, so you claim that my argument is false, and your proof for that is a "non-upgradeable Tivo"?? <sarcasm> That is a _great_ idea. Not being able to patch security holes. Not being able to fix bugs. Not being able to add new features. Makes complete sense. Will be a hit on the market! Every PVR maker will flock from Windows to Linux i'm sure. </sarcasm>
really, do you even _read_ what you write? All your arguments so far were instantly debunkable. This is one of the lowest quality GPL discussions i was ever involved in ...
furthermore, the fact that the GPLv3 had to add carved out exceptions for the anti-Tivo languge is further _proof_ that the whole idea is absurd to begin with! It's like writing a nice new function to implement something, and then when it shows many design flaws, you'd not just admit that it's flawed and would get rid of it and redesign it, you'd instead pretend that it's fine and you'd carve out a few of the more common failure modes and would hack it around in that case.
Quoted 6 lines
Quoted 5 lines
Activities other than copying, distribution and modification are not covered by this License; they are outside its scope.
guess why this section has been completely removed from the GPLv3, without a replacement?
Quoted 6 lines
My guess:
First, because it was redundant, given that the license didn't quite discuss other activities. Unless you count say "imposing restrictions on the exercise of others' freedoms" as other activities, even though these are associated with modification and distribution.
here you prove that you cannot even read what i wrote. I wrote that this section has been removed from the GPLv3. What relevance does it have that in your opinion this section was redundant in the GPLv2?? It would clearly not be redundant in the GPLv3: it would contradict and _completely neutralize_ most of the crap from the GPLv3 that we are talking about here ...
Quoted 8 lines
Second, because GPLv3 does indeed talk about other activities, such as starting lawsuits on patent and pro-DRM grounds, or entering agreements for distribution of software along with limited patent licenses. All of these are still associated, at least to some extent, with modification and distribution, but I guess it was worth clarifying that claiming that such harmful activities are outside the scope of the license isn't a valid excuse to escape the conditions determined by the license.
dont you realize that declaring certain types of activities by hardware makers as being "against freedom" is _exactly_ such an activity that the GPLv2 did not attempt to control? I could tell you offhand a dozen more examples of human activities that restrict the 4 GNU freedoms of users _much more_ than the Tivo ever did: for example censure, opression of free speech, out of control climate, dictatorship, campaign financing laws, the WIN32 API and human stupidity. By your argument we'd have to add prohibition against those restrictions of freedom to the license too, right? Your argument still leads to absurd results, even now that you've modified it a few times already ...
Ingo
Quoted 8 lines · Ingo Molnar
Quoted 5 lines · Alexandre Oliva
Quoted 3 lines · Ingo Molnar
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
Quoted 7 lines
Quoted 2 lines
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
uhm, so you claim that my argument is false, and your proof for that is a "non-upgradeable Tivo"?? <sarcasm> That is a _great_ idea. Not being able to patch security holes. Not being able to fix bugs. Not being able to add new features. Makes complete sense.
Oh, so you think patching security holes, fixing bugs and adding new features are good ideas? What if you can't do it in your TiVo?
Quoted 5 lines
Quoted 4 lines
Quoted 2 lines
guess why this section has been completely removed from the GPLv3, without a replacement?
My guess:
Quoted 8 lines
Quoted 4 lines
First, because it was redundant, given that the license didn't quite discuss other activities. Unless you count say "imposing restrictions on the exercise of others' freedoms" as other activities, even though these are associated with modification and distribution.
here you prove that you cannot even read what i wrote. I wrote that this section has been removed from the GPLv3. What relevance does it have that in your opinion this section was redundant in the GPLv2??
If you didn't mean "removed from the GPLv3 as compared with v2", I misunderstood what you wrote.
The fact that it's redundant is v2 means it is reasonable to take it out. That's the relevance.
Quoted 3 lines
It would clearly not be redundant in the GPLv3: it would contradict and
_completelyneutralize_most of the crap from the GPLv3 that we are talking about here ...
And, per the same reasoning, some of the v2 provisions as well.
Quoted 3 lines
dont you realize that declaring certain types of activities by hardware makers as being "against freedom" is _exactly_ such an activity that the GPLv2 did not attempt to control?
No. And some Linux hackers disagree with your assessment too.
Quoted 4 lines
censure, opression of free speech, out of control climate, dictatorship, campaign financing laws, the WIN32 API and human stupidity. By your argument we'd have to add prohibition against those restrictions of freedom to the license too, right?
-ENONSEQUITUR
How do these stop a user's exercise of the four freedoms of a piece of software licensed under the GPL?
Quoted 2 lines
Your argument still leads to absurd results, even now that you've modified it a few times already ...
I hope you're not saying that my listening to you, recognizing mistakes in my arguments and fixing them up is a bad thing.
But hey, at least I'm not modifying my arguments as much as you are! ;-)
It's pretty easy to shoot a straw man and claim the original argument was broken.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · Alexandre Oliva
On Jun 15, 2007, Ingo Molnar <[email protected]> wrote:
How do these stop a user's exercise of the four freedoms of a piece of software licensed under the GPL?
---
I know you don't see it that way, but I still find it bizarre that "the right to modify the software" should be construed as "implies the right to modify the device that the software was shipped in".
I do agree that it's not a change in "spirit" - I'm sure the GPL authors would have disliked TiVoization 15 years ago as much as they do today, if they had thought about it (regardless of the Stallman interview where he said he didn't care very much about devices).
However, whether it is a change "in spirit" or not, it clearly is a qualitative change that substantially changes the rights granted under the license and it's perfectly reasonable for some authors who liked the GPLv2 to dislike and reject GPLv3.
scott
Quoted 32 lines · On Friday 15 June 2007 20:22:50 Alexandre Oliva wrote:
Quoted 6 lines · Ingo Molnar
Quoted 5 lines · Alexandre Oliva
Quoted 3 lines · Ingo Molnar
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
Quoted 7 lines
Quoted 2 lines
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
uhm, so you claim that my argument is false, and your proof for that is a "non-upgradeable Tivo"?? <sarcasm> That is a _great_ idea. Not being able to patch security holes. Not being able to fix bugs. Not being able to add new features. Makes complete sense.
Oh, so you think patching security holes, fixing bugs and adding new features are good ideas? What if you can't do it in your TiVo?
Quoted 5 lines
Quoted 5 lines
Quoted 2 lines
guess why this section has been completely removed from the GPLv3, without a replacement?
My guess:
Quoted 8 lines
Quoted 4 lines
First, because it was redundant, given that the license didn't quite discuss other activities. Unless you count say "imposing restrictions on the exercise of others' freedoms" as other activities, even though these are associated with modification and distribution.
here you prove that you cannot even read what i wrote. I wrote that this section has been removed from the GPLv3. What relevance does it have that in your opinion this section was redundant in the GPLv2??
Quoted 5 lines
If you didn't mean "removed from the GPLv3 as compared with v2", I misunderstood what you wrote.
The fact that it's redundant is v2 means it is reasonable to take it out. That's the relevance.
It isn't redundant at all. I specifies the definitions of several terms used in the GPLv2 and also defines the exact scope of the license. If you feel that the definition of the terms and the limitation of scope were redundant then you are sadly mistaken.
Quoted 5 lines
Quoted 3 lines
It would clearly not be redundant in the GPLv3: it would contradict and
_completelyneutralize_most of the crap from the GPLv3 that we are talking about here ...And, per the same reasoning, some of the v2 provisions as well.
For a license to be legally enforceable it must be internally consistent. Without that internal consistency it becomes very easy to circumvent it. The GPLv2's definitions and defined scope - as per section 0 - define the limits of the license and are entirely consistent with the rest of it. What it *isn't* consistent with is the FSF's other "propaganda" and the wants of the FSF to make certain activities verboten in GPLv3.
Quoted 5 lines
Quoted 3 lines
dont you realize that declaring certain types of activities by hardware makers as being "against freedom" is _exactly_ such an activity that the GPLv2 did not attempt to control?
No. And some Linux hackers disagree with your assessment too.
And that is their right. However, it appears to a nearly unanimous consensus that it is the truth. It may not be liked by some people, but likes and dislikes don't matter.
Quoted 13 lines
Quoted 4 lines
censure, opression of free speech, out of control climate, dictatorship, campaign financing laws, the WIN32 API and human stupidity. By your argument we'd have to add prohibition against those restrictions of freedom to the license too, right?
-ENONSEQUITURHow do these stop a user's exercise of the four freedoms of a piece of software licensed under the GPL?
Quoted 2 lines
Your argument still leads to absurd results, even now that you've modified it a few times already ...
Quoted 8 lines
I hope you're not saying that my listening to you, recognizing mistakes in my arguments and fixing them up is a bad thing.
But hey, at least I'm not modifying my arguments as much as you are! ;-)
It's pretty easy to shoot a straw man and claim the original argument was broken.
Yep. I've done it to you on more than one occasion, Alexandre. The part that makes me laugh is that you still haven't realized it.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 10 lines · Alexandre Oliva
Quoted 7 lines · Ingo Molnar
Quoted 7 lines
Quoted 4 lines
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
Quoted 7 lines
Quoted 4 lines
uhm, so you claim that my argument is false, and your proof for that is a "non-upgradeable Tivo"?? <sarcasm> That is a _great_ idea. Not being able to patch security holes. Not being able to fix bugs. Not being able to add new features. Makes complete sense.
Oh, so you think patching security holes, fixing bugs and adding new features are good ideas? What if you can't do it in your TiVo?
this has to be one of the most bizarre arguments i've read in this thread as of date. Are you seriously questioning the notion that it's a good and legitimate idea for a hardware vendor to make the system fixable, patchable and upgradable? Are you seriously suggesting that for a hardware vendor to be able to offer such a solution, if they are under the unescapable restriction of content providers that the system itself must be tamper-proof, it should not be able to use a GPL-ed kernel at all? Because that is what your arguments lead to, and that is what the GPLv3 implements. In case you didnt notice: RMS _does not want the Tivo to use a GPLv3 kernel_, simple as that, and the GPLv3 achieves that. He wants Tivo to either to go out of business or to go to WinCE or some other OS. Did you ever think about the meaning of the "anti" word in the "anti-Tivo" expression? Hint: it's not some friendly suggestion of cooperation and working together ;)
Ingo
Quoted 17 lines · Ingo Molnar
Quoted 15 lines · Alexandre Oliva
Quoted 12 lines · Ingo Molnar
Quoted 7 lines
Quoted 4 lines
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
uhm, so you claim that my argument is false, and your proof for that is a "non-upgradeable Tivo"?? <sarcasm> That is a _great_ idea. Not being able to patch security holes. Not being able to fix bugs. Not being able to add new features. Makes complete sense.
Quoted 7 lines
Quoted 2 lines
Oh, so you think patching security holes, fixing bugs and adding new features are good ideas? What if you can't do it in your TiVo?
this has to be one of the most bizarre arguments i've read in this thread as of date. Are you seriously questioning the notion that it's a good and legitimate idea for a hardware vendor to make the system fixable, patchable and upgradable?
No. I'm questioning why the vendor could keep this privilege to itself.
Quoted 4 lines
Are you seriously suggesting that for a hardware vendor to be able to offer such a solution, if they are under the unescapable restriction of content providers that the system itself must be tamper-proof, it should not be able to use a GPL-ed kernel at all?
No, and I've already explained how I believe this can be accomplished with the wording in the GPLv3dd4, although IANAL to tell whether that's correct.
Just make the tivoization machinery require two keys: one that the vendor keeps, one that the vendor gives to the user (maybe without ever knowing it). Neither one can install modifications alone, but the user can approve modifications by the vendor, and the vendor can approve modifications by the user. This is still not ideal, but it at least doesn't permit the vendor to remove features from under the user.
Quoted 2 lines
Because that is what your arguments lead to, and that is what the GPLv3 implements.
You haven't really read that bit of dd3 or dd4, have you?
Or the various portions of this thread in which I showed your assumptions are utterly broken?
Quoted 1 line
RMS
_doesnot want the Tivo to use a GPLv3kernel_,
I know you're not stupid, but I can't tell whether you're malicious or just misinformed.
RMS does not want TiVo (or anyone else) to disrespect users' freedoms, and installing technical measures to prevent users from adapting the software to suit their needs and running their modifications is disrespecting users freedoms.
That he is not opposed to the idea of TiVo using a GPLv3 kernel is easy to see, if you take the time to read the draft instead of spreading false assumptions about it:
this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product
Quoted 2 lines
Hint: it's not some friendly suggestion of cooperation and working together ;)
Hey, wouldn't this be just tit-for-tat? ;-)
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · Alexandre Oliva
Just make the tivoization machinery require two keys: one that the vendor keeps, one that the vendor gives to the user (maybe without ever knowing it). Neither one can install modifications alone, but the user can approve modifications by the vendor, and the vendor can approve modifications by the user. This is still not ideal, but it at least doesn't permit the vendor to remove features from under the user.
So what features has Tivo removed (or threatened to remove) from the GPL licensed parts? I think at some point they disabled the _undocumented_ skip feature in their own proprietary software, and ended up reenabling it when a lot of customers complained. Even if those customers had the ability to replace any GPL licensed parts, it would not have reenabled the feature. And it was an undocumented easter egg type thing at that, it isn't like they widely announced it in their advertising or as a selling point.
So Google is using Linux right. What if they remove some feature? (let's pick a randomg one, i.e. phone number lookup) Should I get a keycard for their machine room to fix the problem, or maybe we should use some secret sharing mechanism to prevent them from removing the feature.
Quoted 4 lines
RMS does not want TiVo (or anyone else) to disrespect users' freedoms, and installing technical measures to prevent users from adapting the software to suit their needs and running their modifications is disrespecting users freedoms.
So if Tivo would allow you to boot your own kernel, but keeps the harddisk encrypted if the booted kernel does not have the right signature? In such a case you can run your own kernel and if you replace the harddrive you can install all the applications you might want. You cannot however use their software, any of the recorded content or obtain any further guide data/service updates.
And how is that any different from taking an off-the-shelf PC and booting your own kernel with Tivo's modifications? Or really different from the current situation.
Tivo complied in as far as they made GPL licensed code available, you can examine it, modify it, compile it. You just can't use it _in combination with TiVo's own software and service_. I didn't think the GPLv2 covered anything related to use and you have retained all the freedoms the GPL promises.
Quoted 7 lines
That he is not opposed to the idea of TiVo using a GPLv3 kernel is easy to see, if you take the time to read the draft instead of spreading false assumptions about it:
this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product
So they keep the system locked down, but include perl/python/emacs and distribute updates in the form of scripts/source code which are either interpreted or compiled to a ramfs filesystem at boot. Time to add another exception?
Jan
Quoted 11 lines · Jan Harkes
Quoted 7 lines · Alexandre Oliva
Just make the tivoization machinery require two keys: one that the vendor keeps, one that the vendor gives to the user (maybe without ever knowing it). Neither one can install modifications alone, but the user can approve modifications by the vendor, and the vendor can approve modifications by the user. This is still not ideal, but it at least doesn't permit the vendor to remove features from under the user.
So what features has Tivo removed (or threatened to remove) from the GPL licensed parts?
Why does it matter? The point is that a tivoizer *could* do that.
Quoted 1 line
So Google is using Linux right. What if they remove some feature?
Are you claiming Google is tivoizing something in their internal infrastructure? They're not distributing or conveying that software, so, nothing wrong with that.
Or are you talking about their search appliance, which I know nearly nothing about?
Quoted 3 lines
So if Tivo would allow you to boot your own kernel, but keeps the harddisk encrypted if the booted kernel does not have the right signature?
The information must suffice to ensure that the continued functioning of the modified object code is in no case prevented or interfered with solely because modification has been made.
Quoted 2 lines
And how is that any different from taking an off-the-shelf PC and booting your own kernel with Tivo's modifications?
TiVo did not sell me that off-the-shelf PC with the Free Software in it. It (hypothetically) sold me a computer with technical measures meant to restrict my ability to adapt the software it shipped to my own needs and to run it for any purpose, while it can still do that. That's a difference.
Quoted 8 lines
Quoted 3 lines
this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product
So they keep the system locked down, but include perl/python/emacs and distribute updates in the form of
scripts/sourcecode which are either interpreted or compiled to a ramfs filesystem at boot. Time to add another exception?
Intent behind this?: weasel out of the obligations of the license. Anyone, probably even a US court, might very well see it that way.
They retain the ability to modify the software, so they ought to pass it on to the user.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 8 lines · Alexandre Oliva
Quoted 3 lines · Ingo Molnar
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
---
This characterization misses something important. For many product devices, like cell phones, the modification is never "behind the user's back", but is done because the user has requested it (to fix a problem or add a new feature). If you go to ROM-based software, the user loses, because problems can't be fixed. For certain kinds of problems the user might be able to get a replacement device, but potentially involving losing any data stored on the device.
The FSF's approval of this distinction (ROM versus replaceable) places the FSF's particular principles over users interests, for no particular reason - if the manufacturer believes that it cannot legally allow software modification, all the restriction does is force them either to make the software unmodifiable (which advances freedom not at all) or to use software under a different license (which advances freedom not at all). The result? The user STILL has no freedom to modify the software and the community around the software is diminished.
To go back to the "behind your back" claim, the only cases I know where the software is replaced behind the users' back are cases were the updates are done by a service (usually not operated by the device manufacturer) that the user has voluntarily requested (like TiVo program guides or cable system subscriptions), which is generally a cases outside the scope of the license in any case.
scott
Quoted 7 lines · Scott Preece
Quoted 6 lines · Alexandre Oliva
Quoted 3 lines · Ingo Molnar
it irreversibly cuts off certain people from being to distribute GPLv3-ed software alongside with certain types of hardware that the FSF's president does not like.
Quoted 6 lines
Quoted 2 lines
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
This characterization misses something important. For many product devices, like cell phones, the modification is never "behind the user's back"
Okay, take out the "behind the users' back", it makes no difference. That was just to highlight the frequent evil intentions behind keeping the keys.
I wonder if giving half the key to the user and keeping the other half would be enough to satisfy the GPLv3 language while still enabling the vendor and user to update the software together.
Quoted 3 lines
The FSF's approval of this distinction (ROM versus replaceable) places the FSF's particular principles over users interests, for no particular reason
Over *users* interest? How so?
Quoted 5 lines
if the manufacturer believes that it cannot legally allow software modification, all the restriction does is force them either to make the software unmodifiable (which advances freedom not at all) or to use software under a different license (which advances freedom not at all).
Right.
But if the manufacturer believes that it can legally allow it, and wants to be able to install, software modifications, then it must decide between giving that up and letting the user do it as well. And this is where the users interests may prevail.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · Alexandre Oliva
Quoted 2 lines · Scott Preece
Quoted 2 lines
That's not true. They can just as well throw the key away and refrain from modifying the installed software behind the users' back.
Quoted 7 lines
Quoted 3 lines
This characterization misses something important. For many product devices, like cell phones, the modification is never "behind the user's back"
Okay, take out the "behind the users' back", it makes no difference. That was just to highlight the frequent evil intentions behind keeping the keys.
---
Yes, but in highlighting the possibility of evil intentions you distort the fact that usually there are no such evil intentions...
---
Quoted 4 lines
I wonder if giving half the key to the user and keeping the other half would be enough to satisfy the GPLv3 language while still enabling the vendor and user to update the software together.
---
I think that's a possibility. I don't see how it's functionally different from the usual case where the manufacturer can't modify the device without the user's consent simply because the user has physical access to the device and the manufacturer doesn't.
---
Quoted 6 lines
Quoted 4 lines
The FSF's approval of this distinction (ROM versus replaceable) places the FSF's particular principles over users interests, for no particular reason
Over *users* interest? How so?
---
Users benefit from the ability to get software updates, from the manufacturer, to resolve problems, fix security vulnerabilities, and provide updated functionality.
---
Quoted 7 lines
Quoted 6 lines
if the manufacturer believes that it cannot legally allow software modification, all the restriction does is force them either to make the software unmodifiable (which advances freedom not at all) or to use software under a different license (which advances freedom not at all).
Quoted 7 lines
Right.
But if the manufacturer believes that it can legally allow it, and wants to be able to install, software modifications, then it must decide between giving that up and letting the user do it as well. And this is where the users interests may prevail.
---
You're harping on the "cannot legally", which is fine but irrelevant. Whether it's a legal requirement or a business decision, the result is the same - neither forcing the manufacturer to make the device non-updatable nor forcing the manufacturer to use different software benefits anyone. I don't know of interesting cases where the manufacturer makes the device non-modifiable out of sheer bloody-mindedness.
I don't believe that the existence of this clause will lead to more manufacturers making their devices modifiable - there are too many other options if they think that non-modifiability is important to them.
[Note that I *do* think it's perfectly appropriate that authors who feel that they don't want their work used in such devices should be able to license them in line with that belief. I just don't think it has any practical value aside from making them feel better.]
scott
Quoted 10 lines · Scott Preece
Quoted 5 lines · Alexandre Oliva
Quoted 3 lines
The FSF's approval of this distinction (ROM versus replaceable) places the FSF's particular principles over users interests, for no particular reason
Over *users* interest? How so?
Users benefit from the ability to get software updates, from the manufacturer, to resolve problems, fix security vulnerabilities, and provide updated functionality.
Which they could have the option to do themselves if the manufacturer didn't prohibit them from doing so.
Quoted 13 lines
Quoted 6 lines
Quoted 5 lines
if the manufacturer believes that it cannot legally allow software modification, all the restriction does is force them either to make the software unmodifiable (which advances freedom not at all) or to use software under a different license (which advances freedom not at all).
Quoted 6 lines
Right.
But if the manufacturer believes that it can legally allow it, and wants to be able to install, software modifications, then it must decide between giving that up and letting the user do it as well. And this is where the users interests may prevail.
Quoted 4 lines
Whether it's a legal requirement or a business decision, the result is the same - neither forcing the manufacturer to make the device non-updatable nor forcing the manufacturer to use different software benefits anyone.
I agree. But that's an incomplete picture.
It's the other part of the picture, that you left out twice, that is the case that is good for the users *and* for the community.
Quoted 9 lines
I don't believe that the existence of this clause will lead to more manufacturers making their devices modifiable - there are too many other options if they think that non-modifiability is important to them.
[Note that I *do* think it's perfectly appropriate that authors who feel that they don't want their work used in such devices should be able to license them in line with that belief. I just don't think it has any practical value aside from making them feel better.]
They can do that with GPLv3. And those who don't want to stop this can then add a special permission. And then everybody wins.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 8 lines · Alexandre Oliva
Quoted 4 lines · Scott Preece
Whether it's a legal requirement or a business decision, the result is the same - neither forcing the manufacturer to make the device non-updatable nor forcing the manufacturer to use different software benefits anyone.
Quoted 4 lines
I agree. But that's an incomplete picture.
It's the other part of the picture, that you left out twice, that is the case that is good for the users *and* for the community.
---
I don't think I "left it out". The point is that if the manufacturer is unwilling to give the right to modify, no change in the language is going to cause the user to have that right.
scott
Quoted 11 lines · Scott Preece
Quoted 10 lines · Alexandre Oliva
Quoted 4 lines · Scott Preece
Whether it's a legal requirement or a business decision, the result is the same - neither forcing the manufacturer to make the device non-updatable nor forcing the manufacturer to use different software benefits anyone.
I agree. But that's an incomplete picture.
Quoted 7 lines
Quoted 2 lines
It's the other part of the picture, that you left out twice, that is the case that is good for the users *and* for the community.
---
I don't think I "left it out". The point is that if the manufacturer is unwilling to give the right to modify, no change in the language is going to cause the user to have that right.
If the alternatives are worse for the manufacturer than letting the user have it, then it will have the intended effects. In the other cases, it won't make much of a difference for anyone else.
The question is: how does tivoization help the community (under the tit-for-tat reasoning)? Does it help more than anti-tivoization?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 4 lines · Scott Preece
Yes, but in highlighting the possibility of evil intentions you distort the fact that usually there are no such evil intentions...
I don't think you can use "usually" and "fact" together like that. Why is it so bad to account for them since they (do) surface and (could) increase significantly in frequency?
For me, the (could) is enough to act upon, regardless of the current likely hood of it happening. Things change frequently.
This, unfortunately comes pre-distorted depending on what you believe. All of us are right but we still don't agree. Quite a fluke.
That's the problem.
Best,
--Tim
Quoted 10 lines · Tim Post
Quoted 4 lines · Scott Preece
Yes, but in highlighting the possibility of evil intentions you distort the fact that usually there are no such evil intentions...
I don't think you can use "usually" and "fact" together like that. Why is it so bad to account for them since they (do) surface and (could) increase significantly in frequency?
---
I agree that it is possible to have different definitions of "evil" and "ethical"...
scott
Quoted 6 lines · Alexandre Oliva
Quoted 5 lines
if the manufacturer believes that it cannot legally allow software modification, all the restriction does is force them either to make the software unmodifiable (which advances freedom not at all) or to use software under a different license (which advances freedom not at all).
Quoted 6 lines
Right.
But if the manufacturer believes that it can legally allow it, and wants to be able to install, software modifications, then it must decide between giving that up and letting the user do it as well. And this is where the users interests may prevail.
with the little tiny problem that this is not what the GPLv3 actually implements. Our point from the very beginning: the GPLv3 "outlaws" certain hardware restrictions _even if they are fully legitimate_. Yes, of course, it also outlaws 'bad' uses of DRM. The GPLv3 tries to carve out some known 'good' uses of DRM (to stop the GPLv3 from being _totally_ unusable in vast areas of the marketplace), but that limited opt-in approach can in no way be the right solution (think about it as a whitelist - wouldnt you want to be able to _add_ to that whitelist?? The GPLv3 hardcodes it.)
In other words: i dont want the police to start shooting innocent people in the streets, in their pursuit of criminals. Yes, this means criminals have an easier job getting away - but _that_ is the price of freedom!
and all these problems of the GPLv3 DRM language derives from the same root issue: RMS is trying to make a manual call about what _technology use_ is 'good' and what is 'evil'. For some of these calls we might even agree. But most fundamentally, a license should _never_ get into the business of trying to 'judge' what _use_ is 'good' and what is 'evil'. As you can see it on this list alone, some people see Tivo's intentions as legitimate, they did the DRM to stay in business but still be able to use free software, employ free software developers, show Mythbox how to do this stuff on Linux, etc. But the GPLv3 completely destroys Tivo's ability to use Linux, were Linux to be under the GPLv3. And by doing that, those contested provisions of the GPLv3 itself become a tool against "freedom".
You tried to find a workaround for that, by suggesting the 'dont do security fixes then', 'use a split key', 'use a rent model' solutions, but dont you realize that by suggesting those you are explicitly against the intent of RMS, who wants to _stop_ Tivo from being able to do DRM? Dont you think it speaks volumes of the GPLv3's quality that you have to go out and search for a _workaround_, for a _back door in the license_, you have to _go against the intent of RMS_, to be able to implement something simple as upgradability? And that even after several attempts you have yet to find a solution that actually enables the Tivo? (and that is not an accident: RMS _does not want_ the Tivo to use a GPLv3'd kernel, and the GPLv3 is structured that way.)
Ingo
Quoted 10 lines · Ingo Molnar
Quoted 8 lines · Alexandre Oliva
Quoted 5 lines
if the manufacturer believes that it cannot legally allow software modification, all the restriction does is force them either to make the software unmodifiable (which advances freedom not at all) or to use software under a different license (which advances freedom not at all).
Right.
Quoted 7 lines
Quoted 4 lines
But if the manufacturer believes that it can legally allow it, and wants to be able to install, software modifications, then it must decide between giving that up and letting the user do it as well. And this is where the users interests may prevail.
with the little tiny problem that this is not what the GPLv3 actually implements.
Except that it does. Go read it, then come back and admit you were mistaken and spreading lies about it.
The vendor must decide between respecting the freedom of the user, or stopping itself from modifying the software too.
Quoted 2 lines
But most fundamentally, a license should _never_ get into the business of trying to 'judge' what _use_ is 'good' and what is 'evil'.
What it does is to seek to carry out its mission (*) of defending users' freedoms. Obstacles that are placed to impede the user from enjoying the freedoms are supposed to not be permitted by the GPL.
(*) it seems that understanding "spirit of a license" is very difficult for you; does the term "mission" help you understand what the GPL means when it says "similar in spirit"?
Quoted 2 lines
But the GPLv3 completely destroys Tivo's ability to use Linux, were Linux to be under the GPLv3.
Sorry, wrong. Barring nonsense.
Quoted 2 lines
You tried to find a workaround for that, by suggesting the 'dont do security fixes then', 'use a split key', 'use a rent model' solutions,
FTR, rent model wasn't me, and it doesn't escape the GPLv3dd4 obligations IIUC, IANAL.
Quoted 2 lines
but dont you realize that by suggesting those you are explicitly against the intent of RMS, who wants to _stop_ Tivo from being able to do DRM?
You misunderstand not only the spirit of the license, but his intentions. Oh, wait! They're the same, that's why.
Respect and defend users' freedoms.
Repeat after me until it sinks.
I know you're not stupid. Why do you pretend to be?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 11 lines · Ingo Molnar
that's fine, but the fundamental question is: where is the moral boundary of the power that the copyright license gives? The FSF seems to believe "nowhere, anything that copyright law allows us to achieve our goals is a fair game" - and the GPLv3 shows that belief. I dont subscribe to that view. I think the proper limit is the boundary where the limit of the software is - because that's the only sane and globally workable way to stop the power-hungry. I.e. the information we produce is covered by the rules of the GPL. It might be used in ways inconvenient to us, it might be put on hardware we dont like (be that a Tivo, a landmine or an abortion instrument) but that does not change the fundamental fact: it's outside the
_moralscope_of our power.
Where is the boundary between hard- and software? I'm employed as hardware designer, and for this purpose, I write programs in a hardware description language, which can be converted into hardware through a synthesis software. I write firmware, which is assembled into binary and gets placed on on-chip memory (ROM or NVM). I've even studied computer science, and electric engineering was just a side-course. I know how transistors work, and how gates are implemented in terms of transistors, but in essense, it's not that relevant unless you want to do analog circuits. Usually, during the development phase, we put the Verilog into an FPGA, where the configuration file still is obviously "software" in any sense it can be. I've even released descriptions of some parts of the work I do under GPL for people to put it into their own FPGAs.
There is no boundary between hard- and software in the sense of that hardware is something fundamentally different. Hardware is just another way to implement programs, and it uses other languages (but SystemC even looks quite close to C). If there is a boundary, it's way below the distinction between a Tivo and a PC, because these two basically consist of a processor, some RAM, some flash, a harddisk, and a video driver.
What's true: We don't have the moral power to define *where* the software goes, but we have the moral power to define *how* users can change the software when they own the hardware (the physical representation).
Signature · 3 lines
Bernd Paysan "If you want it done right, you have to do it yourself" http://www.jwdt.com/~paysan/
Quoted 13 lines · Bernd Paysan
Quoted 11 lines · Ingo Molnar
that's fine, but the fundamental question is: where is the moral boundary of the power that the copyright license gives? The FSF seems to believe "nowhere, anything that copyright law allows us to achieve our goals is a fair game" - and the GPLv3 shows that belief. I dont subscribe to that view. I think the proper limit is the boundary where the limit of the software is - because that's the only sane and globally workable way to stop the power-hungry. I.e. the information we produce is covered by the rules of the GPL. It might be used in ways inconvenient to us, it might be put on hardware we dont like (be that a Tivo, a landmine or an abortion instrument) but that does not change the fundamental fact: it's outside the
_moralscope_of our power.
Quoted 1 line
Where is the boundary between hard- and software? [...]
this is largely irrelevant to my argument: the FSF is clearly trying to extend the scope of the GPL to restrict the distribution of certain hardware+software combinations. The FSF is not really arguing that the boundary between software and hardware is diffuse. (which btw. it clearly is) The FSF simply wants to be able to say via the GPLv3: "to be able to distribute GPL-ed software, the hardware is required to do this and this".
please note an important thing here: "required to do this and this" is the _precise antithesis_ of "freedom". The only significant restriction on freedom the GPLv2 allows is that the covered work (the software) is not to be restricted. And that is a fair deal. Even if any additional restrictions would otherwise be for the "common good" and would further "freedom" in creative ways.
If the FSF's argument and approach was correct then it would be fine to add these restrictions to GPLv4:
- do not distribute non-GPL-compatible software with GPL software on the same hardware.
- send at least 10 free samples of the hardware to the FSF headquarters. (after all true freedom is only achieved if developers are not only allowed to modify the hardware, but are allowed to test it as well, for the freedom of the community.)
- donate $10 to the FSF.
- spray "Linus sucks because he stole RMS's GNU thunder in the 90s and never gave it back!" graffiti on 3 separate walls in your neighborhood.
Each of these items is an additional restriction on either the hardware+software combination that is being distributed or on the person who does the distribution, and each of these items - some abstractly, some more directly - advance a notion of the "four GNU freedoms" in some way. And each of these items has a basis in copyright law and might be legally put into a license and might be enforceable. (ok, probably not the last item ;)
think about it, the list of things that one can do via license to "achieve more freedom" just doesnt stop! My point is: it has to stop at the only boundary that makes sense, and which boundary is clearly spelled out in the spirit and in the letter of the GPLv2: "our work is our work, your work is your work".
Any additional restriction to "help achieve more freedom" just puts us into divisive political and moralistic games that will only fracture us, that will eventually erode the value of our software and hence makes any 'power' we have over that software meaningless in the real world. In the end no-one but the Microsofts of this world will win.
Ingo
Quoted 3 lines · Ingo Molnar
My point is: it has to stop at the only boundary that makes sense, and which boundary
is clearly spelled out in the spirit and in the letter of the GPLv2: "our work is our work, your work is your work".
Agreed - if you want to take my work you are welcome as long as you contribute back your changes. That's the deal that GPL2 enforces and why it has been so successful. GPL3 is a very different beast with a much wider agenda, which makes it far more difficult to achieve consensus on what it should contain.
Personally I would have liked to seen a GPL2.x which fixes some of the issues but stays true to the more limited objective.
Alan
Quoted 3 lines · Alan Milnes
Agreed - if you want to take my work you are welcome as long as you contribute back your changes. That's the deal that GPL2 enforces and why it has been so successful.
That may be a side effect of the GPL, but it's actually not how the GPLv2 works (nor is it the intention). "Contribute back" means upstream. There's no such provision in the GPLv2, you contribute only downstream. And there are cases where you don't need to contribute at all.
E.g. the kernel hacking I'm doing at the moment: I have bought a uClinux blackfin board, for testing my digital audio amplifier. For that, I took one of the blackfin alsa audio drivers, and changed it so that it could talk to my digital audio amplifier. I'm not distributing this software, it's a complete in-house project, so I'm not obliged to contribute back. At the moment, I'm the only person in the world who has both access to the digital audio amplifier and the blackfin board, so releasing this driver in that early stage is a rather pointless excercise.
I think this above explains fairly well the "misunderstandings" that are appearing here. The GPL is not reflective (tit-for-tat), it's transient. If there's a loop in the transient propagation, it becomes reflective through the loop, but not by itself. This was the case in GPLv1, is the case in GPLv2, and will be the case in GPLv3.
Signature · 3 lines
Bernd Paysan "If you want it done right, you have to do it yourself" http://www.jwdt.com/~paysan/
Quoted 9 lines · On Thursday 14 June 2007 11:44:07 Bernd Paysan wrote:
Quoted 3 lines · Alan Milnes
Agreed - if you want to take my work you are welcome as long as you contribute back your changes. That's the deal that GPL2 enforces and why it has been so successful.
That may be a side effect of the GPL, but it's actually not how the GPLv2 works (nor is it the intention). "Contribute back" means upstream. There's no such provision in the GPLv2, you contribute only downstream. And there are cases where you don't need to contribute at all.
And the Linux kernel community has been familiar with this situation all along. It's the bargain the kernel developers struck with each other a decade and a half ago.
Now the FSF is coming along and being Darth Vader: "I am altering the bargain. Pray I don't alter it any further."
Quoted 5 lines
I think this above explains fairly well the "misunderstandings" that are appearing here. The GPL is not reflective (tit-for-tat), it's transient. If there's a loop in the transient propagation, it becomes reflective through the loop, but not by itself. This was the case in GPLv1, is the case in GPLv2, and will be the case in GPLv3.
That's not specifically a limitation of the GPL, that's a limitation of copyright law which forms the basis of the GPL. It covers distribution, not usage.
GPLv2 eliminates the case where I have a modified binary I contributed to, but can't see the source code of those modifications. This has the pragmatic effect of greatly reducing forking in a project, such as the Emacs/Lucid Emacs fork that inspired the "Emacs license" that became GPLv1.
Rob
Quoted 2 lines · Rob Landley
Now the FSF is coming along and being Darth Vader: "I am altering the bargain. Pray I don't alter it any further."
1) it can't possibly do that. the Linux license is something that only the Linux developers can decide.
2) I don't know how the FSF is approaching the Linux developers, but what I've been personally trying to do in this infinite thread was mainly to set the record straight that v3 did not change the spirit of the license, like some have claimed.
3) Another thing I've tried to do was to try to figure out why Linux developers seem to consider v2 better than v3 for their own goals. I must admit I failed. The presented reasons don't seem to distinguish v2 from v3 to me, or rather make v3 sound better.
It's disappointing that I took so much of everyone's time without achieving any of my goals. I hope it was at least useful or enlightening to some.
I'll now try to step out of the discussion, but I guess I'm just as addicted to flames. I don't see that it's getting anywhere, and I don't particularly enjoy the name calling. And then, I was politely invited to go away...
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 3 lines · Alexandre Oliva
It's disappointing that I took so much of everyone's time without achieving any of my goals.
What do you expect, when you tried to entertain a legal picture of the GPLv2 that even the FSF counsel doesn't believe in?
I will state one more time: I think that what Tivo did was and is:
(a) perfectly legal wrt the GPLv2 (and I have shown multiple times why your arguments don't hold logical water - if you actually followed them yourself, you wouldn't be using a redhat.com email address!)
(b) not just legally right, but perfectly morally right too (it wasn't some underhanded "trick" thing - it was following the spirit _and_ the letter of the law)
(c) the only reasonable thing many companies *can* do in the face of laws and regulations and entities like the RIAA/MPAA.
and you should admit that the fact that the FSF counsel says that it couldn't sue Tivo in the US, it means that while my standpoint may not be the _only_ possible one, I'm certainly not "confused" about (a) above.
The (b) and (c) points are not "legal" points, they are about the fact that quite often, morality and practicality are independent of legality, and you should never see law as being the *only* thing that matters. So the reason I bring them up is that it wasn't just "legally ok", they also had good *reasons* for doing it, and there was no hanky-panky about it!
In fact, I consider Tivo one of the good guys, because they were one of the few people that had things like the GPLv2 actually printed out and clearly stated IN THE MAIN PAPER MANUAL. In the very first version of their box. Without anybody twisting any arms at all.
IOW, Tivo really did everything right. I personally think that they were even classy about it.
And that's my opinion. THINK about that for a moment. THINK about the fact that I am the original copyright holder in the main software project they used, and that I state that as neither having ever gotten paid _or_ owning any stock what-so-ever in Tivo.
Dammit, if I cannot say that I think what they did was fine, who can?
So pause there for a moment, and really *think* about the above. Stop seeing Tivo as "the devil".
[ Wait a few seconds here, thinking! ]
Now, we both agree that GPLv3 would change the situation wrt Tivo, don't we?
[ Wait a few more seconds here, thinking about what that means, taking the
above into consideration ]
..so given that I think that what Tivo did is *fine*, the GPLv3 "solution" is not a solution at all, is it?
Quite the reverse. It's a unnecessary restriction that doesn't actually solve anything at all, it just adds problems of its own.
And yet you claim that you cannot understand why I (and others) would consider the GPLv3 to be a "worse" license. It is *obviously* worse to anybody who thought that "Tivoization" wasn't a problem in the first place!
.. but I guess you'll ignore that argument, the way you ignored all the other ones too, and continue to blame your lack of understanding on me being "confused" about the issue.
Linus
Quoted 7 lines · Linus Torvalds
Quoted 3 lines · Alexandre Oliva
It's disappointing that I took so much of everyone's time without achieving any of my goals.
What do you expect, when you tried to entertain a legal picture of the GPLv2 that even the FSF counsel doesn't believe in?
I don't think I made significant legal arguments. My points were about the spirit of the license. That's not legal at all. That's moral and ethical background.
Your aggressive response directed at the FSF came as quite a surprise to me, given the way I got into this conversation.
Quoted 3 lines
I will state one more time: I think that what Tivo did was and is:
(a) perfectly legal wrt the GPLv2
I respectfully disagree, and I know I'm not alone in this assessment. I know other kernel developers agree with it. And they're as entitled to claim failure to comply with the license as you are.
Quoted 2 lines
(and I have shown multiple times why
your arguments don't hold logical water
You countered one of the various arguments I have, and you failed at that. It was another, quite different argument, that got me to realize it didn't work legally. But this says nothing about compliance with the spirit of the license.
Quoted 1 line
(b) not just legally right, but perfectly morally right too
I guess we'll have to agree to disagree on this one.
Quoted 2 lines
(c) the only reasonable thing many companies *can* do in the face of laws and regulations and entities like the RIAA/MPAA.
No, TiVO could just throw the key away. Why doesn't it?
Quoted 3 lines
and you should admit that the fact that the FSF counsel says that it couldn't sue Tivo in the US, it means that while my standpoint may not be the _only_ possible one, I'm certainly not "confused" about (a) above.
I don't think I've ever claimed you were confused about (a). I said we disagreed. That's quite different.
You were confused between the legalese and the spirit.
Quoted 3 lines
The (b) and (c) points are not "legal" points, they are about the fact that quite often, morality and practicality are independent of legality, and you should never see law as being the *only* thing that matters.
Aah, now we're getting somewhere!
Quoted 1 line
Dammit, if I cannot say that I think what they did was fine, who can?
If you and all other Linux copyright holders agreed about it, sure.
Just like you could all grant it an additional permission, just so that they feel safe about it.
Quoted 2 lines
Now, we both agree that GPLv3 would change the situation wrt Tivo, don't we?
Yes. They'd have to give up the ability to update the software, or pass it on to the user. If they can't do the latter, they could still do the former. How bad would this be for them, do you know?
Quoted 2 lines
And yet you claim that you cannot understand why I (and others) would consider the GPLv3 to be a "worse" license.
That's because when you talk about why GPLv2 is better, you always talked about virtues that are just as present in v3, and that AFAICT would in fact by increased by v3.
Quoted 3 lines
.. but I guess you'll ignore that argument, the way you ignored all the other ones too, and continue to blame your lack of understanding on me being "confused" about the issue.
I'm sorry if I come off that way. But we really look at this issue from very different perspectives, and it's difficult for me to try to see it your way. Please cut me some slack here. I'm trying hard, but there is so much noise and so many hard feelings that seeing what the real issues are is not that easy.
Thanks for your understanding,
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 4 lines · Alexandre Oliva
Yes. They'd have to give up the ability to update the software, or pass it on to the user. If they can't do the latter, they could still do the former. How bad would this be for them, do you know?
In other words, you advocate license for technical programs that causes people to make bad technical choices?
Yeah, that's real smart. That's a sign of true intellect, isn't it?
Wrong.
Linus
Quoted 8 lines · Linus Torvalds
Quoted 4 lines · Alexandre Oliva
Yes. They'd have to give up the ability to update the software, or pass it on to the user. If they can't do the latter, they could still do the former. How bad would this be for them, do you know?
In other words, you advocate license for technical programs that causes people to make bad technical choices?
I do place ethical issues over technical ones, if that's what you're asking.
And then, why should the vendor have any say on the software that runs on the hardware I purchased from them, after the purchase?
Heck, I'd feel *safer* if I knew they couldn't modify the code in my box without my permission. Do you support WGA or the Sony Rootkit? How is this any different?
But if they want to keep the ability to change the software in my box, I want that for myself as well. If for no other reason, in case they mess things up.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · Linus Torvalds
I will state one more time: I think that what Tivo did was and is:
(a) perfectly legal wrt the GPLv2 (and I have shown multiple times why your arguments don't hold logical water - if you actually followed them yourself, you wouldn't be using a redhat.com email address!)
Linus, Harald Welte managed to get the keys from Siemens, which sold a "tivoized" Linux router. As typical for German courts, he got the key in a settlement, so there's no citable court verdict (totally, he AFAIK got two court verdicts up to now). But the way settlements work in Germany suggests that it's likely that the court would have decided that way*.
So there is no verdict on this question, but a strong hint that you are wrong, at least under German law (which is not the best money can buy ;-). And as Tivo doesn't sell their crap into Germany, we can't test it with them.
It's you who think the GPLv2 is tit-for-tat, not the FSF, and it's not in the text of the GPLv2; this is your private misinterpretation. The GPLv2 is "transitive", i.e. everybody receives the same rights, nobody is entitled to take rights from others (the only person to choose is the author), and this certainly includes technical means (even if it does not explicitely say so), and the GPLv3 is just the same (it's just more readable). All arguments from your side put up are straw men like "hardware vs. software" and such.
BTW: Hardware as it is done today (chips, printed circuit boards, etc.) is copyrighted as well, as it is much cheaper to copy hardware than to develop it.
*) German judges want to resolve civil cases by settlements. They let the parties pass a few arguments, and indicate which way the judge possibly would decide to help them settle fast. Settlement is way cheaper than a court verdict, so most sane people choose to settle (unfortunately it's often the insane people who go to court ;-).
Signature · 3 lines
Bernd Paysan "If you want it done right, you have to do it yourself" http://www.jwdt.com/~paysan/
Quoted 4 lines
2) I don't know how the FSF is approaching the Linux developers, but what I've been personally trying to do in this infinite thread was mainly to set the record straight that v3 did not change the spirit of the license, like some have claimed.
The FSF have certainly tried to talk to me a bit about it - mostly about some product called GNU/Linux which I had to tell them I'd never heard of and wasn't involved in ;)
Quoted 5 lines
3) Another thing I've tried to do was to try to figure out why Linux developers seem to consider v2 better than v3 for their own goals. I must admit I failed. The presented reasons don't seem to distinguish v2 from v3 to me, or rather make v3 sound better.
At least one important one I think is this:
A large number of people contributed to the GPLv2 kernel. They did so on the basis there was an agreement about how the result could and would be used. The GPLv3 changes that agreement, whether for good or bad depends on who you are and what you do.
What right does Linus or anyone else have to change the rules and unavoidably harm some of the people who contributed on the basis of the previous licence. Any community project is built around a set of expectations and beliefs encoded in culture, licences, documents and so on.
The kernel community was built around GPLv2. A large number of the people involved did so for pragmatic not FSF reasons and are not part of FSF culture. The fact that community isn't interested in GPL3 should not be a suprise, nor should it be seen as it seems you see it to be a failure of the GPL3.
GPLv2 is how we've done it, it has been for fourteen years and numerous people have contributed on that basis. Should we kick some of them out of that community because a third party says "new license good". What matters more to the project itself - respect for those who work on it and their beliefs or an FSF attempt to strengthen free software protection ? Thats an "ends and means" type question but I think it explains the fundamental question very well.
Alan
Quoted 6 lines · Alan Cox
Quoted 4 lines
3) Another thing I've tried to do was to try to figure out why Linux developers seem to consider v2 better than v3 for their own goals. I must admit I failed. The presented reasons don't seem to distinguish v2 from v3 to me, or rather make v3 sound better.
What right does Linus or anyone else have to change the rules
None. It would have to be an agreement between all parties involved. A difficult one, everyone knows.
But see, this is a distraction. It doesn't even begin to address the relevant (to me) question: why Linux developers seem to consider v2 better than v3 for their own goals.
I can appreciate the difficulties that there would be for switching from v2 to v3. This in itself might be a reason to not even try to switch to another license, no matter how much better it could possibly be.
But it doesn't give any hint whatsoever as to why v2 is better than v3. In fact, it simply avoids addressing that point.
Now, of course, each individual contributor may have different reasons to be part of the Linux community, and each individual contributor may have chosen v2 or v2+ or any other v2-compatible set of licensing terms for different reasons.
I'd very much like to hear (err read), from those who think v2 serves their reasons to contribute to Linux better than v3, why that is so.
Thanks,
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 3 lines · Alan Milnes
Agreed - if you want to take my work you are welcome as long as you contribute back your changes. That's the deal that GPL2 enforces and why it has been so successful.
Where did you get this impression that GPLv2 enforces this deal?
It doesn't, and this is *exactly* why I dispute the claim that GPLv2 is tit-for-tat.
Quoted 1 line
GPL3 is a very different beast with a much wider agenda,
The agenda is *precisely* the same: ensure that all users are free to modify and share the licensed software.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · Ingo Molnar
this is largely irrelevant to my argument: the FSF is clearly trying to extend the scope of the GPL to restrict the distribution of certain hardware+software combinations. The FSF is not really arguing that the boundary between software and hardware is diffuse. (which btw. it clearly is) The FSF simply wants to be able to say via the GPLv3: "to be able to distribute GPL-ed software, the hardware is required to do this and this".
Most people arguing for the expansive interpretation do not really care what hardware is combined with what software. They care about the ability for the user (in the GPLv2's terms, someone who receives GPL'ed software) to have comparable ability to modify and (re-)distribute the software as the software distributor does. The issue of GPLed software on DRMed hardware applies equally to digital video recorders, where the hardware and software distributor are usually the same, and video game consoles, where they are not.
There is no good reason to treat a "GPL-incompatible" hardware platform (for example, incompatible due to restrictions on the keys to generate digital signatures) differently than a "GPL-incompatible" patent area. If a software distributor cannot simultaneously comply with the GPL and his other obligations, he should either not distribute the software or be prepared to face the liability from breaching his obligations.
Michael Poole
Quoted 1 line · On Thursday 14 June 2007 07:27:59 Bernd Paysan wrote:
Where is the boundary between hard- and software?
Software's the bit that's infinitely replicable at zero cost. Hardware tends not to be.
Rob
Quoted 5 lines · Rob Landley
Quoted 1 line · On Thursday 14 June 2007 07:27:59 Bernd Paysan wrote:
Where is the boundary between hard- and software?
Software's the bit that's infinitely replicable at zero cost. Hardware tends not to be.
There's no "zero cost" for software replication, either. You have to pay for your media, even if today's price for a GB harddisk space is just 20 Euro-cents. You have to pay for your bandwidth (and even if it's a flat-rate, the maximum amount of data you can get through is bandwidth*(seconds per month) for one month fee). Hardware is replicated as well as software, the cost for hardware replication is higher than for software replication, because more things are to do. The basical principle of producing a CD-ROM and a chip is exactly the same: lithography. You "print" it. A decade ago, ES2 had made chips by direct e-beam lithography, so the offset of the mask costs were eliminated.
With an ES2-like process, you could have your "free software CPU", where you design modifications yourself, send the file to the fab, and get your customized chip back for essentially the same price as a non-customized version (supposed all the tool-chain would be free software, and not horrible expensive Cadence/Synopsys/Mentor software).
Signature · 3 lines
Bernd Paysan "If you want it done right, you have to do it yourself" http://www.jwdt.com/~paysan/
Quoted 3 lines · Ingo Molnar
I think the proper limit is the boundary where the limit of the software is - because that's the only sane and globally workable way to stop the power-hungry.
But see, I'm not talking about getting permission to hack the hardware. I'm only talking about getting permission to hack the Free Software in it.
It's your position that mingles the issues and permits people to use the hardware to deprive users of freedom over the software that they're entitled to have.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 4 lines · Alexandre Oliva
But see, I'm not talking about getting permission to hack the hardware. I'm only talking about getting permission to hack the Free Software in it.
You have that permission. You can hack on it all you want.
Oh, but you want to hack the hardware to accept it? That's a totally different issue. If so, buy a Neuros OSD box.
Really. Go to google, and type in "Neuros OSD".
Do it *now*, and then stop wasting our time. You *can* do what you want to do.
Linus
Quoted 8 lines · Linus Torvalds
Oh, but you want to hack the hardware to accept it? That's a totally different issue. If so, buy a Neuros OSD box.
Really. Go to google, and type in "Neuros OSD".
Do it *now*, and then stop wasting our time. You *can* do what you want to do.
Unfortunatly that device has some closed source kernel drivers, so you don't really have "full" control over the system :(
thanks,
greg k-h
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Forgive me for a little off-topic question but I have a difficulty to understand a technical issue about this all.
The Linux Kernel cannot easily switch licenses because of the large amount of people involved in it (i.e. contributed code on which they have copyright).
But many of FSF's GNU projects are similar - for example GCC has contributors from many many companies and individuals, from which I presume there are who might object to GPLv3.
So how come they can so easily move to GPLv3 ?
Don't they have to have permission from all of those contributors (many of which are Linux companies and distributors who might prefer staying at GPLv2) ?
Quoted 7 lines · Neshama Parhoti
Forgive me for a little off-topic question but I have a difficulty to understand a technical issue about this all.
The Linux Kernel cannot easily switch licenses because of the large amount of people involved in it (i.e. contributed code on which they have copyright).
Quoted 6 lines
But many of FSF's GNU projects are similar - for example GCC has contributors
from many many companies and individuals, from which I presume there are who might object to GPLv3.So how come they can so easily move to GPLv3 ?
The FSF required copyright assignment to themselves in order to accept the changes from the developers. So the FSF owns the whole copyright and can change things whenever they want, to whatever license they want.
This is the exact opposite of the kernel in which all of the original contributors own the copyright.
Hope this helps,
greg k-h
Quoted 2 lines · Greg KH
The FSF required copyright assignment to themselves in order to accept the changes from the developers.
For many strategic projects, but not all of them.
Quoted 2 lines
So the FSF owns the whole copyright and can change things whenever they want, to whatever license they want.
This is not true. Have you ever read the copyright assignment contract? It very clearly constrains the ways the FSF can release the code.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 4 lines · Neshama Parhoti
But many of FSF's GNU projects are similar - for example GCC has contributors
from many many companies and individuals, from which I presume there are who might object to GPLv3.
FSF requires copyright assignment to the FSF on things like the compiler.
Bill
Quoted 4 lines
So how come they can so easily move to GPLv3 ?
Don't they have to have permission from all of those contributors (many of which are Linux companies and distributors who might prefer staying at GPLv2) ?
The FSF uses copyright assignments to ensure the entire project is under FSF control. Linux does not - there are benefits to both approaches.
Quoted 3 lines · Alexandre Oliva
But see, I'm not talking about getting permission to hack the hardware. I'm only talking about getting permission to hack the Free Software in it.
No you're not...you're talking about being able to hack the software *and load it back onto the original hardware*.
Quoted 3 lines
It's your position that mingles the issues and permits people to use the hardware to deprive users of freedom over the software that they're entitled to have.
The software license controls the software. If the hardware has restrictions on it that limit what software it will run, then that is unrelated to the software license.
There is nothing stopping you from taking the code for the tivo, modifying it, distributing it, or even running it on other hardware.
Suppose I had some machine that will only run microsoft-signed binaries.
Would it be at all related to any software license that this machine
won't let me run linux?
Chris
Quoted 7 lines · Chris Friesen
Quoted 3 lines · Alexandre Oliva
But see, I'm not talking about getting permission to hack the hardware. I'm only talking about getting permission to hack the Free Software in it.
No you're not...you're talking about being able to hack the software *and load it back onto the original hardware*.
Yes. You wouldn't impose restrictions on modifying the software like that, now would you? Even though the GPL says you can't impose further restrictions on modification and distribution.
Quoted 7 lines
Quoted 3 lines
It's your position that mingles the issues and permits people to use the hardware to deprive users of freedom over the software that they're entitled to have.
The software license controls the software. If the hardware has restrictions on it that limit what software it will run, then that is unrelated to the software license.
As in, the license controls the software. If a patent creates restrictions that limit what you can do with the software, then that is unrelated to the software license.
As in, the license controls the software. If a discriminatory contract limits what you can do with the software, then that is unrelated to the software license.
As in, the license controls the software. If I send you the source code, but it happens to be protected by a key that only the hardware can decode, and it won't decode for you, then that is unrelated to the software license.
Is that so, really?
Quoted 2 lines
There is nothing stopping you from taking the code for the tivo, modifying it, distributing it, or even running it on other hardware.
True. But TiVO is still imposing further restrictions on how I can modify the software stored in their device, while reserving that ability to itself. This is wrong. This is not "in kind". This is not "tit-for-tat". Tit-for-tat is: if they can, then I can too, and if I can't, then they can't either.
Quoted 3 lines
Suppose I had some machine that will only run microsoft-signed binaries. Would it be at all related to any software license that this machine won't let me run linux?
That would be an unfortunate machine to have, but if Linux or some other GPLed software was not shipped in it, then I don't see how this is relevant to this discussion. It's not about the hardware, it's about the software in it, and about passing on the freedoms related with it.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 6 lines · On Thursday 14 June 2007 15:13:31 Alexandre Oliva wrote:
Quoted 5 lines · Chris Friesen
Quoted 3 lines · Alexandre Oliva
But see, I'm not talking about getting permission to hack the hardware. I'm only talking about getting permission to hack the Free Software in it.
Quoted 6 lines
Quoted 2 lines
No you're not...you're talking about being able to hack the software *and load it back onto the original hardware*.
Yes. You wouldn't impose restrictions on modifying the software like that, now would you? Even though the GPL says you can't impose further restrictions on modification and distribution.
replace != modify
Quoted 5 lines
Quoted 5 lines
Quoted 4 lines
It's your position that mingles the issues and permits people to use the hardware to deprive users of freedom over the software that they're entitled to have.
Quoted 7 lines
Quoted 3 lines
The software license controls the software. If the hardware has restrictions on it that limit what software it will run, then that is unrelated to the software license.
As in, the license controls the software. If a patent creates restrictions that limit what you can do with the software, then that is unrelated to the software license.
No - because this case is covered in GPLv2. Lose the straw-men.
Quoted 3 lines
As in, the license controls the software. If a discriminatory contract limits what you can do with the software, then that is unrelated to the software license.
Incorrect. This is, again, covered by the GPLv2. Straw-man argument.
Quoted 4 lines
As in, the license controls the software. If I send you the source code, but it happens to be protected by a key that only the hardware can decode, and it won't decode for you, then that is unrelated to the software license.
Straw-man. Situation covered by the GPLv2.
Quoted 10 lines
Is that so, really?
Quoted 2 lines
There is nothing stopping you from taking the code for the tivo, modifying it, distributing it, or even running it on other hardware.
True. But TiVO is still imposing further restrictions on how I can modify the software stored in their device, while reserving that ability to itself. This is wrong. This is not "in kind". This is not "tit-for-tat". Tit-for-tat is: if they can, then I can too, and if I can't, then they can't either.
But that right has never been guaranteed by the GPL. It might have been the *intent* of RMS when he wrote GPLv1 and the *intent* of the FSF when they wrote GPLv2, but intent is worth exactly *NOTHING* in the law *UNLESS* that intent is spelled out.
Anyway, as I've pointed out before: replace != modify
You can *replace* parts of a program and it will be a modification, you can *replace* components of a piece of Hardware and it will be a modification but replacing one software component of a device with another is *NOT* a modification. Why? Because the hardware hasn't changed at all - the hardware is merely there so the software can perform its job. And since you are *replacing* the *ENTIRE* piece of software, it isn't a modification of the software.
Quoted 9 lines
Quoted 3 lines
Suppose I had some machine that will only run microsoft-signed binaries. Would it be at all related to any software license that this machine won't let me run linux?
That would be an unfortunate machine to have, but if Linux or some other GPLed software was not shipped in it, then I don't see how this is relevant to this discussion. It's not about the hardware, it's about the software in it, and about passing on the freedoms related with it.
Exactly. However, you are making it about the hardware by making the claim that "replacing a program, in its entirety, with another is a modification". It isn't. A modification is when you replace or change a *portion* of a program. By your logic I could write an operating system that is 100% binary compatible with Linux and I'd be *required* to release it under the GPL, because, even though it *replaces* Linux, it's still a "modification".
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 11 lines · Daniel Hazelton
Quoted 9 lines · On Thursday 14 June 2007 15:13:31 Alexandre Oliva wrote:
Quoted 8 lines · Chris Friesen
Quoted 3 lines · Alexandre Oliva
It's your position that mingles the issues and permits people to use the hardware to deprive users of freedom over the software that they're entitled to have.
The software license controls the software. If the hardware has restrictions on it that limit what software it will run, then that is unrelated to the software license.
Quoted 5 lines
Quoted 3 lines
As in, the license controls the software. If a patent creates restrictions that limit what you can do with the software, then that is unrelated to the software license.
No - because this case is covered in GPLv2. Lose the straw-men.
It's not a straw man. See, I was just showing that there's precedent to ensuring that other tricks can't be used to deny users the freedoms that the GPL is meant to defend.
By pointing out this is in the GPLv2, you acknowledge the point I wanted to make.
So what is it that makes hardware so different that it can be used as a trick to deny users freedoms, if other tricks can't?
Quoted 4 lines
But that right has never been guaranteed by the GPL. It might have been the *intent* of RMS when he wrote GPLv1 and the *intent* of the FSF when they wrote GPLv2, but intent is worth exactly *NOTHING* in the law *UNLESS* that intent is spelled out.
That's the different between legal terms and the spirit. And the promise of the GPL is to retain the spirit, to defend the freedoms.
Quoted 6 lines
Quoted 5 lines
That would be an unfortunate machine to have, but if Linux or some other GPLed software was not shipped in it, then I don't see how this is relevant to this discussion. It's not about the hardware, it's about the software in it, and about passing on the freedoms related with it.
Quoted 6 lines
Exactly. However, you are making it about the hardware by making the claim that "replacing a program, in its entirety, with another is a modification". It isn't. A modification is when you replace or change a *portion* of a program. By your logic I could write an operating system that is 100% binary compatible with Linux and I'd be *required* to release it under the GPL, because, even though it *replaces* Linux, it's still a "modification".
I'm not sure I agree with the reasoning here, but I'm already convinced that the argument about modification by replacement won't fly.
But then again I ask you: why do you think TiVO is making these hardware locks? What do they want to cause or stop?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}On 15/06/07, Alexandre Oliva <[email protected]> wrote:
[snip]
Quoted 4 lines
So what is it that makes hardware so different that it can be used as a trick to deny users freedoms, if other tricks can't?
[snip]
Why can't you understand that the GPL v2 is a *software* license, it doesn't cover hardware at all.
If I take some GPLv2 software, modify it and then distribute it on a CD-ROM and provide the source code as well, then I have complied with the terms of the license.
If I take the same software, make the same modifications and distribute the software in a ROM inside some piece of hardware, but still allow people access to a copy of the source code used to build whatever I put inside that ROM, then I've also complied with the license.
In neither case can you modify the copy on the hardware (be it ROM chip or CD-ROM), but that's not required by the license. As long as you have access to the source code it's OK. The license says nothing about you having to be able to update it on the hardware. The license only says you need access to the source code.
No one is taking away your freedom to change the source or redistribute it or whatever. The only thing locked hardware prevents you from doing is installing modified software on that specific piece of hardware, but that is completely outside the scope of the *software* license.
[snip]
Quoted 4 lines
But then again I ask you: why do you think TiVO is making these hardware locks? What do they want to cause or stop?
I can't know for a fact what TiVO wants, but I can guess.
1) Maybe they want to prevent you installing modified software on their hardware, then contacting them when you break it, costing them money in customer support etc.
2) Perhaps they don't want to risk being liable if you modify the software on their box in a way that allows you to use it as a means to break the law.
3) Maybe they don't want you to modify the software running on their hardware in such a way as to use the software to obtain intimate details about their hardware that could be used by a competitor to create a product superiour to theirs.
All quite valid reasons in my opinion.
Signature · 3 lines
Jesper Juhl <[email protected]> Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html
Quoted 2 lines
Why can't you understand that the GPL v2 is a *software* license, it doesn't cover hardware at all.
The GPLv2 is a copyright license not a software licence, indeed there is no such thing as a 'software licence'. It deals with the circumstances and manner in which you are permitted (by the author) to make copies of their work, to modify their work and in some cases to perform their work (plus other sundry rights). Copyright law doesn't care whether the object in question is as abstract as computer source code (providing it has been 'fixated' in some form) or a two hundred foot high art installation - or a combination of the two.
So irrespective of the whole pointless debate going on you are trying to draw lines that don't exist in the first place.
Quoted 1 line
I can't know for a fact what TiVO wants, but I can guess.
You could also do your research.
Quoted 1 line
All quite valid reasons in my opinion.
and all wrong.
Look up the owning and controlling interests in Tivo and you'll find the correct reason - stopping you doing evil things like keeping movies you've recorded or uploading them to the internet [which ironically of course is the entire effect of the whole 'convergence' thing]
Alan
Quoted 12 lines · Alan Cox
Quoted 2 lines
Why can't you understand that the GPL v2 is a *software* license, it doesn't cover hardware at all.
The GPLv2 is a copyright license not a software licence, indeed there is no such thing as a 'software licence'. It deals with the circumstances and manner in which you are permitted (by the author) to make copies of their work, to modify their work and in some cases to perform their work (plus other sundry rights). Copyright law doesn't care whether the object in question is as abstract as computer source code (providing it has been 'fixated' in some form) or a two hundred foot high art installation - or a combination of the two.
Right. My bad.
Quoted 7 lines
So irrespective of the whole pointless debate going on you are trying to draw lines that don't exist in the first place.
Quoted 1 line
I can't know for a fact what TiVO wants, but I can guess.
You could also do your research.
I have absolutely no idea where to go look something like that up :-(
Quoted 9 lines
Quoted 1 line
All quite valid reasons in my opinion.
and all wrong.
Look up the owning and controlling interests in Tivo and you'll find the correct reason - stopping you doing evil things like keeping movies you've recorded or uploading them to the internet [which ironically of course is the entire effect of the whole 'convergence' thing]
Hmm, wouldn't that be my guess nr. 2? A way to use the hardware to break the law...
Anway, the whole point of my post was mainly to /try/ and say that the GPL gives you a right to obtain source code for modifications, but it doesn't say anything about being able to run a compiled version of that source on any specific hardware.
Signature · 3 lines
Jesper Juhl <[email protected]> Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html
On Friday 15 June 2007 02:59:31 Jesper Juhl wrote:
<snip>
Quoted 9 lines
Quoted 8 lines
Quoted 1 line
All quite valid reasons in my opinion.
and all wrong.
Look up the owning and controlling interests in Tivo and you'll find the correct reason - stopping you doing evil things like keeping movies you've recorded or uploading them to the internet [which ironically of course is the entire effect of the whole 'convergence' thing]
Quoted 8 lines
Hmm, wouldn't that be my guess nr. 2? A way to use the hardware to break the law...
Anway, the whole point of my post was mainly to /try/ and say that the GPL gives you a right to obtain source code for modifications, but it doesn't say anything about being able to run a compiled version of that source on any specific hardware.
And you are correct. It is also clear, thanks to language directly in the GPLv2 itself, that there is no "intent" of the license to cover that situation.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 8 lines · Daniel Hazelton
Quoted 2 lines · On Friday 15 June 2007 02:59:31 Jesper Juhl wrote:
it doesn't say anything about being able to run a compiled version of that source on any specific hardware.
And you are correct. It is also clear, thanks to language directly in the GPLv2 itself, that there is no "intent" of the license to cover that situation.
You're again confusing legal terms with the intent. The legal terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that.
That said, the letter of the GPL explicitly says that the act of running the program is not restricted, and that this is outside the scope of the license. It's a copyright license, and per US law, running software is not regulated by copyright; this is not so elsewhere, and local interpretation indicates that the intent of the license is indeed to grant unlimited permission to run the program and modified versions thereof, based on the free software definition.
But then, when someone says "I won't let you run modified versions of this software on this hardware I'm selling you", is this not a further restriction on the exercise of the rights granted in the license?
And, per the spirit, if the manufacturer can still install and run modified versions of the software on that hardware, is it not failing to comply with the spirit of passing on all the rights that you have?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · On Friday 15 June 2007 15:37:04 Alexandre Oliva wrote:
Quoted 4 lines · Daniel Hazelton
Quoted 2 lines · On Friday 15 June 2007 02:59:31 Jesper Juhl wrote:
it doesn't say anything about being able to run a compiled version of that source on any specific hardware.
Quoted 7 lines
Quoted 3 lines
And you are correct. It is also clear, thanks to language directly in the GPLv2 itself, that there is no "intent" of the license to cover that situation.
You're again confusing legal terms with the intent. The legal terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that.
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing. Just as it is not the *intent* of RMS, the FSF or *ANY* person (or legal entity) that had a hand in crafting the GPLv2 or GPLv3 which is looked at when determining the "intent" of the license. It is the intent of the person and/or "legal entity" that has placed their work under said license.
What is so hard to understand about that ?
Quoted 7 lines
That said, the letter of the GPL explicitly says that the act of running the program is not restricted, and that this is outside the scope of the license. It's a copyright license, and per US law, running software is not regulated by copyright; this is not so elsewhere, and local interpretation indicates that the intent of the license is indeed to grant unlimited permission to run the program and modified versions thereof, based on the free software definition.
And that may be what courts in Brazil believe, independent of the fact that the license itself *intentionally* limits itself to "copying, distribution and modification". That legal decision, in fact, may not have been motivated by the actual belief that that was the intent of the license - it could (and, from looking at the situation and available facts, might actually have been) motivated by political reasons.
Quoted 4 lines
But then, when someone says "I won't let you run modified versions of this software on this hardware I'm selling you", is this not a further restriction on the exercise of the rights granted in the license?
In Brazil, because the courts there have rendered a judgment that the license requires the unlimited running of the covered work. But that is *unfairly* applying a license on a piece of software to the hardware on which it runs. Based on your logic the hardware manufacturer would have to enable people to run code compiled for an entirely different processor. Not that it matters in the least.
Quoted 3 lines
And, per the spirit, if the manufacturer can still install and run modified versions of the software on that hardware, is it not failing to comply with the spirit of passing on all the rights that you have?
Not in the least. They have the rights to "copy, modify and distribute" the "source code for a work". That is *EXACTLY* the set of rights they have to the code, and it is *EXACTLY* the set of rights they pass on. The GPL does not apply to any *BINARY* form of the work, except for the fact that you are required to provide the source code that was used to generate the binary.
The GPL *clearly* defines "source code" as:
"the preferred form of the work for making modifications to it" It goes on to state:
"For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable."
(Note that, since the "signing" of the TiVO kernel is part of the installation they *should* be including the script that does the signing. However, since the SHA1 key that is part of the signing process is *not* a "script" (even in the loosest possible definition of the term) they do not have to provide it.)
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 14 lines · Daniel Hazelton
Quoted 5 lines · On Friday 15 June 2007 15:37:04 Alexandre Oliva wrote:
Quoted 4 lines · Daniel Hazelton
Quoted 2 lines · On Friday 15 June 2007 02:59:31 Jesper Juhl wrote:
it doesn't say anything about being able to run a compiled version of that source on any specific hardware.
Quoted 7 lines
Quoted 3 lines
And you are correct. It is also clear, thanks to language directly in the GPLv2 itself, that there is no "intent" of the license to cover that situation.
You're again confusing legal terms with the intent. The legal terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that.
Quoted 6 lines
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing. Just as it is not the *intent* of RMS, the FSF or *ANY* person (or legal entity) that had a hand in crafting the GPLv2 or GPLv3 which is looked at when determining the "intent" of the license. It is the intent of the person and/or "legal entity" that has placed their work under said license.
No disagreement. You keep forgetting that I'm not here to say what Linux licensing means or doesn't mean.
I'm here to point out that GPLv3 does not break the spirit of the GPL.
Quoted 1 line
What is so hard to understand about that ?
/echo
Quoted 5 lines
(Note that, since the "signing" of the TiVO kernel is part of the installation they *should* be including the script that does the signing. However, since the SHA1 key that is part of the signing process is *not* a "script" (even in the loosest possible definition of the term) they do not have to provide it.)
It's not build script, it's just regular source code, indeed.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 13 lines · Alexandre Oliva
Quoted 5 lines
Quoted 4 lines
You're again confusing legal terms with the intent. The legal terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that.
Quoted 7 lines
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing. Just as it is not the *intent* of RMS, the FSF or *ANY* person (or legal entity) that had a hand in crafting the GPLv2 or GPLv3 which is looked at when determining the "intent" of the license. It is the intent of the person and/or "legal entity" that has placed their work under said license.
Quoted 2 lines
No disagreement. You keep forgetting that I'm not here to say what Linux licensing means or doesn't mean.
it is _you_ forgetting to read what you wrote just 1 mail ago above. _Read_ it:
"The legal terms provide an indication of the intent, but the
preamble, along with the free software definition it alludes to, do an even better job at that.".
Your point was totally bogus, and you have been pointed out that your point was bogus. And your answer - instead of admitting that you were wrong once again (i'm not even asking you to apologize for wasting our time) - to pretend that there is "no disagreement" and to patronize your discussion partner with a "you keep forgetting ..." phrase and a non-sequitor statement? How low can this discussion get? I'm truly amazed ...
Ingo
Quoted 18 lines · Ingo Molnar
Quoted 13 lines · Alexandre Oliva
Quoted 5 lines
Quoted 4 lines
You're again confusing legal terms with the intent. The legal terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that.
Quoted 7 lines
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing. Just as it is not the *intent* of RMS, the FSF or *ANY* person (or legal entity) that had a hand in crafting the GPLv2 or GPLv3 which is looked at when determining the "intent" of the license. It is the intent of the person and/or "legal entity" that has placed their work under said license.
Quoted 2 lines
No disagreement. You keep forgetting that I'm not here to say what Linux licensing means or doesn't mean.
Quoted 6 lines
it is _you_ forgetting to read what you wrote just 1 mail ago above. _Read_ it:
"The legal terms provide an indication of the intent, but the
preamble, along with the free software definition it alludes to, do an even better job at that.".
You still don't seem to get the difference between spirit and letter, and the difference between author of the license and licensor of the software (I guess I wasn't clear about this).
When I talk about the spirit of the GPL, I'm talking about how its authors designed it, how they described their motivations in the preamble, in the Free Software Definition, in speeches explaining it, etc.
When I talked about the meaning of the Linux licensing above, I'm talking about the intent of the Linux authors when choosing the GPLv2.
They aren't necessarily the same.
Some Linux authors may have read the preamble and understood something else. Some may have simply skipped it, and focused in the legal terms. Some even understood the legal terms in different senses. But they have all agreed to license their code under GPLv2, whatever the motivations each one of them had, and none of them has to match the spirit of the GPL.
Now, the spirit of the GPL, the intent behind its design, is something that may be entirely different. And when I say that GPLv3 didn't change the spirit of the GPL, I'm saying that from the perspective of someone who understands very deeply the philosphy and motivations behind it.
Please understand that these are two separate issues. That the spirit doesn't change, and that you don't share that spirit, or that you didn't think that was the spirit, doesn't justify a claim that the revision is changing the spirit. It's not. The spirit is, and has always been, to defend users' freedoms (the 4 freedoms of the Free Software definition), such that Free Software remains Free.
The fact that they can be different means that GPLv3 may not match the goals you had when you chose GPLv2 for your contributions. That's unfortunate. But this doesn't mean that GPLv3 is changing the spirit of the GPL. It's merely exposing that your goals don't match the spirit of the GPL. There's nothing wrong about this, GPLv2 is not a bad license, just like the LGPLv2 is not a bad license. They just don't defend users' freedoms as well as GPLv3 will do.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 32 lines · Alexandre Oliva
Quoted 30 lines · Ingo Molnar
Quoted 19 lines · Alexandre Oliva
Quoted 9 lines · Daniel Hazelton
Quoted 8 lines
Quoted 3 lines
Quoted 3 lines
It is also clear, thanks to language directly in the GPLv2 itself, that there is no "intent" of the license to cover that situation.
You're again confusing legal terms with the intent. The legal terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that.
Quoted 7 lines
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing. Just as it is not the *intent* of RMS, the FSF or *ANY* person (or legal entity) that had a hand in crafting the GPLv2 or GPLv3 which is looked at when determining the "intent" of the license. It is the intent of the person and/or "legal entity" that has placed their work under said license.
Quoted 2 lines
No disagreement. You keep forgetting that I'm not here to say what Linux licensing means or doesn't mean.
it is _you_ forgetting to read what you wrote just 1 mail ago above. _Read_ it:
"The legal terms provide an indication of the intent, but the
preamble, along with the free software definition it alludes to, do an even better job at that.".
[...]
Quoted 5 lines
Now, the spirit of the GPL, the intent behind its design, is something that may be entirely different. And when I say that GPLv3 didn't change the spirit of the GPL, I'm saying that from the perspective of someone who understands very deeply the philosphy and motivations behind it.
why do you start babbling about the GPLv3 when the false statement you made was about the GPLv2 - as visible _very_ clearly in the first quote above?? Again, as a reminder, this point was presented to you (see the quotes above), in the discussion about whether the Tivo is fine by the GPLv2 or not:
' It is also clear, thanks to language directly in the GPLv2
itself, that there is no "intent" of the license to cover that situation. '
to which you replied with this falsehood:
' You're again confusing legal terms with the intent. The legal
terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that. '
and you have simply been pointed out that what you say is trivially false - the simple legal fact is that the GPLv2 does not "embedd" more than what is its letter and, to the lesser extent that letter may be ambigious, what Linus' (and other copyright holders') intent is and was. Not RMS's external intentions or the "free software definition" you mention.
how about just simply admitting that you were wrong about this, instead of putting up (and beating down) yet another non-sequitor strawman argument? Is that really that hard and embarrasing to do? I've yet to see a _single_ instance of you admitting in this thread that "oops, it seems i was really wrong about this point. Sorry.". Or are you one of those perfect humans who are never wrong? ;-)
Ingo
Quoted 3 lines · Ingo Molnar
Again, as a reminder, this point was presented to you (see the quotes above), in the discussion about whether the Tivo is fine by the GPLv2 or not:
That's false. I've explicitly avoided discussions on whether the legal terms of GPLv2 permit tivoization.
What I've done was to discuss whether the tivoization was in line with mission (since you seem to have a problem grasping the notion of spirit of a license) of the license, because some people (yourself included) claimed GPLv3 changed the spirit.
That others tried to steer the discussion away from this apparently incomprehensible concept of spirit of the license, to the point of their and your getting utterly confused and making nonsensical claims about inconsistencies in my reasoning, moving it into the legal terms that I was not willing to discuss, is not my fault. My participation here was about intent, about spirit, about mission of the GPL.
As I've already made it clear, this doesn't necessarily match intent of people who chose GPLv2 as the license, and that's fine. Copyright holders know the spirit as they understood it, and that's how they meant to license their work.
But the spirit (mission) of the GPL is the one the FSF wrote about in the preamble. No amount of "but the legal terms say such and such" or "but I'm the licensor" distractions can change that.
Quoted 11 lines
' You're again confusing legal terms with the intent. The legal
terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that. 'and you have simply been pointed out that what you say is trivially false - the simple legal fact is that the GPLv2 does not "embedd" more than what is its letter and, to the lesser extent that letter may be ambigious, what Linus' (and other copyright holders') intent is and was. Not RMS's external intentions or the "free software definition" you mention.
Which shows you don't understand the notion of "spirit of license" (as opposed to intent of licensing, which I AFAIK invented today to try to dispell this confusion), and that the fact that the letter of the license doesn't have bearing about the intent of the author of the license, which is what the spirit of the license is about.
Quoted 1 line
how about just simply admitting that you were wrong about this,
I'm not. You are. Really. Until you understand the difference between "letter of license", "intent of licensor" and "spirit of license", you won't be able to understand this.
Quoted 2 lines
I've yet to see a _single_ instance of you admitting in this thread that "oops, it seems i was really wrong about this point. Sorry.".
I've done that, IIRC more than once. Sorry to disappoint you.
Quoted 1 line
Or are you one of those perfect humans who are never wrong? ;-)
:-) I wish :-)
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 11 lines · Alexandre Oliva
Which shows you don't understand the notion of "spirit of license" (as opposed to intent of licensing, which I AFAIK invented today to try to dispell this confusion), and that the fact that the letter of the license doesn't have bearing about the intent of the author of the license, which is what the spirit of the license is about.
Quoted 1 line
how about just simply admitting that you were wrong about this,
I'm not. You are. Really. Until you understand the difference between "letter of license", "intent of licensor" and "spirit of license", you won't be able to understand this.
I just can't stand this anymore.
Might I submit that nobody can be wrong in a hypothetical? The chances of the kernel changing to GPL v3 are the same as fort knox being ripped off. IM GLAD! This means I can continue to merge.
Folks I will summarize myself one last time and stop disrupting your list. I (personally) think enough time has been taken from productivity and I'm becoming awfully concerned about the integrity of Alexandre's ass. He's not here to irritate you, he's here to do what he thinks is right. I admire him for that considering the adversity.
(its not a mantra, you can keep reading, I promise!)
If I found advantage in a bug In Linux, and you fixed it, I have a choice to not apply the patch so that the bug remains useful to me.
The FSF found needs to produce another license to serve all of the people who have interest in the FSF. This does not mean you need to *apply* it. Its there if you want it, but its up to you.
Just as I told Linus, This is a kernel, not *&$(*# sed, and its a decision I'm happy was not taken lightly.
NOBODY should be influencing you with what you do with your contributions!! Make up your own minds in your own ways or have fun in the parrot cage where you sit in your own shit until someone else cleans it up. Who likes that?
I am asking as strongly as one can ask, *please* put this to bed. Continuing at this point can serve no useful purpose and everyone involved shares a common intent to be useful! Can we agree on that spirit?
Opinions have been expressed over, and over and over again. Everyone knows what everyone thinks.
We're all doing more harm than good. Who hates waste? I do.
Best,
--Tim
Quoted 22 lines · On Friday 15 June 2007 17:45:16 Alexandre Oliva wrote:
Quoted 14 lines · Daniel Hazelton
Quoted 5 lines · On Friday 15 June 2007 15:37:04 Alexandre Oliva wrote:
Quoted 4 lines · Daniel Hazelton
Quoted 2 lines · On Friday 15 June 2007 02:59:31 Jesper Juhl wrote:
it doesn't say anything about being able to run a compiled version of that source on any specific hardware.
Quoted 7 lines
Quoted 3 lines
And you are correct. It is also clear, thanks to language directly in the GPLv2 itself, that there is no "intent" of the license to cover that situation.
You're again confusing legal terms with the intent. The legal terms provide an indication of the intent, but the preamble, along with the free software definition it alludes to, do an even better job at that.
Quoted 6 lines
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing. Just as it is not the *intent* of RMS, the FSF or *ANY* person (or legal entity) that had a hand in crafting the GPLv2 or GPLv3 which is looked at when determining the "intent" of the license. It is the intent of the person and/or "legal entity" that has placed their work under said license.
Quoted 4 lines
No disagreement. You keep forgetting that I'm not here to say what Linux licensing means or doesn't mean.
I'm here to point out that GPLv3 does not break the spirit of the GPL.
And you have been given evidence that it does. It may not break the spirit as you, or anyone else that believes that the FSF is the singular fountain of truth, see it. But it does break the spirit as a large fraction of the populace see it. Why do they like GPL3? Because they don't like what TiVO did - they see it and say "but that's not fair", so when they see that GPL3 makes it a license violation they accept GPL3 without understanding the bigger picture.
To *anyone* who is considered an "Adult" in *ANY* nation and still wants to scream "its not fair" when something doesn't go their way or somebody does something they don't like I have this to say: Grow up! Life *ISN'T* fair - at all.
Quoted 11 lines
Quoted 1 line
What is so hard to understand about that ?
/echo
Quoted 5 lines
(Note that, since the "signing" of the TiVO kernel is part of the installation they *should* be including the script that does the signing. However, since the SHA1 key that is part of the signing process is *not* a "script" (even in the loosest possible definition of the term) they do not have to provide it.)
It's not build script, it's just regular source code, indeed.
Not even. If it's anything it's "input to a program". QED: It isn't covered by the license. Hell, give me a week and access to a signed TiVO kernel and I'll produce a signed kernel that is functionally equivalent - it won't run on a TiVO, but I'll have replicated the "signing script" and process that TiVO uses. It isn't until you start extending the definition of source code in strange ways that the key becomes "source code". Hell, for all *anyone* knows (that isn't employed by TiVO) they could enter the key *manually* - ie:
interactive input.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 3 lines · Daniel Hazelton
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing.
That's not true. Again, ianal, etc etc, but:
"Intent" *does* matter, and if you wrote down the intent at the time you entered some legal agreement, that actually also has non-zero bearing (as it can be used to _show_ intent more clearly than claiming fifteen years later "but, your honour, I _intended_ to do something else").
BUT!
Intent only matters if the contract reading itself isn't otherwise clear!
In other words, intent can never *override* the contract. But if there is some actual legally unclear area, the intent can be used by a judge/jury/arbitrator to _guide_ them to the decision. And the better documented that intent is, the better it is in that respect. Intent _after_ the fact is not very meaningful, is it?
So the preamble *does* matter legally, it just matters a lot less than the actual legal wording itself.
Put another way: the "intent" of the parties can be, and is, used to read the contract, but it's actually a mutual thing for a contract (since you have two or more parties in question - the intent of _one_ party doesn't really matter, if he cannot show that the other party agreed with that intent! And to get back to rms' personal history: _his_ intent, and _his_ personal history, and _his_ other documents don't matter for an agreement that doesn't include him).
So as such, a preamble can be used to show the intent of the parties, and thus to guide any reading.
Of course, so can "real life". If a certain reading of a contract doesn't make sense (and that has nothing to do with "intent"), then that obviously cannot be what either party really agreed to.
Oh, and documented intent can be used for laches, ie it can be a _defense_ against an accusation. If somebody has publicly stated some intent, he can't really complain later when somebody else tried to fulfill his intent, can he?
Linus
PS. "Intent" can obviously have a _totally_ different legal meaning too. The difference between "murder" and "manslaughter" is about pre-meditation
- ie "intent". But that's not an issue in licenses/contracts, except for a very specific _kind_ of "contract" - the kind the Mob puts out on people it wants to get rid of ;)
Quoted 10 lines · Linus Torvalds
Quoted 2 lines
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing.
That's not true. Again, ianal, etc etc, but:
"Intent" *does* matter, and if you wrote down the intent at the time you entered some legal agreement, that actually also has non-zero bearing (as it can be used to _show_ intent more clearly than claiming fifteen years later "but, your honour, I _intended_ to do something else").
yeah. What comes up periodically in GPLv3 discussions as 'proof' of what the GPL means are totally detached statements of the FSF and of RMS, often written a decade _after_ the GPL has been chosen for a license of the Linux kernel. (the whole anti-Tivo line was invented well after the fact.) And those statements have little bearing on the interpretation of the license of GPL-ed works. (unless, of course, the author of a GPL-ed work agrees with those statements and intends them to be his interpretation of the license.)
Ingo
Quoted 2 lines
And the preamble, not being part of the active portion of the license, has absolutely *ZERO* bearing. Just as it is not the *intent* of RMS, the FSF or
Wrong (again)
The pre-amble is incredibly important as is the intent of the license creator and even more so of the author.
When trying to solve a dispute the process starts with the legal equivalent of banging the two parties head together in the hope they see sense. If that fails then the legal wording is considered in detail. Where it is ambiguous the surrounding context is considered in order to understand the probable intent of the case.
Finally the stated intent of the author is considered in defence (The doctrine of estoppel), and at least in UK law whether their intent was honest (The doctrine of clean hands)
Legal disputes almost always end up about the things that are not clear (if they were clear one side would shut up and put up) so the preambles and statements are terribly important when this occurs, along with the context and history.
Thus for example the fact Linus has said he believes what Tivo does is ok means he can no longer sue Tivo for doing it. They are relying on his promise.
Alan
[ Damn. I moved you to my flamers list, and then I started reading it. I'm
addicted to flaming. Sue me. I really do enjoy it too much. If I didn't do software development, my full-time job would probably be to troll various internet sites and try to set up flame wars. I'm bad, I know.
It's an addiction. I'm not proud. ]
Quoted 3 lines · Alexandre Oliva
But then again I ask you: why do you think TiVO is making these hardware locks? What do they want to cause or stop?
Actually, they didn't want to lock down the hardware at all. The first versions of the Tivo was really quite hackable - and people started hacking them.
They were basically forced to add lockdown by the content vendors. You can call them evil for "caving in", but hey, it was their whole market. They really had no choice. Being a company actually limits you in some ways..
If you don't want to cave in to content providers, use a regular PC and soemthing like MythTV. You will probably also have to use the analogue hole, and will have a really hard time unscrambling digital cable TV signals of your own, but hey, you can see it as a challenge. At least in places where it's not illegal.
And yes, there are bad laws in the US. But blaming Tivo for them is ludicrous. And the *laws* won't get fixed by software licensing either, quite the reverse. The GPLv3 will just make free software that uses it *less* relevant in that space, rather than more.
For example, I'd rather have some GPLv2'd DVD player software that does *not* come with a de-css key (I can get that key myself quite easily), and that thus gets distributed in a "useless" form, than have a GPLv3'd DVD player that cannot be distributed at all, because it needs the magic unlocking key, and distributing the css key is illegal in some countries.
Or if I was an mplayer developer (which I'm not - so I have absolutely *zero* say in the mplayer license - please don't take this as anythign like that), I'd prefer for mplayer to be GPLv2, simply because that way I could see my software in some high-end (legal) DVD players that actually complied with the insane laws that exist. Sure, to comply with the laws and not get sued, they might have to limit the hardware, but hey, in other saner places of the world (like Finland), you can use the GPLv2'd software legally *without* those concerns.
See? The more permissive license actually allows more people to get involved. And the only thing that really *matters* (the source code) can be distributed and improved on by all these different people, even if some of them may have their hands bound by legal issues.
Btw, the same is true of things like FCC rules in the US. All that is evil does not come from the RIAA and MPAA. It's entirely possible that a cellphone manufacturer would have to lock down the control logic that sets the power levels - and that is something that is against the license of the GPLv3.
So the GPLv3 actually _hinders_ people who might otherwise help the community from helping, by making the license so strict that those people (who are nice people, but have their options limited by stupid laws and regulations) cannot use the GPLv3.
Linus
Quoted 3 lines
If you don't want to cave in to content providers, use a regular PC and soemthing like MythTV. You will probably also have to use the analogue hole
Only in the USA. In most of the world its considered quite normal that you can plug a USB disk into your PVR, save stuff to it and then plug it into your PC.
Alan
Quoted 1 line · Linus Torvalds
It's an addiction. I'm not proud.
I guess this makes it two of us :-(
Quoted 1 line
They were basically forced to add lockdown by the content vendors.
They can do that. They will still be able to do that with v3.
All they have to do is to throw away the keys that enable themselves to modify the code further.
Quoted 2 lines
For example, I'd rather have some GPLv2'd DVD player software that does *not* come with a de-css key
libdvdread and libdvdcss are separate packages.
Quoted 4 lines
So the GPLv3 actually _hinders_ people who might otherwise help the community from helping, by making the license so strict that those people (who are nice people, but have their options limited by stupid laws and regulations) cannot use the GPLv3.
Just like v2 hinders their many customers.
Are you so sure v2 is better in this regard?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 6 lines · Alexandre Oliva
Quoted 3 lines · Ingo Molnar
I think the proper limit is the boundary where the limit of the software is - because that's the only sane and globally workable way to stop the power-hungry.
Quoted 7 lines
But see, I'm not talking about getting permission to hack the hardware. I'm only talking about getting permission to hack the Free Software in it.
It's your position that mingles the issues and permits people to use the hardware to deprive users of freedom over the software that they're entitled to have.
where does this false sense of entitlement come from? The hardware maker ows you nothing but what is written into the GPLv2. Not more, not less.
(In fact, most hardware makers that utilize free software today give back _substantially more_ to the community than the license requires! For example they are currently the largest employers of free software developers - although nothing in the license forces them to do so. Why? Because the economic rules that the GPLv2 creates are healthy.)
you are not "entitled" to dictate the hardware's design (or any other copyrighted work's design), even if the license gives you the power to do so. By your argument we'd have to put the following items into the license too:
- free on-site training for free software developers about the hardware's inner workings. (It is justified to teach free software the same know-how as in-house engineers of the hardware maker. Without this, users are hindered in their freedom to use and effectively modify (fix) the software.)
- free access to all the hardware diagnostics tools that the hardware maker has. (Without that it might be impossible to modify the software as efficiently as the hardware maker's own engineers can do it.)
- free samples of the hardware to be sent to free software developers, upon request. (The hardware maker's own engineers have free access to samples. Otherwise free software users might not get the same level of driver support as the hardware maker can achieve.)
- free access to the hardware manufacturing equipment. (If i wish to modify the free software in a way that requires more RAM than the hardware has, i need access to the manufacturing equipment to produce a new version of the hardware that can run that free software. The hardware maker has this right and flexibility to modify the software, so i should have that same right too.)
see how quickly your argument becomes totally ludicrous, if brought to its logical conclusion?
This "right to modify" and "have the same rights as the hardware maker" arguments are _totally_ bogus, they were made up after the fact, just because quite apparently RMS had a fit over Tivo and started this verbal (and legal) vendetta. The FSF is now attempting to rewrite history and pretends that this "always was in the GPLv2" and applies this newly thought up concept to the GPLv3 in a way that substantially departs from the spirit of the GPLv2. Which spirit the GPLv2 explicitly promised to uphold in Section 9. Which could make any contrary section of the GPLv3 unenforceable, when applied to "GPLv2 or later" licensed software.
Ingo
Quoted 9 lines · Ingo Molnar
This "right to modify" and "have the same rights as the hardware maker" arguments are _totally_ bogus, they were made up after the fact, just because quite apparently RMS had a fit over Tivo and started this verbal (and legal) vendetta. The FSF is now attempting to rewrite history and pretends that this "always was in the GPLv2" and applies this newly thought up concept to the GPLv3 in a way that substantially departs from the spirit of the GPLv2. Which spirit the GPLv2 explicitly promised to uphold in Section 9. Which could make any contrary section of the GPLv3 unenforceable, when applied to "GPLv2 or later" licensed software.
That, BTW, is perhaps the worst problem with v2 (inherited by v3). WTF _is_ "the spirit of the license" and who gets to decide if two licenses are in the same spirit? As soon as we get to "well, original authors of the license are the final authority on that", we are in the "I've always said ..." country.
Look, humans _suck_ at revision control, especially that of our intentions and opinions. It doesn't even require malice, all ancedotes about spouses/mothers-in-law/etc. nonwithstanding.
We all easily fall into belief that we had always meant what we mean now; that even if we said something different, it was just a poor wording; that if we had known what we know now, we would certainly had come to the same conclusions we have come to now.
"In the same spirit" is just about the weakest requirement in that area. I.e. the most prone to drift, especially when one is an ideologist and thus has severely decayed integrity to start with. Call it a professional disease of crystal ball users - or a prerequisite for playing a visionary, if you will ;-/
Quoted 11 lines · Al Viro
Quoted 9 lines · Ingo Molnar
This "right to modify" and "have the same rights as the hardware maker" arguments are _totally_ bogus, they were made up after the fact, just because quite apparently RMS had a fit over Tivo and started this verbal (and legal) vendetta. The FSF is now attempting to rewrite history and pretends that this "always was in the GPLv2" and applies this newly thought up concept to the GPLv3 in a way that substantially departs from the spirit of the GPLv2. Which spirit the GPLv2 explicitly promised to uphold in Section 9. Which could make any contrary section of the GPLv3 unenforceable, when applied to "GPLv2 or later" licensed software.
Quoted 3 lines
That, BTW, is perhaps the worst problem with v2 (inherited by v3). WTF _is_ "the spirit of the license" and who gets to decide if two licenses are in the same spirit? [...]
yeah. I see this as: "RMS does not want to let go of the community". This clause amounts to "power to relicense" _vast_ amounts of free software and this is by far the worst problem with the "GPLv3 process".
The GPLv3 process was pretended to be "open", but regardless of what the "GPL comittees" said, in the end it was one person: the president of the FSF (Richard Stallman) who singlehandedly decided what went into the GPLv3 draft and what not. For example he singlehandedly has ignored all the criticism that the the "Tivo" section has received.
And note how hypocritic RMS's position is here. Where is that freedom when it comes to the licensing process? Why does RMS have more rights over modifications to the license than all the other free software developers have? Should not he give that freedom to others too? Shouldnt there be a fair and just election, a vote? You know, that democracy thing.
And with his current attitude he affects somewhere around of 1 billion lines of free software. Via a license that is just a few hunded lines long.
I believe RMS should accept the fact that most of that code was written without people having bought into his ideology, and he should accept _responsibility_ for the power he has acquired by genius or by accident (your choice) and he should try to _understand_ how those people tick - instead of trying to further his own personal agenda.
He shouldnt say what amounts to "oh, my original intent was this and that, if you didnt understand it and still wrote code and used the default 'or later' license, it's your damn fault".
He should accept that what happened happened, after he wrote 100,000 lines of original GNU code another ten thousand people wrote about a _ten thousand times more_ code. He should also accept that the "open source" community is about many other things, and it is alot more varied than his thinking is. He does not have to _like_ Tivo, but he should try to _understand_ them, and he should be compassionate about other people's right to have their own opinion and their own approaches to freedom. It is very clear that he has not even attempted to do that so far.
And the best way to start would be to significantly limit the 'same spirit' clause by putting in something like this:
'in the event of a section of this license being rules unenforceable by
a court of law the FSF has the option to modify the license in the most minimal fashion to make that section enforceable again'.
but this would mean RMS would have to give up power irreversibly. Will that ever happen?
Ingo
Quoted 4 lines · Ingo Molnar
* Alexandre Oliva <[email protected]> wrote:
you are not "entitled" to dictate the hardware's design (or any other copyrighted work's design),
Agreed.
Quoted 2 lines
By your argument we'd have to put the following items into the license too:
No, you're confusing two very different situations.
In the case of TiVO, it's getting out of its way to make sure users can't enjoy one of the freedoms that the license says it ought to pass on.
In the cases you mentioned, the company would have to get out of its way to put the other parties on equal grounds.
The former is bad, it's against the spirit of the license, it's a further restriction.
The latter would be nice to have, but it would be wrong to demand it.
You're picturing the difference between blocking the way such that you can't get there, and actually taking you there. What the GPL seeks is just that you don't get in the way.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 8 lines · Alexandre Oliva
Quoted 4 lines · Ingo Molnar
* Alexandre Oliva <[email protected]> wrote:
you are not "entitled" to dictate the hardware's design (or any other copyrighted work's design),
Agreed.
hey, that's progress. If you concede this single point then your arguments about the Tivo situation all fall like domino stones. Just watch it happen please:
Quoted 8 lines
Quoted 2 lines
By your argument we'd have to put the following items into the license too:
No, you're confusing two very different situations.
In the case of TiVO, it's getting out of its way to make sure users can't enjoy one of the freedoms that the license says it ought to pass on.
the GPLv2 license says no such thing, and you seem to be mighty confused about how software licenses work.
the GPL applies to software. It is a software license.
the Tivo box is a piece of hardware.
a disk is put into it with software copied to it already: a bootloader, a Linux kernel plus a handful of applications. The free software bits are available for download.
the Tivo box is another (copyrighted) work, a piece of hardware.
so how can, in your opinion, the hardware that Tivo produces, "take away" some right that the user has to the GPL-ed software? Because they distribute the software and the hardware in the same package, and because the hardware (as _ALL_ hardware on this planet) has certain limitations? It was _your_ choice to buy that particular hardware+software combination, with whatever limitations the hardware has. One such limitation of the hardware might be that its color is butt-ugly pink. Another limitation might be that the buttons on it are too small for elderly people to press. A third limitation might be that it's not a general purpose computer and that it's not freely programmable by the end user. Bugger, what did you expect? Why didnt you buy a green PVR? Why didnt you buy a PVR with larger buttons? Why didnt you buy a general purpose computer? Did perhaps the Tivo look like a nice general purpose PC to you when you bought it?
Quoted 2 lines
In the cases you mentioned, the company would have to get out of its way to put the other parties on equal grounds.
how about quoting what i wrote and rebutting it specifically if you disagree with it, instead of writing a non-sequitor generality? You are involved in compiler development, so you should have the mental ability to follow logical arguments and you should be able to conduct a meaningful and objective discussion. Lets look at one of the examples i gave you:
Quoted 4 lines
Quoted 4 lines
- free access to all the hardware diagnostics tools that the hardware maker has. (Without that it might be impossible to modify the software as efficiently as the hardware maker's own engineers can do it.)
by your argument, the user has some "right to modify the software", on that piece of hardware it bought which had free software on it, correct? By your argument, the "right to modify the software" becomes meaningless if you cannot soft-upgrade your Tivo, if you have solder off the ROM to install your own ROM with a bootloader that does not do the SHA1 check, correct?
But by that _very same argument_, you are hindered _much more_ by not having proper hardware diagnostics tools and no access to hardware specifications. If you dont know how the hardware works, you cannot fix bugs in the software. So by your argument, the user has an inherent right to get on equal footing with the hardware manufacturer to modify the software on that specific hardware? There's no ifs and when. "having to solder off the ROM" is a "restriction on modifiability" just as much as "having less information about the hardware's inner workings". In fact, ask any kernel developer, "having to solder off the ROM" is a lot _smaller_ restriction than "having no information about the hardware's inner workings".
Ingo
Quoted 2 lines
the GPLv2 license says no such thing, and you seem to be mighty confused about how software licenses work.
There is no such thing as a software licence. It is a copyright licence.
Quoted 1 line
the GPL applies to software. It is a software license.
You can GPL a new graphical logo you painted on your toilet seat, you can GPL hardware designs. It might not be a good licence for either but it is a valid licence.
Quoted 1 line
the Tivo box is a piece of hardware.
A Tivo box is a collection of literary works protected by copyright, designs protected by design patents and copyright, names and logos protected by trademarks, functionalities protected by patents and many more things. These are the things that restrict what I may do with it and how I may treat it. The collection of bits of metal and sand aren't really of relevance in terms of licencing.
If it was a generic housebrick with none of these things attached then within the law I can do what I like with it including copying it. A book is a copyright work but the copyright is about the literary work and the fact it is on paper is largely irrelevant. What determines your usage rights for those pieces of paper are the literary work it carries not the pieces of paper (unless made of a new patented paper material or similar)
Quoted 3 lines
a disk is put into it with software copied to it already: a bootloader, a Linux kernel plus a handful of applications. The free software bits are available for download.
Except the keys - which may nor may not be required depending upon how a court (not a mailing list) interprets the phrases
"The source code for a work means the preferred form of the work for making modifications to it"
and
"For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable."
If you ask the legal profession about this seriously the answer you get is bluntly "There is no caselaw I am aware of", which means that nobody knows. Obviously Tivo and their legal counsel have formed an opinion and have based their actions upon that opinion.
Quoted 1 line
the Tivo box is another (copyrighted) work, a piece of hardware.
You can't copyright hardware. Sorry but if you are going to try and have a detailed logical argument you need to start from a rigorous base point.
Alan
Quoted 5 lines · Alan Cox
Quoted 2 lines
the GPLv2 license says no such thing, and you seem to be mighty confused about how software licenses work.
There is no such thing as a software licence. It is a copyright licence.
a "software license" is a common shortcut for "copyright license for copies of software works". It's a commonly used phrase. In fact it is used by the FSF itself too:
http://www.fsf.org/licensing/essays/free-sw.html
"To decide whether a specific software license qualifies as a free
^^^^^^^^^^^^^^^^
software license, we judge it based on these criteria to determine whether it fits their spirit as well as the precise words."
Quoted 5 lines
Quoted 1 line
the GPL applies to software. It is a software license.
You can GPL a new graphical logo you painted on your toilet seat, you can GPL hardware designs. It might not be a good licence for either but it is a valid licence.
yeah - the GPL can be applied to most types of works recognized by copyright law.
Quoted 8 lines
Quoted 1 line
the Tivo box is a piece of hardware.
A Tivo box is a collection of literary works protected by copyright, designs protected by design patents and copyright, names and logos protected by trademarks, functionalities protected by patents and many more things. These are the things that restrict what I may do with it and how I may treat it. The collection of bits of metal and sand aren't really of relevance in terms of licencing.
If you are into technicalities then you fail to achieve that "rigorous base" by a wide margin. The Tivo box is not "a collection of literary works", it is a piece of matter, that also happens to contain fixated copies of literary (and other) works. The Tivo box is just one copy of those works - it is not "a collection of literary works". (Only if there was just a single Tivo box on the planet then could that box itself be meaningfully called a collection of works - a single and unique "master copy" of a work can be called the work itself.)
and that distinction, although fine, is very important. Look at GPLv2 section 0:
" 0. This License applies to any program or other work which contains a
notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. "
the work is not the copy! The work is a more 'abstract' entity. The word "copyright" comes straight from that: the right to create specific copies of the work. And that's another reason why it's nonsensical to suggest that somehow the GPLv2 gives us the right to influence the hardware environment that the copy of the kernel got fixated into. We dont. ( unless that hardware environment too is a copy of a GPL-ed work or it is a copy of a work that is a modification of or derives from a GPL-ed work - but in the Tivo case it isnt. It's a collection of copies of works and derivation does not "jump" from the harddisk to the hardware. )
More down the technicalities road: the Tivo box also contains many items that are not copies of works protected by copyright: common types of screws that are not original forms of expression that are creative enough enough to gain copyright protection. Or numbers painted on various places. Or computer-originated random output. Copies of works that have entered the public domain and thus are not under the scope of copyright protection.
Neither is the Tivo box "collection of functionalities protected by patents", if then it is an embodiment of a method and apparatus, which invention is under patent protection (there are other types of patents as well), or which invention might not be under patent protection but have a patent application pending. (which might or might not issue at the end of the patent application process.)
Quoted 9 lines
Quoted 3 lines
a disk is put into it with software copied to it already: a bootloader, a Linux kernel plus a handful of applications. The free software bits are available for download.
Except the keys - which may nor may not be required depending upon how a court (not a mailing list) interprets the phrases
"The source code for a work means the preferred form of the work for making modifications to it"
i think it is clear what is intended with this section: that for example using automatic tools to strip out comments and obfuscating the source code does not fly, because what matters is the _form of the software_ the developer usually makes his modifications under. So this in essence defines the scope of the actual source code that must be made available so that it works on a general purpose computer, not the specific hardware environment under which the developer operates.
so i believe it is a ... fairly creative bending of the wording of this section to attempt to extend it to the hardware environment. You dont get my ssh keys either [*] that i use on my test-boxes, and those test boxes are very much part of the preferred way for me to produce kernel patches. But you get my kernel patches for sure! [ that is, if they dont crash the testboxes :-) ] Am i violating the GPLv2?
Quoted 6 lines
and
"For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable."
i think it is pretty clear what this section intends: not allowing people to become cute by stripping out of makefiles from the source tarball. If someone else tries to run that software on a general purpose computer (which has enough physical resources for that purpose) it should be possible.
But to read this to require a toaster that a piece of free software came installed on to be modifiable by the licensee who choses to excercise his rights under the GPL, in the same way as the original developer was able to modify that toaster is ... quite creative too i think, and leads to many absurd results.
Quoted 4 lines
If you ask the legal profession about this seriously the answer you get is bluntly "There is no caselaw I am aware of", which means that nobody knows. Obviously Tivo and their legal counsel have formed an opinion and have based their actions upon that opinion.
i guess i'll take Linus' word that the FSF's own lawyers agreed that the distribution of the Tivo box does not break the GPLv2. (although the cynic in me might say that this could be a self-serving position on their behalf done for tactical reasons, to increase the perceived 'justification' for the GPLv3.)
Ingo
[*] actually, you can get them if you want to, because i very much trust
you :-)
Quoted 10 lines
Quoted 6 lines
A Tivo box is a collection of literary works protected by copyright, designs protected by design patents and copyright, names and logos protected by trademarks, functionalities protected by patents and many more things. These are the things that restrict what I may do with it and how I may treat it. The collection of bits of metal and sand aren't really of relevance in terms of licencing.
If you are into technicalities then you fail to achieve that "rigorous base" by a wide margin. The Tivo box is not "a collection of literary works", it is a piece of matter, that also happens to contain fixated
The physical matter is irrelevant. I am perfectly entitled to own, shape and fiddle with sand and bits of metal. If I wish to remove the software from the tivo, melt it down and cast the result into the shape of an obscene gesture and wear it at the tivo shareholder meeting so be it. At that point it would be my work made from melting the tivo that was the protected work - same matter.
Quoted 5 lines
copies of literary (and other) works. The Tivo box is just one copy of those works - it is not "a collection of literary works". (Only if there was just a single Tivo box on the planet then could that box itself be meaningfully called a collection of works - a single and unique "master copy" of a work can be called the work itself.)
Each copy is an instance of the work. My copy does not change its status, nor its legal situation if someone rounds up every other tivo and melts them down. I guess if you want to be pedantic the Tivo contains "an instance of the work"
Quoted 5 lines
the work is not the copy! The work is a more 'abstract' entity. The word "copyright" comes straight from that: the right to create specific copies of the work. And that's another reason why it's nonsensical to suggest that somehow the GPLv2 gives us the right to influence the hardware environment that the copy of the kernel got fixated into. We
We have every legal right to do so. I am perfectly permitted to try to grant you the right to reproduce my work only if you pay me $25 and the reproductions are provided in a silver box with flashy blue lights. I am perfectly permitted as author of a work to tell you "no". You as box maker are perfectly at liberty to tell me where to go stick my offer and just not use my work.
I can influence your hardware all I like. What I cannot do is influence you in any way if you decide not to take any action involving my copyright. Nor can I through copyright require certain kinds of condition (eg control other works on the same media) as that requires contract law and a proper contract, nor certain things that are deemed to be unlawful by the state (The GPL gives me the right to modify the code to break into the DoD, steal all their secrets and mail them to the Iraqi government, the law of the USA not unsuprisingly takes that right away).
Quoted 7 lines
More down the technicalities road: the Tivo box also contains many items that are not copies of works protected by copyright: common types of screws that are not original forms of expression that are creative enough enough to gain copyright protection. Or numbers painted on various places. Or computer-originated random output. Copies of works that have entered the public domain and thus are not under the scope of copyright protection.
And this matters because ?
Quoted 6 lines
Neither is the Tivo box "collection of functionalities protected by patents", if then it is an embodiment of a method and apparatus, which invention is under patent protection (there are other types of patents as well), or which invention might not be under patent protection but have a patent application pending. (which might or might not issue at the end of the patent application process.)
Ok I guess thats a question of level of abstraction, like being "an instance"
Quoted 4 lines
Quoted 2 lines
"The source code for a work means the preferred form of the work for making modifications to it"
i think it is clear what is intended with this section: that for example
The Lawyers don't. As experts in their field I generally trust their view on this. Also remember that lawyers assess legality not morality so there are other questions to ask than "will I get sued".
Quoted 5 lines
But to read this to require a toaster that a piece of free software came installed on to be modifiable by the licensee who choses to excercise his rights under the GPL, in the same way as the original developer was able to modify that toaster is ... quite creative too i think, and leads to many absurd results.
Agreed. But GPLv2 has many absurdities such as the way it handles copyight notices. It wasn't designed when GUI apps were the norm, it predates web hosted services and the GPL mobile phone was, I suspect, not on the drafters radar let alone in their pocket.
If my toaster is ROM based then it is difficult to argue that the preferred form for modification is anything but the code and usual build files. If the system is writable then it is possible if not reasonable to argue that the preferred form includes the information needed to load that modified image, or should do so.
What this means for the FSF goals if Tivo get up one morning and switch their system firmware to ROM however is interesting 8)
Alan
Quoted 2 lines · Alan Cox
What this means for the FSF goals if Tivo get up one morning and switch their system firmware to ROM however is interesting 8)
I'm not the FSF, and I don't speak for it, but it seems to me that this would be "mission accomplished".
The goal AFAIK is not to force people to enable others to hack the hardware or software to their liking. The goal is respect for the freedoms, it's not making it more difficult for others to do what you can and want to do. I guess it also goes under the name "Golden Rule". Others might phrase it as tit-for-tat, or quid pro quo.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · Alexandre Oliva
Quoted 2 lines · Alan Cox
What this means for the FSF goals if Tivo get up one morning and switch their system firmware to ROM however is interesting 8)
I'm not the FSF, and I don't speak for it, but it seems to me that this would be "mission accomplished".
This is insane. You start with a lofty ideal involving "freedom", and when you end up with a meaningless technicality (and in technical terms a change for the worse) you consider it a victory?
Yes, I know that this is what happens in politics (look here, our laws had an effect!), but I have more respect for you than to think you fall for these kinds of games. I do not wish to revise my opinion.
Bernd
Quoted 6 lines · Bernd Schmidt
Quoted 5 lines · Alexandre Oliva
Quoted 2 lines · Alan Cox
What this means for the FSF goals if Tivo get up one morning and switch their system firmware to ROM however is interesting 8)
Quoted 6 lines
Quoted 2 lines
I'm not the FSF, and I don't speak for it, but it seems to me that this would be "mission accomplished".
This is insane. You start with a lofty ideal involving "freedom", and when you end up with a meaningless technicality (and in technical terms a change for the worse) you consider it a victory?
It accomplishes the mission in that everyone is on the same grounds. Same freedom for everyone. If the vendor tries to keep a privilege over the software to itself, denying it to its customers, it's failing to comply with the spirit of the license. It's really this simple. Is this so hard to understand?
The goal is not to push vendors away from GPLed software. If they can't permit modification of the software, that's fine, they can still accomplish this.
What they can't do is deny it to customers while they retain it to themselves. This is unfair, this is wrong, and this disrespects users' freedoms. Therefore, the GPL should not permit it.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 21 lines · Alexandre Oliva
Quoted 6 lines · Bernd Schmidt
Quoted 5 lines · Alexandre Oliva
Quoted 2 lines · Alan Cox
What this means for the FSF goals if Tivo get up one morning and switch their system firmware to ROM however is interesting 8)
Quoted 6 lines
Quoted 2 lines
I'm not the FSF, and I don't speak for it, but it seems to me that this would be "mission accomplished".
This is insane. You start with a lofty ideal involving "freedom", and when you end up with a meaningless technicality (and in technical terms a change for the worse) you consider it a victory?
It accomplishes the mission in that everyone is on the same grounds. Same freedom for everyone. If the vendor tries to keep a privilege over the software to itself, denying it to its customers, it's failing to comply with the spirit of the license. It's really this simple. Is this so hard to understand?
Quoted 7 lines
The goal is not to push vendors away from GPLed software. If they can't permit modification of the software, that's fine, they can still accomplish this.
What they can't do is deny it to customers while they retain it to themselves. This is unfair, this is wrong, and this disrespects users' freedoms. Therefore, the GPL should not permit it.
How the hell does that improve the situation for users? Alexandre, please realize that you are preaching to non-believers. I realize that you have accepted the FSF credo, but if you want that conversation to go anywhere you have to separate the things you believe in from the things you can rationally explain. Apologetics of your variety is not going to cut it. _Can_ you separate the things relying on your beliefs from the things that can stand on their own? If you can't do that, please stop wasting everyone's time and bandwidth. It's a secular maillist; what any of us might happen to believe in is personal and frankly, none of your damn business.
Quoted 1 line · Al Viro
How the hell does that improve the situation for users?
Maybe it doesn't. How does it make it worse?
Maybe just providing an incentive for the vendor to respect users' freedoms will do the trick, and *some* vendors will do, while those who can't will keep the status quo.
And then we're likely to be better off.
Quoted 3 lines
I realize that you have accepted the FSF credo, but if you want that conversation to go anywhere you have to separate the things you believe in from the things you can rationally explain.
I've already explained what the spirit of the GPL is.
I've already explained that the anti-Tivoization provision is in line with it.
I've already asked in what sense Tivoization makes for a better tit-for-tat, and got no reply whatsoever, rational or otherwise.
I have already hinted at why it makes things worse.
You don't have to believe what I believe to analyze the arguments rationally, just like I don't have to believe what you believe to analyze your arguments rationally.
We may still get to different conclusions as to what is better, if we have different values guiding us.
But whatever conclusion you arrive at won't change the plain fact that Tivoization is against the spirit of the GPL, because it is a means to restrict users' freedoms that the GPL is designed to defend.
It's really this simple. I'm not trying to convince you of anything other than that the spirit of the GPL is not being changed at all. You don't have to agree with that spirit in order to accept this simple fact. And while people keep on spreading this lie, I'll be inclined to point out that it's false.
See, this is not about promoting GPLv3, or "pushing it down your throats", as some have claimed. This feeling is just a symptom of the high rejection for the FSF ideology, that appears to blind so many smart people from rational reasoning on matters that touch the FSF ideology.
This is not even about showing that the letter of GPLv2 prohibited Tivoization. My arguments concerning Tivoization were all about the spirit of the license, and unfortunately so many people seem unable to tell the spirit from the letter that they keep on moving the discussion to legal technicalities, and then they shoot straw men and feel happy that they shot an argument. But the argument stands untouched, and the straw man was already dead before. Wasted time.
As far as I'm concerned, Linux is released under GPLv2, and that's a good thing. It's unlikely to change. I wish it changed for better, but that's just me, and my contributions to Linux in term of code are really minimal. I have no say on that.
But as someone involved in the GPLv3 development, it saddens me when people lie about it. I feel it's my moral obligation to set the record straight. And that's what I've been trying to do.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 6 lines · Alexandre Oliva
Quoted 1 line · Al Viro
How the hell does that improve the situation for users?
Maybe it doesn't. How does it make it worse?
Now not even the vendor can upgrade the software in the hardware and fix problems for the user. The user loses.
Signature · 3 lines
Jesper Juhl <[email protected]> Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html
Quoted 3 lines · Jesper Juhl
Now not even the vendor can upgrade the software in the hardware and fix problems for the user. The user loses.
You're seeing that the wrong way.
The correct response is (and I quote from the manual, pick one talking point at random):
"This is a great step for freedom, as the users now have exactly the
same rights as the vendors."
"When we talk about free software, we don't talk about 'free as in
beer', we talk about 'free as in buggy and unfixable'"
"You're now at least no less free than anybody else!"
"Oh, except for the fact that those other people still design the
hardware you are using, and the programs you watch. But we have a plan for that too! We will make the GPLv4 outlaw Disney and Britney Spears!"
"In order to protect your freedoms, we sometimes have to take some
freedoms away. In particular, the freedom of critical thinking got revoked last year, because people were just too 'confused'"
"There are no American Infidels in Baghdad. Never!"
There's a long list of those things, but sadly I didn't have time to copy them all when I sneaked into the FSF main office in my ninja suit under the cover of darkness.
Linus
Quoted 3 lines · Linus Torvalds
"In order to protect your freedoms, we sometimes have to take some
freedoms away. In particular, the freedom of critical thinking got revoked last year, because people were just too 'confused'"
ITYM "upgraded to Loyalty To The Cause, which is the best antidote against doubt and confusion plaguing the so-called rationalists".
Quoted 9 lines · Jesper Juhl
Quoted 6 lines · Alexandre Oliva
Quoted 1 line · Al Viro
How the hell does that improve the situation for users?
Maybe it doesn't. How does it make it worse?
Now not even the vendor can upgrade the software in the hardware and fix problems for the user. The user loses.
Assuming the vendor's intent as for patching the software is to help the user. If the vendor doesn't want to let the user do that independently, why should this assumption hold?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 2 lines · Alexandre Oliva
I've already explained what the spirit of the GPL is.
No. You've explained one thing only: that you cannot see that people don't *agree* on the "spirit".
You think that there is only one "spirit", and that you own the code-book, and that your spirit is thus the only right one.
This is where we started. The same way you seem to think that "freedom" has only the meaning *you* and the FSF give it, and that somehow the spirit of the GPL includes the "four freedoms" that aren't even _mentioned_ in it.
THAT IS NOT TRUE.
But equally importantly, it's not even *relevant*. Nobody is suing the FSF for contract violation for changing the spirit. Yes, people have brought out the argument that the GPLv3 actually even changes the spirit, and you don't seem to realize that people can have different opinions. You just repeat YOUR OWN OPINION about the spirit over and over again.
But even if the spirit changes, so what? The GPL doesn't actually say "same in spirit". It says "similar in spirit", implying that the spirit is "similar".
In other words, your arguments are wholly irrelevant.
Quoted 2 lines
I've already explained that the anti-Tivoization provision is in line with it.
.. and we have already explained to you that it's irrelevant.
So let's get back to the *real* issue:
- The GPLv2 was ok with Tivo.
I really tried to explain to you *why* that was, but by now, I can't be bothered any more. Even if you cannot understand it, just accept it.
And if you have a hard time accepting it, just accept the fact that the FSF thinks Tivo cannot be sued, which is just another way of saying "they didn't actually break the license".
- *I* think Tivo is fine. Other people think Tivo is fine. Other people have told you they think what Tivo did is fine. Some people have even said that they don't like Tivo, but that they don't think the license should stop Tivo.
- The GPLv3 tries to stop Tivo.
Instead of mumbling about your spirit and feelings (I need to be a whole lot more drunk before I start caring), how about you look at those three statements, and then admit that you see why the people in bullet#2 think that
GPLv2 is a better license than GPLv3
I don't *care* how you mangle the "spirit of the GPLv2", because that was never the issue.
What I care about is that the GPLv3 is a _worse_license_ than GPLv2, and that I'd be stupid to select the worse of two licenses, wouldn't I?
So just stop *whining* about this.
The GPLv3 is the worse license, for anybody who thinks that what Tivo did should not be against the license. It really is that simple.
And again: you don't even have to *like* Tivo to realize that the license itself shouldn't try to spell out some anti-Tivo measures. As I've _also_ tried to explain, the anti-Tivo measures are actually more than just "anti Tivo". They are also "anti-anything-else-that-might-want-to-lock-down-a- specific-version-for-security-or-regulatory-reasons".
But in the end, it really hinges on a very simple concept:
- Not everybody thinks like you or agrees with you.
- In particular, the original copyright author in the kernel does *not* think like you, and *realized* that he doesn't really like the FSF religious agenda years and years ago, and made sure that the FSF cannot control the licensing of the Linux kernel.
If you don't accept those two simple *facts*, I don't know what's wrong with you.
Linus
Quoted 6 lines · Linus Torvalds
Quoted 2 lines · Alexandre Oliva
I've already explained what the spirit of the GPL is.
No. You've explained one thing only: that you cannot see that people don't *agree* on the "spirit".
They don't have to.
Just like nobody but you can tell why you chose the GPLv2, nobody but RMS can tell why he wrote the GPL. And the intent behind writing the GPL is what defines its spirit.
Quoted 2 lines
Yes, people have brought out the argument that the GPLv3 actually even changes the spirit,
And that's the point that I'm fighting here. It does not change the spirit. It's still ensuring that Free Software remains Free:
respecting and defending the four freedoms defined in the Free Software definition.
Quoted 4 lines
Quoted 2 lines
I've already explained that the anti-Tivoization provision is in line with it.
.. and we have already explained to you that it's irrelevant.
It is relevant. It was the point that my participation was intended to address.
I guess it is just too hard to accept that an FSFer could not be trying to force GPLv3 down your throat or some other such nonsense.
Quoted 1 line
- The GPLv2 was ok with Tivo.
There's disagreement about this, even among developers of the kernel Linux, and you know it.
I know you're always right and I pretend to respect that ;-), but why do you think your opinion should prevail over theirs? Don't you realize that they're as entitled as you are to enforce the license, and in the way *they* (not you) perceive and meant to license their code?
And then again, this is not something I'm overly concerned about. I probably don't have enough contributions to Linux for my take on it to make any difference whatsoever.
This is not the real issue at all. The real issue, that brought me here and got you to name calling me and the FSFs, is that there were false claims about the GPLv3 that I wanted to dispell, particularly the point about its changing the spirit. The anti-tivoization provisions are in the spirit of the GPL, and so much so that a number of people perceive them as already covered by GPLv2.
Quoted 1 line
- The GPLv3 tries to stop Tivo.
A minor nit, but no, it doesn't. It tries to stop the practice of tivoization on programs licensed under the GPLv3.
TiVo has a number of choices, and so do other tivoizers, even if they adopt software under the GPLv3.
Quoted 1 line
What I care about is that the GPLv3 is a _worse_license_ than GPLv2,
Even though anti-tivoization furthers the quid-pro-quo spirit that you love about v2, and anti-tivoization is your only objection to v3? That's what I don't understand. This is so obviously contradictory to me that it's almost funny, and you've so far dodged my questions about this and refrainied from commenting on this contradiction so much that it looks like it's a blind spot in your mind.
Quoted 1 line
I'd be stupid to select the worse of two licenses, wouldn't I?
Yes. That's precisely why I don't understand your stance. Because I expect you to be intelligent, but starting from your stated motivation for choosing GPLv2, and from the consequences of the anti-tivoization provisions, you'd satisfy your motivations better with v3.
Tivoization reduces the motivation for customers of tivoized devices to improve the software. You end up with contributions from the manufacturers alone, instead of from all the user community.
With explicit anti-tivoization provisions, you may very well lose contributions from some tivoizers, but for those who change their stance, you gain far more contributors. You don't need a lot of tivoizers to take the path of freedom for you to win big time in the bottom line that you posed as the only relevant one.
You see why I don't understand your position?
Quoted 2 lines
They are also "anti-anything-else-that-might-want-to-lock-down-a- specific-version-for-security-or-regulatory-reasons".
It's not, this is false. "Lock down" is permitted. It just won't work if the business model depends on modifying stuff behind the user's back. But other cases of "lock down" are permitted:
this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM).
Quoted 6 lines
- Not everybody thinks like you or agrees with you.
- In particular, the original copyright author in the kernel does *not* think like you, and *realized* that he doesn't really like the FSF religious agenda years and years ago, and made sure that the FSF cannot control the licensing of the Linux kernel.
I hereby acknowledge, one more time, that I accept these facts.
Since we're in such a good mood now, would you mind acknoledging some other simple facts, such that we can end this discussion?
- the spirit of the GNU GPL, written by RMS in the FSF, is to keep Free Software Free, respecting and defending the freedoms of users of software licensed under the GPL
It can serve other goals, and some people, yourself included, chose it for other reasons, but the intent, the spirit of the license is what its author intended it to be, just like the intent behind each contribution to Linux is whatever the author of the contribution meant it to be.
- GPLv3 does not change this spirit
On the contrary, it advances this spirit. Given that defending these freedoms is the mission of the FSF, it's no surprise that it does revise the GPL to do it. It's not like it has a choice.
- Tivoization reduces the incentive for contributions
Customers of tivoized devices can't enjoy or even test the benefits of their modifications to the software on the device where the modifications would be most useful for them.
- anti-tivoization provisions encourage tivoizers who can respect users' freedoms to do so
If the choice is that or not being able to change the software for the user or adopting another platform, they may very well choose this option, and then you get not only more users and mind-share, but also far more contributors, and the community of developers that forms around the product benefits the former-tivoizer as well.
Are these so hard to accept?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 4 lines · On Sunday 17 June 2007 01:09:01 Alexandre Oliva wrote:
Quoted 3 lines · Linus Torvalds
Quoted 1 line · Alexandre Oliva
I've already explained what the spirit of the GPL is.
Quoted 8 lines
Quoted 2 lines
No. You've explained one thing only: that you cannot see that people don't *agree* on the "spirit".
They don't have to.
Just like nobody but you can tell why you chose the GPLv2, nobody but RMS can tell why he wrote the GPL. And the intent behind writing the GPL is what defines its spirit.
"Charging for programs is an crime against humanity"
Quoted 7 lines
Quoted 2 lines
Yes, people have brought out the argument that the GPLv3 actually even changes the spirit,
And that's the point that I'm fighting here. It does not change the spirit. It's still ensuring that Free Software remains Free:
respecting and defending the four freedoms defined in the Free Software definition.
See, you can't even keep the FSF's "Free Software Definition" and its inherent "religion" out of the discussion. Sure, the FSF can claim that the GPL is intended as a way to "defend" the "Four Freedoms" defined *BY* *THEM*, but unless alluded to in the license, the only bearing it can have, anywhere, is on the "intent" of the license, as seen by the FSF. And if the "ability to run a "covered work" on any piece of hardware" is "freedom 0" then binary distribution is in violation of the "spirit" - I can't run an x86 binary on a PPC. Isn't that a "designed in hardware restriction" that violates the "spirit" of the license ?
Quoted 41 lines
Quoted 4 lines
Quoted 2 lines
I've already explained that the anti-Tivoization provision is in line with it.
.. and we have already explained to you that it's irrelevant.
It is relevant. It was the point that my participation was intended to address.
I guess it is just too hard to accept that an FSFer could not be trying to force GPLv3 down your throat or some other such nonsense.
Quoted 1 line
- The GPLv2 was ok with Tivo.
There's disagreement about this, even among developers of the kernel Linux, and you know it.
I know you're always right and I pretend to respect that ;-), but why do you think your opinion should prevail over theirs? Don't you realize that they're as entitled as you are to enforce the license, and in the way *they* (not you) perceive and meant to license their code?
And then again, this is not something I'm overly concerned about. I probably don't have enough contributions to Linux for my take on it to make any difference whatsoever.
This is not the real issue at all. The real issue, that brought me here and got you to name calling me and the FSFs, is that there were false claims about the GPLv3 that I wanted to dispell, particularly the point about its changing the spirit. The anti-tivoization provisions are in the spirit of the GPL, and so much so that a number of people perceive them as already covered by GPLv2.
Quoted 1 line
- The GPLv3 tries to stop Tivo.
A minor nit, but no, it doesn't. It tries to stop the practice of tivoization on programs licensed under the GPLv3.
TiVo has a number of choices, and so do other tivoizers, even if they adopt software under the GPLv3.
Quoted 8 lines
Quoted 1 line
What I care about is that the GPLv3 is a _worse_license_ than GPLv2,
Even though anti-tivoization furthers the quid-pro-quo spirit that you love about v2, and anti-tivoization is your only objection to v3? That's what I don't understand. This is so obviously contradictory to me that it's almost funny, and you've so far dodged my questions about this and refrainied from commenting on this contradiction so much that it looks like it's a blind spot in your mind.
Not in the least. They still have to release their changes if they want to sell their devices. Or are you so blinded by your belief that the FSF and RMS can't be wrong that you can't understand that?
Quoted 6 lines
Quoted 1 line
I'd be stupid to select the worse of two licenses, wouldn't I?
Yes. That's precisely why I don't understand your stance. Because I expect you to be intelligent, but starting from your stated motivation for choosing GPLv2, and from the consequences of the anti-tivoization provisions, you'd satisfy your motivations better with v3.
It is only *YOUR* opinion that the GPLv3 is the better license. As Linus has explained, he doesn't share your viewpoint.
Quoted 3 lines
Tivoization reduces the motivation for customers of tivoized devices to improve the software. You end up with contributions from the manufacturers alone, instead of from all the user community.
No, it reduces their motivation to improve the software on *those* devices. If they like the software enough to actually download the source, they probably also liked it enough to install it on their computer *AND* will modify it to make it work better on their computer.
Quoted 5 lines
With explicit anti-tivoization provisions, you may very well lose contributions from some tivoizers, but for those who change their stance, you gain far more contributors. You don't need a lot of tivoizers to take the path of freedom for you to win big time in the bottom line that you posed as the only relevant one.
With your argument about reduced motive shotten down this portion falls apart.
Quoted 20 lines
You see why I don't understand your position?
Quoted 2 lines
They are also "anti-anything-else-that-might-want-to-lock-down-a- specific-version-for-security-or-regulatory-reasons".
It's not, this is false. "Lock down" is permitted. It just won't work if the business model depends on modifying stuff behind the user's back. But other cases of "lock down" are permitted:
this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM).
Quoted 6 lines
- Not everybody thinks like you or agrees with you.
- In particular, the original copyright author in the kernel does *not* think like you, and *realized* that he doesn't really like the FSF religious agenda years and years ago, and made sure that the FSF
cannot control the licensing of the Linux kernel.
Quoted 8 lines
I hereby acknowledge, one more time, that I accept these facts.
Since we're in such a good mood now, would you mind acknoledging some other simple facts, such that we can end this discussion?
- the spirit of the GNU GPL, written by RMS in the FSF, is to keep Free Software Free, respecting and defending the freedoms of users of software licensed under the GPL
Agreed. The disagreement is about what that spirit is. I feel that its spirit is in the free and open exchange of ideas, as personified by the software people write. I *ALSO* feel that it's spirit lies in the phrase "do whatever you want with the software as long - but if you add your own ideas to it, give them back to the people like your inspiration was given to you."
You, the FSF and, apparently, RMS, feel it is about the "Four Freedoms" as defined by RMS. I'm quite sure that my view is much more common among the people that *DON'T* think that the FSF and RMS are never wrong.
Quoted 11 lines
It can serve other goals, and some people, yourself included, chose it for other reasons, but the intent, the spirit of the license is what its author intended it to be, just like the intent behind each contribution to Linux is whatever the author of the contribution meant it to be.
- GPLv3 does not change this spirit
On the contrary, it advances this spirit. Given that defending these freedoms is the mission of the FSF, it's no surprise that it does revise the GPL to do it. It's not like it has a choice.
Again, that is *your* version of the "spirit".
Quoted 6 lines
- Tivoization reduces the incentive for contributions
Customers of tivoized devices can't enjoy or even test the benefits of their modifications to the software on the device where the modifications would be most useful for them.
I agree to the "can't enjoy or test" bits. But I don't believe that it reduces anything. Personally I feel that anything that exposes people to "Free Software" is a *BONUS*.
Quoted 4 lines
- anti-tivoization provisions encourage tivoizers who can respect users' freedoms to do so
Quoted 8 lines
If the choice is that or not being able to change the software for the user or adopting another platform, they may very well choose this option, and then you get not only more users and mind-share, but also far more contributors, and the community of developers that forms around the product benefits the former-tivoizer as well.
Are these so hard to accept?
Yes. Because a number of your "facts" are massively flawed. Now, please, you've proven to me that you can't, in fact, do any *objective* thinking about this topic.
When you are ready to drop your pre-conceptions and think *objectively* about the topic, come back and talk. Until then, go away.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 8 lines · Daniel Hazelton
Quoted 7 lines · On Sunday 17 June 2007 01:09:01 Alexandre Oliva wrote:
Quoted 5 lines · Linus Torvalds
Quoted 1 line · Alexandre Oliva
I've already explained what the spirit of the GPL is.
No. You've explained one thing only: that you cannot see that people don't *agree* on the "spirit".
Quoted 7 lines
Quoted 5 lines
They don't have to.
Just like nobody but you can tell why you chose the GPLv2, nobody but RMS can tell why he wrote the GPL. And the intent behind writing the GPL is what defines its spirit.
"Charging for programs is an crime against humanity"
Assuming he actually said that, I have no doubt that it would pre-dates by far even the Free Software Definition, let alone the GPL.
Quoted 2 lines
See, you can't even keep the FSF's "Free Software Definition" and its inherent "religion" out of the discussion.
Of course not. That's what the spirit of the GPL is all about. And the spirit of the GPL is what the discussion is all about for me.
Quoted 4 lines
Sure, the FSF can claim that the GPL is intended as a way to "defend" the "Four Freedoms" defined *BY* *THEM*, but unless alluded to in the license, the only bearing it can have, anywhere, is on the "intent" of the license, as seen by the FSF.
Exactly! And since the *Free* *Software* Foundation wrote the license, and documented the goals in the preamble, referring to keeping *free* *software* *free*, it is quite safe to say that this *is* indeed the intent, the spirit of the GPL.
Quoted 5 lines
And if the "ability to run a "covered work" on any piece of hardware is "freedom 0" then binary distribution is in violation of the "spirit" - I can't run an x86 binary on a PPC. Isn't that a "designed in hardware restriction" that violates the "spirit" of the license ?
It's not. freedom and ability have two very different meanings.
Freedom to run the software for any purpose means that people won't stop you from doing that. It may take you some work, such as porting the software, rebuilding it, etc. But if, at the end of that effort, you find that it will run on your development machine, but not in a machine where the original software runs on, and that's because the manufacturer imposed prohibitions on running unauthorized versions of the software, then the manufacturer of the hardware is very clearly disrespecting your freedom #0 WRT that software.
Demanding the ability to run the software for any purpose, without any effort whatsoever, would indeed be nonsensical.
(BTW, covered work is a legal term, only present in the legal portion of the license, which I'm actively avoiding, because I'm not a lawyer, and my point is about the spirit. but I'm sure I wrote that before ;-)
Quoted 8 lines
Quoted 6 lines
Quoted 1 line
I'd be stupid to select the worse of two licenses, wouldn't I?
Yes. That's precisely why I don't understand your stance. Because I expect you to be intelligent, but starting from your stated motivation for choosing GPLv2, and from the consequences of the anti-tivoization provisions, you'd satisfy your motivations better with v3.
It is only *YOUR* opinion that the GPLv3 is the better license.
Are you even reading what I write?
Quoted 9 lines
Quoted 3 lines
Tivoization reduces the motivation for customers of tivoized devices to improve the software. You end up with contributions from the manufacturers alone, instead of from all the user community.
No, it reduces their motivation to improve the software on *those* devices. If they like the software enough to actually download the source, they probably also liked it enough to install it on their computer *AND* will modify it to make it work better on their computer.
Sure, but that's a different point. They could do that with or without tivoization.
The point is that, if they have an issue with the program in the device, and they'd like to improve it, but they find that they won't be able to use their modification to get the device to do what they want, they're less likely to make the change.
Now multiply this by all customers, and see how much you're losing by permitting tivoization, assuming that at least some tivoizers would change their minds towards respecting users' freedoms, if faced with an anti-tivoization licensing provision.
Quoted 1 line
With your argument about reduced motive shotten down this portion falls apart.
A distraction doesn't shoot down an argument.
Quoted 5 lines
Quoted 3 lines
- the spirit of the GNU GPL, written by RMS in the FSF, is to keep Free Software Free, respecting and defending the freedoms of users of software licensed under the GPL
Agreed. The disagreement is about what that spirit is.
What is the 'agreed' supposed to mean, then? ;-)
Quoted 5 lines
I feel that its spirit is in the free and open exchange of ideas, as personified by the software people write. I *ALSO* feel that it's spirit lies in the phrase "do whatever you want with the software as long - but if you add your own ideas to it, give them back to the people like your inspiration was given to you."
You're entitled to have these motivations to release software under GPLv2, or any other license that you believe furthers these goals.
But you have no say whatsoever on what intent RMS had when he wrote the GPL, just like he has no say whatsoever on what intent you have when you choose the GPL for your program.
Quoted 2 lines
You, the FSF and, apparently, RMS, feel it is about the "Four Freedoms" as defined by RMS.
s/feel/know/
Quoted 2 lines
I'm quite sure that my view is much more common among the people that *DON'T* think that the FSF and RMS are never wrong.
Others can choose the GPL for other reasons. There's nothing wrong with this.
What's wrong is to then complain that the GPL is changing the spirit, just because the revised version allegedly no longer matches their reasons.
Quoted 7 lines
Quoted 5 lines
- GPLv3 does not change this spirit
On the contrary, it advances this spirit. Given that defending these freedoms is the mission of the FSF, it's no surprise that it does revise the GPL to do it. It's not like it has a choice.
Again, that is *your* version of the "spirit".
Again, this is the FSF version of the spirit, the only one that matters as far as "the spirit of the GPL" is concerned.
Quoted 7 lines
Quoted 5 lines
- Tivoization reduces the incentive for contributions
Customers of tivoized devices can't enjoy or even test the benefits of their modifications to the software on the device where the modifications would be most useful for them.
I agree to the "can't enjoy or test" bits.
See, no "argument shot down" ;-)
Quoted 1 line
But I don't believe that it reduces anything.
It reduces the incentive for these users to collaborate.
Or, if you want to put it in a positive tone, the ability to enjoy and test their modifications would grow the number of contributors. More "giving back in kind". More tit-for-tat.
Quoted 2 lines
Personally I feel that anything that exposes people to "Free Software" is a *BONUS*.
I don't think tivoized software qualifies as Free Software any more.
Sure, they still get the sources and that's still Free Software, but the tivoized binaries aren't.
Quoted 1 line
Yes. Because a number of your "facts" are massively flawed.
Because you say so? Even though you agree with points after claiming to have shot them down? Now that's rational!
Quoted 2 lines
Now, please, you've proven to me that you can't, in fact, do any *objective* thinking about this topic.
/me hands Daniel a mirror
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 11 lines · On Sunday 17 June 2007 02:27:42 Alexandre Oliva wrote:
Quoted 10 lines · Daniel Hazelton
Quoted 9 lines · On Sunday 17 June 2007 01:09:01 Alexandre Oliva wrote:
Quoted 5 lines · Linus Torvalds
Quoted 1 line · Alexandre Oliva
I've already explained what the spirit of the GPL is.
No. You've explained one thing only: that you cannot see that people don't *agree* on the "spirit".
They don't have to.
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
Just like nobody but you can tell why you chose the GPLv2, nobody but RMS can tell why he wrote the GPL. And the intent behind writing the GPL is what defines its spirit.
"Charging for programs is an crime against humanity"
Assuming he actually said that, I have no doubt that it would pre-dates by far even the Free Software Definition, let alone the GPL.
The quote actually pre-dates the FSF
Quoted 5 lines
Quoted 2 lines
See, you can't even keep the FSF's "Free Software Definition" and its inherent "religion" out of the discussion.
Of course not. That's what the spirit of the GPL is all about. And the spirit of the GPL is what the discussion is all about for me.
The "spirit" is no different than "intent". Different words that mean the exact same thing.
Quoted 9 lines
Quoted 4 lines
Sure, the FSF can claim that the GPL is intended as a way to "defend" the "Four Freedoms" defined *BY* *THEM*, but unless alluded to in the license, the only bearing it can have, anywhere, is on the "intent" of the license, as seen by the FSF.
Exactly! And since the *Free* *Software* Foundation wrote the license, and documented the goals in the preamble, referring to keeping *free* *software* *free*, it is quite safe to say that this *is* indeed the intent, the spirit of the GPL.
The intent of the GPL, as seen by the FSF, *DOESN'T* *MATTER* *AT* *ALL* when the software isn't licensed by the FSF. Or did you forget that part of the discussion?
Quoted 7 lines
Quoted 5 lines
And if the "ability to run a "covered work" on any piece of hardware is "freedom 0" then binary distribution is in violation of the "spirit" - I can't run an x86 binary on a PPC. Isn't that a "designed in hardware restriction" that violates the "spirit" of the license ?
It's not. freedom and ability have two very different meanings.
This isn't what you've argued before. The hardware doesn't allow me to run the software, so its a designed-in limitation on the freedom of the end-user.
Quoted 8 lines
Freedom to run the software for any purpose means that people won't stop you from doing that. It may take you some work, such as porting the software, rebuilding it, etc. But if, at the end of that effort, you find that it will run on your development machine, but not in a machine where the original software runs on, and that's because the manufacturer imposed prohibitions on running unauthorized versions of the software, then the manufacturer of the hardware is very clearly disrespecting your freedom #0 WRT that software.
But a PPC binary won't run on an x86 either :)
No, I'm in agreement with you here. But I'm smart enough to not buy something that does this to me.
Quoted 8 lines
Demanding the ability to run the software for any purpose, without any effort whatsoever, would indeed be nonsensical.
(BTW, covered work is a legal term, only present in the legal portion of the license, which I'm actively avoiding, because I'm not a lawyer, and my point is about the spirit. but I'm sure I wrote that before ;-)
I use it because that is the term used in the GPLv2. And since the GPLv3 (dd4) is no longer specific to software "covered work" is the best choice.
Quoted 10 lines
Quoted 8 lines
Quoted 6 lines
Quoted 1 line
I'd be stupid to select the worse of two licenses, wouldn't I?
Yes. That's precisely why I don't understand your stance. Because I expect you to be intelligent, but starting from your stated motivation for choosing GPLv2, and from the consequences of the anti-tivoization provisions, you'd satisfy your motivations better with v3.
It is only *YOUR* opinion that the GPLv3 is the better license.
Are you even reading what I write?
Yes. But you are interpreting Linus' intentions using your own preconceived beliefs, rather than looking at the situation objectively.
Quoted 4 lines
Quoted 4 lines
Quoted 3 lines
Tivoization reduces the motivation for customers of tivoized devices to improve the software. You end up with contributions from the manufacturers alone, instead of from all the user community.
Quoted 8 lines
Quoted 5 lines
No, it reduces their motivation to improve the software on *those* devices. If they like the software enough to actually download the source, they probably also liked it enough to install it on their computer *AND* will modify it to make it work better on their computer.
Sure, but that's a different point. They could do that with or without tivoization.
Exactly. Tivoization doesn't make a damned bit of difference.
Quoted 4 lines
The point is that, if they have an issue with the program in the device, and they'd like to improve it, but they find that they won't be able to use their modification to get the device to do what they want, they're less likely to make the change.
They complain to the manufacturer, file a report with a consumer watchdog agency and start advising people against buying the device.
Quoted 4 lines
Now multiply this by all customers, and see how much you're losing by permitting tivoization, assuming that at least some tivoizers would change their minds towards respecting users' freedoms, if faced with an anti-tivoization licensing provision.
Apply the same logic to my above statement and tell me - how much money does the company lose ?
Quoted 4 lines
Quoted 2 lines
With your argument about reduced motive shotten down this portion falls apart.
A distraction doesn't shoot down an argument.
Quite thoroughly shot down.
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
- the spirit of the GNU GPL, written by RMS in the FSF, is to keep Free Software Free, respecting and defending the freedoms of users of software licensed under the GPL
Agreed. The disagreement is about what that spirit is.
What is the 'agreed' supposed to mean, then? ;-)
It means that I agree that the GPL is about "respecting and defending freedoms"
Quoted 6 lines
Quoted 5 lines
I feel that its spirit is in the free and open exchange of ideas, as personified by the software people write. I *ALSO* feel that it's spirit lies in the phrase "do whatever you want with the software as long - but if you add your own ideas to it, give them back to the people like your inspiration was given to you."
Quoted 6 lines
You're entitled to have these motivations to release software under GPLv2, or any other license that you believe furthers these goals.
But you have no say whatsoever on what intent RMS had when he wrote the GPL, just like he has no say whatsoever on what intent you have when you choose the GPL for your program.
I never claimed I did. I was just pointing out that your belief that there is only one "spirit" of the license is complete and utter BS.
Quoted 4 lines
Quoted 2 lines
You, the FSF and, apparently, RMS, feel it is about the "Four Freedoms" as defined by RMS.
s/feel/know/
I said exactly what I meant.
Quoted 9 lines
Quoted 2 lines
I'm quite sure that my view is much more common among the people that *DON'T* think that the FSF and RMS are never wrong.
Others can choose the GPL for other reasons. There's nothing wrong with this.
What's wrong is to then complain that the GPL is changing the spirit, just because the revised version allegedly no longer matches their reasons.
It isn't, because from my own, and other peoples, viewpoint it *IS*.
Quoted 10 lines
Quoted 7 lines
Quoted 5 lines
- GPLv3 does not change this spirit
On the contrary, it advances this spirit. Given that defending these freedoms is the mission of the FSF, it's no surprise that it does revise the GPL to do it. It's not like it has a choice.
Again, that is *your* version of the "spirit".
Again, this is the FSF version of the spirit, the only one that matters as far as "the spirit of the GPL" is concerned.
So you admit you believe what the FSF says?
Quoted 10 lines
Quoted 7 lines
Quoted 5 lines
- Tivoization reduces the incentive for contributions
Customers of tivoized devices can't enjoy or even test the benefits of their modifications to the software on the device where the modifications would be most useful for them.
I agree to the "can't enjoy or test" bits.
See, no "argument shot down" ;-)
Quoted 7 lines
Quoted 1 line
But I don't believe that it reduces anything.
It reduces the incentive for these users to collaborate.
Or, if you want to put it in a positive tone, the ability to enjoy and test their modifications would grow the number of contributors. More "giving back in kind". More tit-for-tat.
If the platform doesn't allow the running of modified binaries, why would the modifications matter? Sure, TiVO might like them - hell, they might even pay for them - but would anyone else?
So modifications for that "closed execution" platform might suffer, but that is the *ONLY* thing that will suffer.
Quoted 7 lines
Quoted 2 lines
Personally I feel that anything that exposes people to "Free Software" is a *BONUS*.
I don't think tivoized software qualifies as Free Software any more.
Sure, they still get the sources and that's still Free Software, but the tivoized binaries aren't.
Yes, they are. But only because the law states that they are.
Quoted 4 lines
Quoted 1 line
Yes. Because a number of your "facts" are massively flawed.
Because you say so? Even though you agree with points after claiming to have shot them down? Now that's rational!
I agreed with *portions* of some of the statements. However, did I actually say which facts I found flawed?
Quoted 4 lines
Quoted 2 lines
Now, please, you've proven to me that you can't, in fact, do any *objective* thinking about this topic.
/me hands Daniel a mirror
I've been looking at this objectively the entire time. That you don't understand that is just more proof that you aren't.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 3 lines · Daniel Hazelton
The intent of the GPL, as seen by the FSF, *DOESN'T* *MATTER* *AT* *ALL* when the software isn't licensed by the FSF. Or did you forget that part of the discussion?
You're mixing up spirit of license with intent of licensing (or something else, I don't remember exactly the term I invented to try to make this distinction clear)
The former is the intent of the FSF for writing the GPL.
The latter is the motivations of the copyright holder to license his work under that license.
The FSF has no say on the latter. The copyright holders have no say on the former.
When people claim the GPL changed its spirit, they're claiming the FSF changed its intent.
It didn't.
Clear now?
Quoted 11 lines
Quoted 7 lines
Quoted 5 lines
And if the "ability to run a "covered work" on any piece of hardware is "freedom 0" then binary distribution is in violation of the "spirit" - I can't run an x86 binary on a PPC. Isn't that a "designed in hardware restriction" that violates the "spirit" of the license ?
It's not. freedom and ability have two very different meanings.
This isn't what you've argued before. The hardware doesn't allow me to run the software, so its a designed-in limitation on the freedom of the end-user.
If it doesn't allow you to run the software *because* of a designed-in limitation on the freedom of the end-user, then it's a disrespect for the freedom. I've never changed my position in this regard. Maybe I wasn't clear, or you misunderstood, or the network corrupted the bits in transit ;-)
Quoted 2 lines
No, I'm in agreement with you here. But I'm smart enough to not buy something that does this to me.
tit for tat or each for oneself?
Quoted 6 lines
Quoted 4 lines
(BTW, covered work is a legal term, only present in the legal portion of the license, which I'm actively avoiding, because I'm not a lawyer, and my point is about the spirit. but I'm sure I wrote that before ;-)
I use it because that is the term used in the GPLv2.
Not in the preamble, which discusses the spirit I'm talking about.
Quoted 2 lines
And since the GPLv3 (dd4) is no longer specific to software "covered work" is the best choice.
The preamble of GPLv3 is pretty much the same.
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines
Quoted 3 lines
Tivoization reduces the motivation for customers of tivoized devices to improve the software. You end up with contributions from the manufacturers alone, instead of from all the user community.
No, it reduces their motivation to improve the software on *those* devices. If they like the software enough to actually download the source, they probably also liked it enough to install it on their computer *AND* will modify it to make it work better on their computer.
Quoted 4 lines
Quoted 2 lines
Sure, but that's a different point. They could do that with or without tivoization.
Exactly. Tivoization doesn't make a damned bit of difference.
For this case. But it's not the case where I claimed it made a difference.
Quoted 7 lines
Quoted 4 lines
The point is that, if they have an issue with the program in the device, and they'd like to improve it, but they find that they won't be able to use their modification to get the device to do what they want, they're less likely to make the change.
They complain to the manufacturer, file a report with a consumer watchdog agency and start advising people against buying the device.
And? How is this going to achieve more contributions in kind?
Quoted 7 lines
Quoted 4 lines
Now multiply this by all customers, and see how much you're losing by permitting tivoization, assuming that at least some tivoizers would change their minds towards respecting users' freedoms, if faced with an anti-tivoization licensing provision.
Apply the same logic to my above statement and tell me - how much money does the company lose ?
Dunno. Not much? I know I complain to hardware manufacturers that ship broken BIOSes, to no avail. It doesn't look like they care, or that it makes a difference.
And then, I'm not talking about a case in which the thing is broken (in which case the user might have a real case)
Think of improvements I'd like to make, that I probably won't do because the hardware won't let me run it. So you'll never see those contributions I and all the other untivoized users could make, and you won't ever know what you're missing.
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines
With your argument about reduced motive shotten down this portion falls apart.
A distraction doesn't shoot down an argument.
Quite thoroughly shot down.
If you say so, it must be right, in spite of all objective evidence, eh? :-)
Quoted 10 lines
Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
- the spirit of the GNU GPL, written by RMS in the FSF, is to keep Free Software Free, respecting and defending the freedoms of users of software licensed under the GPL
Agreed. The disagreement is about what that spirit is.
What is the 'agreed' supposed to mean, then? ;-)
It means that I agree that the GPL is about "respecting and defending freedoms"
But still, somehow, that's not its spirit, you say. I don't get it.
Quoted 5 lines
Quoted 3 lines
What's wrong is to then complain that the GPL is changing the spirit, just because the revised version allegedly no longer matches their reasons.
It isn't, because from my own, and other peoples, viewpoint it *IS*.
Which shows you don't know what the spirit really is. It is, and it has always been, what you agreed above that the GPL was about.
What others think the spirit is doesn't affect what the spirit is. It just says what others think the spirit is, and how off the mark they are in their assessment of the spirit of the license (= intent behind its creation)
Quoted 6 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
- GPLv3 does not change this spirit
On the contrary, it advances this spirit. Given that defending these freedoms is the mission of the FSF, it's no surprise that it does revise the GPL to do it. It's not like it has a choice.
Quoted 6 lines
Quoted 4 lines
Quoted 1 line
Again, that is *your* version of the "spirit".
Again, this is the FSF version of the spirit, the only one that matters as far as "the spirit of the GPL" is concerned.
So you admit you believe what the FSF says?
I guess this should be pretty obvious that I believe this, yes.
I don't know that I can make a general assertion about my believing what the FSF says, but I don't remember having had reasons to disbelieve it.
Quoted 10 lines
Quoted 3 lines
Or, if you want to put it in a positive tone, the ability to enjoy and test their modifications would grow the number of contributors. More "giving back in kind". More tit-for-tat.
If the platform doesn't allow the running of modified binaries, why would the modifications matter? Sure, TiVO might like them - hell, they might even pay for them - but would anyone else?
So modifications for that "closed execution" platform might suffer, but that is the *ONLY* thing that will suffer.
You're looking at the downside. Look at the upside: all the contributions you'll get from users of formerly-tivoized platforms. Being able to put their efforts to work in their own self-interest, on their own hardware, they are likely to give more back "in kind" than the vendor ever could.
And if the vendor doesn't go that way, what you lose are the limited contributions of that vendor.
Big potential win, small potential loss. Sounds like a no-brainer to me, really.
Quoted 9 lines
Quoted 7 lines
Quoted 2 lines
Personally I feel that anything that exposes people to "Free Software" is a *BONUS*.
I don't think tivoized software qualifies as Free Software any more.
Sure, they still get the sources and that's still Free Software, but the tivoized binaries aren't.
Yes, they are. But only because the law states that they are.
I can't make sense of what you're saying.
Are you disputing that tivoized binaries are non-Free Software?
What law could possibly support this claim?
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines
Now, please, you've proven to me that you can't, in fact, do any *objective* thinking about this topic.
/me hands Daniel a mirror
I've been looking at this objectively the entire time.
So you say. I believe I have too. That we disagree doesn't mean any of us is not being objective. It may mean we have different backgrounds, we're talking past each other, we're not understanding each other, and a number of other possibilities.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 9 lines · Alexandre Oliva
Quoted 2 lines · Linus Torvalds
No. You've explained one thing only: that you cannot see that people don't *agree* on the "spirit".
They don't have to.
Just like nobody but you can tell why you chose the GPLv2, nobody but RMS can tell why he wrote the GPL. And the intent behind writing the GPL is what defines its spirit.
Given that a number of people who don't buy into FSF ideology (let's call them "open source proponents" to contrast them with the "free software people") have concluded that the GPLv2 achieves their personal goals, and have chosen the GPLv2 as the license for their projects, I'd argue that the spirit that is embodied in the GPLv2 is actually a larger thing than what the FSF intended, and more inclusive.
When these same people now disagree with the GPLv3, it indicates that something has been lost, and the spirit of the _license_ has changed. The _intention_ behind writing the license may or may not have been the same (who can tell, after 20-odd years?), but this is separate from the spirit embodied in the license itself - the latter has, in my mind anyway, clearly been changed. You might prefer to say "clarified", but it comes down to the same thing.
But personally, I find the discussion about whether the spirit changed or not somewhat beside the point and not very interesting. What's really going to cause problems is the fact that the actual wording took a turn for the worse.
Bernd
Quoted 11 lines · Bernd Schmidt
Quoted 9 lines · Alexandre Oliva
Quoted 2 lines · Linus Torvalds
No. You've explained one thing only: that you cannot see that people don't *agree* on the "spirit".
They don't have to.
Just like nobody but you can tell why you chose the GPLv2, nobody but RMS can tell why he wrote the GPL. And the intent behind writing the GPL is what defines its spirit.
Quoted 6 lines
Given that a number of people who don't buy into FSF ideology (let's call them "open source proponents" to contrast them with the "free software people") have concluded that the GPLv2 achieves their personal goals, and have chosen the GPLv2 as the license for their projects, I'd argue that the spirit that is embodied in the GPLv2 is actually a larger thing than what the FSF intended, and more inclusive.
This sounds like a good argument, but it doesn't hold water.
Consider this: We manufacture bread toasters and sell them in the market with great success. They're big and bulky. So the engineers work on reducing its size, but in a way they can still fit perfectly a slice of bread. When we launch bread toaster, people complain that this new product cannot toast bagels any more, that we've changed the spirit of the bread toaster.
See? Just because you could use it for other purposes doesn't make the intent behind it any different.
Quoted 2 lines
When these same people now disagree with the GPLv3, it indicates that something has been lost, and the spirit of the _license_ has changed.
It just shows that they've never agreed with the spirit of the license in the first place. They just saw it could do something else, and used it for this reason. There's nothing wrong about this.
What's wrong is to complain that those who introduced the license with a specific and public intent, and that advancing that intent with a new revision of the license, are changing the intent.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · Alexandre Oliva
Quoted 2 lines
What I care about is that the GPLv3 is a _worse_license_ than GPLv2,
Even though anti-tivoization furthers the quid-pro-quo spirit that you love about v2, and anti-tivoization is your only objection to v3?
You apparently do not understand "quid-pro-quo".
Another way of stating it might be "same for same".
A third way of stating it is "software for software". No, the romans never said that, but I just did, to make it just more obvious that the whole point is that you are expected to answer IN KIND!
I do *not* ask for hardware access.
I do *not* ask for money.
And the reason I'm harping on "money" is that "money" is something *different* from what I give out. I give out software. I don't expect money in return.
Money is *irrelevant*. It's allowed (and certainly much appreciated), but it's not required.
See? Can you agree with that? Can you agree that that is actually part of what the whole "open source" spirit is all about (I'll avoid the word "free software", since you have defined it so rigorously personally that it makes no sense any more).
Now, replace "money" with "access to the hardware", and read the exact *same* sentences again:
And the reason I'm harping on "access to hardware" is that "access to hardware" is something *different* from what I give out. I give out software. I don't expect access to hardware in return.
Access to hardware is *irrelevant*. It's allowed (and certainly much appreciated), but it's not required.
See?
Exact same words. Exact same spirit. Just using "access to hardware" instead of "money".
You have been showing that you have a really hard time understanding that very *simple* argument.
Quoted 3 lines
Quoted 1 line
I'd be stupid to select the worse of two licenses, wouldn't I?
Yes. That's precisely why I don't understand your stance.
If you don't understand it after the above, I really can only say:
"You are either terminally stupid, or you're not allowing yourself
to see an obvious argument, because it destroys your world-view".
The latter is very possible. It's a very human thing. It's why apparently a lot of people in the US have a hard time believing in evolution. Are they terminally stupid? Yeah, that is quite possible. But it is also possible that they are of average intelligence, and they just cannot mentally _afford_ to follow the argument - it destroys the silyl stories they heard as children, and requires them to think too hard about the veracity of the source.
Linus
PS. Since some people talked about the game theory aspects of "tit-for-tat", I'd like to point out that what is usually considered an even *better* strategy than "tit-for-tat" is actually "tit-for-tat with forgiveness".
In particular, "tit-for-tat with forgiveness" is considered better when there is ambiguity (like "communication difficulties" - does that sound familiar?) in the encouter. You allow some leeway, and don't always retaliate!
So the FSF is DOING THE WRONG THING! They are turning "tit-for-tat" not into "tit-for-tat with forgiveness", but into "tit-for-tat with preemptive strikes".
That is a *LOSING* strategy in game theory. So a game theorist could very well argue with good reason to believe he is right that the GPLv3 is actually a worse license even from a purely theoretical standpoint!
Quoted 7 lines · Linus Torvalds
Quoted 5 lines · Alexandre Oliva
Quoted 2 lines
What I care about is that the GPLv3 is a _worse_license_ than GPLv2,
Even though anti-tivoization furthers the quid-pro-quo spirit that you love about v2, and anti-tivoization is your only objection to v3?
Quoted 7 lines
You apparently do not understand "quid-pro-quo".
Another way of stating it might be "same for same".
A third way of stating it is "software for software". No, the romans never said that, but I just did, to make it just more obvious that the whole point is that you are expected to answer IN KIND!
Yes. And this was precisely what meant when I wrote "quid-pro-quo" above.
Quoted 6 lines
If you don't understand it after the above, I really can only say:
"You are either terminally stupid, or you're not allowing yourself
to see an obvious argument, because it destroys your world-view".The latter is very possible. It's a very human thing.
/me hands Linus a mirror
Serious, what's so hard to understand about:
no tivoization => more users able to tinker their formerly-tivoized computers => more users make useful modifications => more contributions in kind
?
Sure, there's a downside too:
no tivoization => fewer contributions from manufacturers that demand on tivoization
My perception is that the first easily dominates the second, and so you are better off without tivoization.
Quoted 12 lines
it is also possible that they are of average intelligence, and they just cannot mentally _afford_ to follow the argument - it destroys the silyl stories they heard as children, and requires them to think too hard about the veracity of the source.
PS. Since some people talked about the game theory aspects of "tit-for-tat", I'd like to point out that what is usually considered an even *better* strategy than "tit-for-tat" is actually "tit-for-tat with forgiveness".
Quoted 8 lines
In particular, "tit-for-tat with forgiveness" is considered better when there is ambiguity (like "communication difficulties" - does that sound familiar?) in the encouter. You allow some leeway, and don't always retaliate!
So the FSF is DOING THE WRONG THING! They are turning "tit-for-tat" not into "tit-for-tat with forgiveness", but into "tit-for-tat with preemptive strikes".
Wrong. It enables copyright holders to decide whether forgiveness is appropriate, rather than forcing them to forgive. Being forced to forgive deception is not tit-for-tat, and it's a losing strategy.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}On 17/06/07, Alexandre Oliva <[email protected]> wrote:
[snip]
Quoted 9 lines
Serious, what's so hard to understand about:
no tivoization => more users able to tinker their formerly-tivoized computers => more users make useful modifications => more contributions in kind
?
Quoted 9 lines
Sure, there's a downside too:
no tivoization => fewer contributions from manufacturers that demand on tivoization
My perception is that the first easily dominates the second, and so you are better off without tivoization.
I have to disagree.
Let's say I'm the owner of a company selling some device that uses a GPLv2 OS and some GPLv2 applications to do the job. Let's say that for some reason I don't want the end users of my device to tinker with the software inside my device. Obviously I release the source for any modifications I may have made, but I use the hardware to prevent users from installing modified versions on the device (basically I TiVO'ize the device).
Now I think you can agree to these things being positive:
- My use of GPLv2 software in this device results in my employees being exposed to open source software at work (who knows, some may even start using such software at home as a result). A good thing.
- The source code with my own modifications that I distribute as required by the GPLv2 can potentially be of use to other developers working on other GPLv2 software and those other developers are free to use those modifications. Also a good thing.
- When creating marketing material for my device I'll most likely include information about the fact that I'm using WhatEverOS that is GPL'ed as well as other Open Source components. This in turn results in many people becoming aware that such software exist. I have to say that this is also a good thing.
- When dealing with hardware companies supplying bits and pieces for my device I'll probably push for components that already have open source drivers, so my partners will find out there is value in having open source drivers for their stuff and hopefully end up supporting that. Yet another good thing.
- If I end up being happy with my choice of GPLv2 OS & GPLv2 apps there's a, not insignificant, chance that I'll start helping out with the development of those components or maybe sponsor other developers with money to do so. Again we have a positive benefit.
The only downside is that the end user purchasing the device can't install modified versions of the software on it.
Now let's try it in a GPLv3 universe. Since I can no longer create my device without having to allow the end user to install modified software on it I probably turn to some closed source OS like WinCE or QNX (or maybe I use BSD, but now I can't be bothered to give my modifications back any longer since their license says I don't have to). I'm still happy, I can still sell my device and make money just like I used to, but:
- My employees are no longer exposed to Open Source software at work. Bummer, no new users from there.
- I no longer distribute the source for whatever modifications I make in-house. Damn, that's some nice software the community is missing out on.
- Now my adverticing material is highlighting the use of some proprietary OS and apps. What a drag, no more free adverticing for open source software - actually, quite the opposite.
- The next time I call my suppliers I just ask them to provide me whatever closed source drivers they have for BigCommercialClosedSourceOS and I'm happy since the drivers probably work just fine. Hmm, no more pressure on hardware companies to engage in developing open drivers.
- Now that I'm paying all this money in licenses for all this proprietary software and not using any open source software at all, there's zero to no chance I'll throw any money, developers or whatever at any open source projects. Dang, we just lost some corporate funding.
So, as I see it, tivoization isn't all bad. In fact I think the positives outweigh the negatives by quite a large margin. Sure, with GPLv3 you may win the battle and force some manufacturers away from your software if they can't/won't open their hardware up to end user modifications. But you'll lose the war in that you'll be killing the momentum that GPL'ed software has currently since you'll be driving a lot of players away from it.
I'd rather have the few benefits we get from some company using GPL'ed software in tivoized hardware than get nothing at all because the GPLv3 drives that company into the arms of some proprietary vendor.
Signature · 3 lines
Jesper Juhl <[email protected]> Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html
Quoted 7 lines · Jesper Juhl
Let's say I'm the owner of a company selling some device that uses a GPLv2 OS and some GPLv2 applications to do the job. Let's say that for some reason I don't want the end users of my device to tinker with the software inside my device. Obviously I release the source for any modifications I may have made, but I use the hardware to prevent users from installing modified versions on the device (basically I TiVO'ize the device).
BTW: Another reason a vendor might lock down the device is for security. For example, Juniper routers (which now run a significant portion of the "core" of the Internet) run FreeBSD on the routing engine. They include several GNU software utilities (for example gawk, diff, and gdb). Starting with JUNOS 7.6 (IIRC), end-users can no longer build and run their own binaries on the routing engine. This means that the GPLv2 code cannot be modified in-place (similar to TiVo altough done using different means).
The reason is that if there ever is a security hole in the routing engine software (FreeBSD kernel, OpenSSH, etc.), it would be a really bad thing if crackers could load arbitrary software (rootkits, spam software, etc.) directly on Internet core routers. If you think spam zombies on cable modems or DSL are bad, imagine them on 100 megabit links!
Quoted 18 lines · Chris Adams
Quoted 7 lines · Jesper Juhl
Let's say I'm the owner of a company selling some device that uses a GPLv2 OS and some GPLv2 applications to do the job. Let's say that for some reason I don't want the end users of my device to tinker with the software inside my device. Obviously I release the source for any modifications I may have made, but I use the hardware to prevent users from installing modified versions on the device (basically I TiVO'ize the device).
BTW: Another reason a vendor might lock down the device is for security. For example, Juniper routers (which now run a significant portion of the "core" of the Internet) run FreeBSD on the routing engine. They include several GNU software utilities (for example gawk, diff, and gdb). Starting with JUNOS 7.6 (IIRC), end-users can no longer build and run their own binaries on the routing engine. This means that the GPLv2 code cannot be modified in-place (similar to TiVo altough done using different means).
Quoted 6 lines
The reason is that if there ever is a security hole in the routing engine software (FreeBSD kernel, OpenSSH, etc.), it would be a really bad thing if crackers could load arbitrary software (rootkits, spam software, etc.) directly on Internet core routers. If you think spam zombies on cable modems or DSL are bad, imagine them on 100 megabit links!
To be fair here, this could also be accomplished by having to flip a physical switch on the router, especially if you did something funky like:
[---] push this button for a 5 minute access pass to upload new
software through physical cable port 1.
More complex, but not unreasonable.
Bron.
Quoted 8 lines · Bron Gondwana
To be fair here, this could also be accomplished by having to flip a physical switch on the router, especially if you did something funky like:
[---] push this button for a 5 minute access pass to upload new
software through physical cable port 1.More complex, but not unreasonable.
Well, there is no restriction on putting files on the routing engine's storage devices (flash and hard drive); it is running OpenSSH, so scp/sftp work fine, and you can drop to a shell easily. The restriction is that the kernel won't run unsigned binaries.
Also, flipping physical switches is pretty much an unreasonable expectation for core router operation. These are often in other locations, sometimes other telcos' central offices (where you have to pay to have "remote hands" do something and then hope they don't screw it up). You can easily go the entire life of a device where the primary operators never physically see the device.
Signature · 3 lines
Chris Adams <[email protected]> Systems and Network Administrator - HiWAAY Internet Services I don't speak for anybody but myself - that's enough trouble.
Quoted 15 lines · Chris Adams
Quoted 8 lines · Bron Gondwana
To be fair here, this could also be accomplished by having to flip a physical switch on the router, especially if you did something funky like:
[---] push this button for a 5 minute access pass to upload new
software through physical cable port 1.More complex, but not unreasonable.
Well, there is no restriction on putting files on the routing engine's storage devices (flash and hard drive); it is running OpenSSH, so scp/sftp work fine, and you can drop to a shell easily. The restriction is that the kernel won't run unsigned binaries.
Quoted 6 lines
Also, flipping physical switches is pretty much an unreasonable expectation for core router operation. These are often in other locations, sometimes other telcos' central offices (where you have to pay to have "remote hands" do something and then hope they don't screw it up). You can easily go the entire life of a device where the primary operators never physically see the device.
Every server I run is like that, but if something is important enough I can remote control a robot over to push the button for me (actually, I think they implement this under the hood by having a human read the ticket I submit and go push the button for me manually, but that could be my imagination. So long as the button gets pushed the black box is functioning)
Bron.
Quoted 6 lines · Chris Adams
The reason is that if there ever is a security hole in the routing engine software (FreeBSD kernel, OpenSSH, etc.), it would be a really bad thing if crackers could load arbitrary software (rootkits, spam software, etc.) directly on Internet core routers. If you think spam zombies on cable modems or DSL are bad, imagine them on 100 megabit links!
Not sure if it's a good example, keep in mind that at the first exploitable software bug any hardware DRM breaks apart.
But since you made a BSD-embedded example, this shows how the only really important thing is that by using linux instead of BSD, they can't make huge improvements or important security bugfixes to the routing engine, without us being able to incorporate them in our "home firewalls", that's the whole difference with BSD and it explains the spirit of the gpl pretty well and in the end why linux by definition can receive more contributions and in turn be technically superior. Whatever the vendor does with the gpl code is generally up to him, and if it uses the closed approach it'll allow somebody else to sell a "open" router (potentially at an higher price). Economy
101. The worry that nobody will step in and sell an "open" equivalent is a red herring. Infact I wouldn't be so certain that openmoko would exist if the current linux cellphones would be already totally open! Now I know this all probably sounds boring talk, but I think it's much closer to reality than the prospect of a trusted computing and/or DRM apocalypse.
Quoted 10 lines · Jesper Juhl
On 17/06/07, Alexandre Oliva <[email protected]> wrote:
[snip]Quoted 5 lines
Serious, what's so hard to understand about:
no tivoization => more users able to tinker their formerly-tivoized computers => more users make useful modifications => more contributions in kind
I have to disagree.
Your analysis stopped at the downside of prohibiting tivoization. You didn't analyze the potential upsides, so you may indeed come to different conclusions, and they may very well be wrong.
It's very human to look only at the potential downside of an action and conclude it's a bad action.
But it's more rational to look at the potential upside as well, evaluate the likelihood of each in the grand scheme of things, and then decide whether the potential upside will make up for the potential downside.
Quoted 2 lines
Let's say that for some reason I don't want the end users of my device to tinker with the software inside my device.
Ok, keep the *want* in mind. This is very important.
Quoted 1 line
Now I think you can agree to these things being positive:
Yes, even if I'd phrase them slightly differently.
Quoted 2 lines
The only downside is that the end user purchasing the device can't install modified versions of the software on it.
And therefore you severely limit the number of end users who might turn into contributors because of self interest in hacking the device to suit their needs.
Quoted 3 lines
Now let's try it in a GPLv3 universe. Since I can no longer create my device without having to allow the end user to install modified software on it
False assumption. You can create the device using GPLv3 software in it. So your acccounting of necessary downsides is only one of the possibilities. The other possibility would be to have the program in ROM, of course, which would come with a completely different set of downsides, but that would retain all of the "these things being positive" you mentioned above.
And, remember, since you merely don't *want* the end user of the device to tinker with the software, you have the option to do let them do that.
And, if you do, they may find in themselves reasons and incentives to change the software in the device, and the improvements are likely to get back to the community and thus back to you. Everybody wins.
This is the upside that you left out from your analysis, and from every other analysis that set out to "prove" that anti-tivoization is bad that I've seen so far.
It appears that people are so concerned about whatever little they might lose from requiring respect for users' freedoms that they don't even consider what they might win, and that they *would* win if at least some of the vendors were to make an choice more favorable to their users and the community.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · Alexandre Oliva
Quoted 4 lines · Jesper Juhl
On 17/06/07, Alexandre Oliva <[email protected]> wrote:
[snip]Quoted 1 line
Serious, what's so hard to understand about:
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
no tivoization => more users able to tinker their formerly-tivoized computers => more users make useful modifications => more contributions in kind
I have to disagree.
Your analysis stopped at the downside of prohibiting tivoization. You didn't analyze the potential upsides,
Maybe that's because I don't really see any up sides.
As I see it, if we prevent tivoization, then the most likely outcome will be that a very few number of vendors will switch to ROM based solutions or similar (everyone lose, both vendor and user), a few vendors that currently tivoize hardware may open up their hardware but I doubt that will be very many, and the vast majority of vendors will move to *BSD or proprietary software since they simply can't or won't open up their hardware.
So no, I don't think there are any upsides. We'll lose a huge number of developers, testers and users inside the business comunity and we'll lose a lot of exposure (like "hey, did you know TiVO actually runs Linux inside? Isn't that cool?)... Gaining a few hobyists at the expense of driving a huge number of businesses away from GPL'ed software does not look like an upside to me.
Quoted 3 lines
so you may indeed come to
different conclusions, and they may very well be wrong.
Just because I come to a different conclusion than you doesn't nessesarily make it wrong.
Quoted 3 lines
It's very human to look only at the potential downside of an action and conclude it's a bad action.
And you believe yourself to be immune to that - right?
Quoted 5 lines
Quoted 2 lines
Let's say that for some reason I don't want the end users of my device to tinker with the software inside my device.
Ok, keep the *want* in mind. This is very important.
No, it is not. When I wrote that I meant "don't want" as in "really don't want to since it'll destroy our business" or "really really don't want to since we would be breaking the law" etc.
Quoted 11 lines
Quoted 1 line
Now I think you can agree to these things being positive:
Yes, even if I'd phrase them slightly differently.
Quoted 2 lines
The only downside is that the end user purchasing the device can't install modified versions of the software on it.
And therefore you severely limit the number of end users who might turn into contributors because of self interest in hacking the device to suit their needs.
Most people don't care about hacking their devices, and of the few who do only a subset have the skill and only a subset of those will actually contribute anything back. This is a *small* set of people and gaining that small set at the expense of losing the large number of contributers from various companies doesn't make sense to me.
Quoted 6 lines
Quoted 3 lines
Now let's try it in a GPLv3 universe. Since I can no longer create my device without having to allow the end user to install modified software on it
False assumption. You can create the device using GPLv3 software in it.
Not as long as I want to prevent the user from tampering with it, no.
Quoted 6 lines
So your acccounting of necessary downsides is only one of the possibilities. The other possibility would be to have the program in ROM, of course, which would come with a completely different set of downsides, but that would retain all of the "these things being positive" you mentioned above.
But do you really expect a vendor to put a device on the market where they also lock themselves out of upgrading it and releasing new software for it? That's just rediculous.
Quoted 4 lines
And, remember, since you merely don't *want* the end user of the device to tinker with the software, you have the option to do let them do that.
See above.
Quoted 4 lines
And, if you do, they may find in themselves reasons and incentives to change the software in the device, and the improvements are likely to get back to the community and thus back to you. Everybody wins.
For a few select individuals that may be true. But for the majority of the population it won't mean a thing.
Quoted 4 lines
This is the upside that you left out from your analysis, and from every other analysis that set out to "prove" that anti-tivoization is bad that I've seen so far.
I'm sorry, but I don't think it holds water.
Quoted 5 lines
It appears that people are so concerned about whatever little they might lose from requiring respect for users' freedoms that they don't even consider what they might win, and that they *would* win if at least some of the vendors were to make an choice more favorable to their users and the community.
Contrary to you, I don't believe any significant number of companies will do that. It's simply better for business to just use other software in that case.
Signature · 3 lines
Jesper Juhl <[email protected]> Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html Plain text mails only, please http://www.expita.com/nomime.html
Quoted 5 lines · Jesper Juhl
Quoted 2 lines · Alexandre Oliva
Your analysis stopped at the downside of prohibiting tivoization. You didn't analyze the potential upsides,
Maybe that's because I don't really see any up sides.
You do:
Quoted 2 lines
a few vendors that currently tivoize hardware may open up their hardware but I doubt that will be very many
You just don't think they'd prevail over the downsides. *This* is an opinion I can respect, even if it's as much of a guestimate as mine. I'm sure both are highly influenced by personal opinions, wishful thinking and fears. This is all very human.
Quoted 5 lines
Quoted 2 lines
so you may indeed come to different conclusions, and they may very well be wrong.
Just because I come to a different conclusion than you doesn't nessesarily make it wrong.
Agreed. I didn't say they were. I said they could be. Can you prove they're right? Do you even have any supporting evidence to back your guestimates? Heck, you may even have more than I do.
I openly admit mine is mostly theoretical. I extrapolate the initial success of GNU+Linux on the PC environment, due in a large part to the ability for users to tinker with their computers, and expect it not to be so significantly different for other kinds of computers.
For sure you'll get a far lower *percentage* of hackers in consumer devices than on PCs, whose users used to be far more technically-inclined and thus more propense to become hackers when GNU+Linux started than these days.
But then I think of all of these computer users who helped make GNU+Linux what it is today, and other hackers that hadn't discovered this inclination before because they haven't had access to hackable computers. They could be tinkering with their DVRs, cell phones, wireless routers et al, and bringing the same kind of exciting community development to these kinds of computers.
I'm saddened that the major Linux developers are willing to trade all of this (which I openly admit may be just a figment of my imagination, or just a tip of an iceberg) for some professional contributions (good) and some additional exposure that won't do justice to their software (bad), because these users will miss a big part of the picture by not being able to tinker with the software in the environment where they use the software.
Quoted 4 lines
Quoted 2 lines
It's very human to look only at the potential downside of an action and conclude it's a bad action.
And you believe yourself to be immune to that - right?
Last I looked, I was still human. So no. I try to use logic to reason out such behaviors when I realize they might be in action. But, as the saying goes, logic is a tool we use to justify our intutions. Or, logical reasoning is a tool to make the wrong decisions with a greater amount of confidence ;-)
Quoted 3 lines
Quoted 1 line
You can create the device using GPLv3 software in it.
Not as long as I want to prevent the user from tampering with it, no.
<broken record>mumble ROM mumble</broken record>
Quoted 3 lines
But do you really expect a vendor to put a device on the market where they also lock themselves out of upgrading it and releasing new software for it?
Depends on how badly they want to use the GPLed software.
Don't you guys think Linux is so technically superior that some vendors might prefer to stick with it (should it move to GPLv3, or tivoization be found to be already forbidden by GPLv2 in a US court or elsewhere) even if this means going to ROM or respecting users' freedoms?
Or are Linux advantages so thin and fragile (if they exist at all) that you're just hoping nobody realizes there are better choices out there, and you're desperate for vendors not to realize this?
Quoted 2 lines
For a few select individuals that may be true. But for the majority of the population it won't mean a thing.
Agreed. You're thinking of percentages (fewer percent hackers among consumers of user products than among PC users). I'm thinking all hackers in PCs could become hackers of such devices as well, and then some.
Quoted 5 lines
Quoted 3 lines
This is the upside that you left out from your analysis, and from every other analysis that set out to "prove" that anti-tivoization is bad that I've seen so far.
I'm sorry, but I don't think it holds water.
Fair enough.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 6 lines · Alexandre Oliva
Quoted 3 lines
A third way of stating it is "software for software". No, the romans never said that, but I just did, to make it just more obvious that the whole point is that you are expected to answer IN KIND!
Yes. And this was precisely what meant when I wrote "quid-pro-quo" above.
Ok, so we're on the same page.
"software for software".
Quoted 6 lines
Quoted 4 lines
"You are either terminally stupid, or you're not allowing yourself
to see an obvious argument, because it destroys your world-view".The latter is very possible. It's a very human thing.
/me hands Linus a mirror
"I'm a damn handsome dude, ain't I?"
Quoted 1 line
Serious, what's so hard to understand about:
You're talking about something totally different.
Answer my argument:
- I think Tivoization is *good*.
- Your license stops something *good*.
Ergo:
- Your license is *bad*.
Quoted 3 lines
Wrong. It enables copyright holders to decide whether forgiveness is appropriate, rather than forcing them to forgive. Being forced to forgive deception is not tit-for-tat, and it's a losing strategy.
There is NOTHING TO FORGIVE!
Your whole idiotic argument misses the point:
What Tivo did is *good* in my opinion!
Can't you get that through your skull?
They gave the software back! Be happy! They *followed* the rules. They *followed* the tit-for-tat.
Linus
Quoted 5 lines · Linus Torvalds
Quoted 1 line · Alexandre Oliva
/me hands Linus a mirror
"I'm a damn handsome dude, ain't I?"
Heh. I beg to differ ;-)
Quoted 3 lines
Quoted 1 line
Serious, what's so hard to understand about:
You're talking about something totally different.
No, I'm not. You can say tivoization is *good* however much you like. This doesn't dispute in any way my claim that no tivoization would be *better*, that you'd get contributions from the people that, because of tivoization, don't feel compelled to develop and contribute, because they can't use the fruits of their efforts in the device where they would be most useful for them.
Quoted 3 lines
What Tivo did is *good* in my opinion!
Can't you get that through your skull?
No. I disagree. We can agree to disagree on that.
Quoted 2 lines
They gave the software back! Be happy! They *followed* the rules. They *followed* the tit-for-tat.
But they removed incentive for far more users to do so. So you get fewer contributions than you could without tivoization.
"Can't you get that through your skull?" :-)
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · Alexandre Oliva
Quoted 3 lines
What Tivo did is *good* in my opinion!
Can't you get that through your skull?
No. I disagree. We can agree to disagree on that.
Sorry, no we cannot. You seem to not accept that "in my opinion".
That's not somethign we can disagree on. My opinion is _mine_ to state. There's no room for disagreement. It's my opinion, and your "agreement" is not optional.
So when I say "I think what Tivo did is good", then you'd better just say, "ok, that's your opinion, and I respect you for it".
Otherwise you're a douche-bag and an idiot.
And once you realize that _I_ think that Tivo did is good, you have to accept that I think that the GPLv3 is the worse license in my opinion.
So stop blathering about anything else.
Just accept it. Just repeat after me: "Linus thinks that the GPLv3 is a bad license, and Linus is not confused".
Don't call me "confused". Don't bother talking about what _you_ think, or what the FSF thinks is the "spirit". Don't say that you cannot understand it.
Because if you cannot understand it, the only thing it shows is _your_ lack of understanding.
Linus
Quoted 8 lines · Linus Torvalds
Quoted 5 lines · Alexandre Oliva
Quoted 3 lines
What Tivo did is *good* in my opinion!
Can't you get that through your skull?
No. I disagree. We can agree to disagree on that.
Sorry, no we cannot. You seem to not accept that "in my opinion".
Sorry. I stand corrected. I didn't mean to disagree with the "in my opinion". I guess I was too distracted by the beauty of your kind words ;-) Sorry, it takes some effort to take focus away from that.
I agree that this is your opinion, and you're entitled to it, no matter how much I disagree with this opinion.
I disagree that what TiVo did is good. I think it is wrong on ethical and moral grounds, and I think it is bad for the user, for the community, and quite possibly even for TiVo itself.
We can agree to disagree as to our opinions, if you want.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 2 lines · Alexandre Oliva
We can agree to disagree as to our opinions, if you want.
That's all I ever asked for.
This whole thread started with me saying:
I see the smiley, but I hate it how the FSF thinks others are morons and cannot read or think for themselves.
Any time you disagree with the FSF, you "misunderstand" (insert condescending voice) the issue.
_Please_ don't continue that idiocy. Disagreement and thinking that the FSF is controlling and putting its fingers where they don't belong is _not_ misunderstanding. It's just not "blind and unquestioning obedience".
so all I asked for in the first place was that you stop claiming that I had "misunderstood" anything.
That's really all I've always asked for:
- I chose the GPLv2, and I understand it.
- you don't have to agree with my choice, but you *do* have to accept it if you want to work on Linux. Because it's the only license that Linux has ever been released under since early 1992.
So as long as you follow the GPLv2 (as a _legal_ license), I don't care if you like it or not. I don't care if you think you are a modern-day Napoleon, or if you are a demented squirrel. I don't care if you are an axe-murderer, or if you make sex toys with Linux. I don't care if your hardware is open or closed.
I care about one thing, and one thing only: I care that you respect my choice of license for the projects _I_ started. Nothing more.
And it doesn't matter one whit if *you* would have made a different choice. You are not me. You don't hold any power over me, and *your* choices are your own - not mine.
Choice of license is personal. Many people think that the BSD license is better than _any_ version of the GPL. Are they wrong? No, it's _their_ choice. Is it relevant for the kernel? No, their preference of license is simply irrelevant. They can choose to accept the license that the kernel is under, or go play somewhere else.
I think the GPLv2 is superior to the GPLv3. That is simply not something you can argue against. You can just say "ok, it's your choice". You can ask me *why*, and I've told you at length, but in the end, it doesn't matter.
And no, it's not because I'm "special", and I get to make all decisions. It's simply because I am _me_, and when it comes to my own opinions, I actually _do_ get to make all the decisions.
You can disagree, and choose to use the GPLv3. You just cannot do it for the *kernel*, because they kernel has always been under the GPLv2, and the GPLv3 is simply not compatible, and asks for things that the kernel license has never asked for.
But if you prefer the GPLv3, that's _your_ choice, and that choice can certainly guide you in the licensing of _your_ projects where _you_ are the copyright holder. And I will never complain.
Linus
Quoted 2 lines · Linus Torvalds
I care about one thing, and one thing only: I care that you respect my choice of license for the projects _I_ started. Nothing more.
I do. Really.
Once the issue about the spirit of the GPL is (hopefully) settled with all concerned about it, my job would have been done if it hadn't been for my having got interested in this other issue:
Quoted 4 lines
I think the GPLv2 is superior to the GPLv3. That is simply not something you can argue against. You can just say "ok, it's your choice". You can ask me *why*, and I've told you at length, but in the end, it doesn't matter.
Let me explain why I don't see that you've told me at length why you consider GPLv2 superior to GPLv3.
1. I asked you why GPLv2 is better, and you said it was because it promoted giving back in kind.
2. I asked you what you didn't like about GPLv3, and you said it was Tivoization.
3. Then I argued that, since Tivoization enables tivoizers to remove some motivation for potential developers (= their customers) to contribute, you trade the potential contributions of all those users for the contributions of tivoizers, apparently assuming that all tivoizers would simply move away from the community, taking their future contributions away from your community, rather than moving to a position in which you'd get not only the contributions from the company itself, but also from all their users.
This last piece of the theorem that proves that GPLv2 is more aligned with your stated goals than GPLv3 is the one that is missing, and so far you've dodged that portion entirely. That's the 'connecting the dots' that I mentioned earlier. You haven't even acknowledged its existence, going back to points 1. and 2. as if they were enough, as if 3. didn't show a contradiction between them.
Now, it may be that 3. is wrong, or that you think it is wrong. But you've never said so, or explained why you think so. You've simply disregarded that point entirely.
Do you understand now why I feel you haven't answered the 'why'?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 3 lines · Alexandre Oliva
1. I asked you why GPLv2 is better, and you said it was because it promoted giving back in kind.
Where I explained that "in kind" was about *software*.
Quoted 2 lines
2. I asked you what you didn't like about GPLv3, and you said it was Tivoization.
Right. The GPLv3 asks you to give back *money*.
That's like the Microsoft license agreements. I don't like them either.
Oh, and replace "money" with "access to hardware", to make that thing technically correct. But the point is, that's what I don't like about the GPLv3.
Quoted 2 lines
3. Then I argued that, since Tivoization enables tivoizers to remove some motivation for potential developers (= their customers)
That's simply not my *reason* for doing "tit-for-tat". My basic reason for "tit-for-tat" was not about "lots of potential developers", but simply because I think it's the right choice for me!
Can you not understand that? I simply DO NOT LIKE TO CONTROL PEOPLE!
I just want software back. I think it is *wrong* for me to ask for anything else. It's literally my personal "moral choice": I think the hardware manufacturers need to make their _own_ choices when it comes to _their_ designs.
I feel that I have the moral right to ask for modifications to the kernel (because I started it), but I *personally* am very unhappy about asking people to also give their hardware access. That's *their* choice.
Is that really so hard to understand? I ask you to respect _my_ choice wrt license for my software, but the same way I expect others to respect my choices, I also myself need to respect *their* choices.
So to me, it's the hardware manufacturers choice to to select the license for their hardware, exactly the same way it was *my* choice to select it for my software. I believe in basically *one* freedom: the freedom to make our own choices!
But if you actually want to discuss "number of developers" and their motications, I actually have another few arguments for you:
- I just personally think your math is bogus. I think more people think like I do, than people think like you and the FSF does.
But I don't even depend on that. Because:
- I think that *technical*quality* is more important than *quantity*.
And I think you have already proven a point: the GPLv3 seems to attract people who make the wrong *technical* decisions.
Put another way: I'll much rather attract one Al Viro to the project, than a hundred rabid FSF followers.
See? Because I think that one Al Viro will make *more* of a difference than a hundred people who think that their *religion* is more important than making the technically correct choice is!
With the GPLv2, you need to give your software modifications back, but the GPLv2 never *ever* makes any technical limitations on the end result.
In the GPLv3 world, we have already discussed in this thread how you can follow the GPLv3 by making the TECHNICALLY INFERIOR choice of using a ROM instead of using a flash device.
Quite frankly, I don't *want* to attract develpers that are not technically "up to snuff". And if you think that making the technically worse decisions is the "rigth decision", then hey, you're clearly not in the same technical quality range as I am, or Al Viro is.
Am I elitist? HELL YES! I think some people are simply *better* engineers than other people. I've met my share of outstanding engineers, and I've met average engineers.
I am firmly of the opinion that one of the signs of an outstanding engineer is making the right technical choices. The GPLv2 is ok with that. The GPLv3 is not. The GPLv3 makes *limits* what you can do from a technical angle, in a way that the GPLv2 does not.
The GPLv2 requires that you give source code out. But if you want to make your hardware in a way that it only runs signed versions, because of some reason like an FCC rule, or banking rule, or just because you damn well want, the GPLv2 doesn't stop that.
The GPLv3 doesn't stop it *either*, but the GPLv3 requries that you make the INFERIOR TECHNICAL CHOICE.
In other words: the GPLv3 is for people who care more about the opinions of the FSF than about the technology.
And why the hell should I trust people like that to make the right technical choices in *other* matters? They have already shown themselves to make bad technical choices.
Quoted 1 line
Do you understand now why I feel you haven't answered the 'why'?
Ok, so now I have. I have three *different* and independent answers for you:
a) I don't personally feel like I have the "moral authority" to require hardware designers to give access to their hardware to me.
I can tell them that I *like* open hardware more than closed hardware, but they designed the hardware, and as a result I think it's *their* choice.
In contrast, I _do_ think I have the moral authority to ask for modifications to the _software_ back. Because they didn't design it, they just improved on it.
b) I think you're simply wrong in your math. I think more people like the middle-ground and not-frothing-at-the-mouth spirit of "open source" over the religious dogma of "free software".
I think Linux has pretty much proved my point. Look at Hurd, then look at Linux. Am I *that* much better than the Hurd developers (yes, of course I am, but let's assume not). Or is it just that my approach of being more _pragmatic_ about things rather than talking about those "four freedoms" all the time was just much easier for people to accept?
c) Even if you're not wrong in the math, I've seen the kind of people who argue for it, and quite frankly, I think they are making bad technical decisions. You arguing for a ROM over a flash is an excellent example.
You seem to never even have given a second _thought_ to the fact that you actually advocated what is technically the inferior choice.
See? Those are three totally different reasons why I think the GPLv2 is the right license for me, and for the kernel.
Linus
Quoted 11 lines · Linus Torvalds
a) I don't personally feel like I have the "moral authority" to require hardware designers to give access to their hardware to me.
I can tell them that I *like* open hardware more than closed hardware, but they designed the hardware, and as a result I think it's *their* choice.
In contrast, I _do_ think I have the moral authority to ask for modifications to the _software_ back. Because they didn't design it, they just improved on it.
Btw, this may explain the differences between rms/FSF and me.
Richard Stallman is out to "improve the world". He thinks he has not just the moral authority and right, but the _obligation_ to tell others how to live their lives.
So rms feels he has to "spread his message", the same way those annoying people come to your door on Saturday in order to "save you". rms literally is saving you from a sin.
Me, not so much. I'm not out to "improve the world". I'm not saving you from any sin. I just have this project I started, and I think that if you improve on the project, you should let me use your improvements.
And I don't ask you use that license anywhere else. For your other projects, you should follow *their* licenses. I'm just asking you to honor my choice of license for those projects I started, and I designed, and where I conned other people into helping me under the same rules.
Linus
Quoted 4 lines · Linus Torvalds
Quite frankly, I don't *want* to attract develpers that are not technically "up to snuff". And if you think that making the technically worse decisions is the "rigth decision", then hey, you're clearly not in the same technical quality range as I am, or Al Viro is.
You are confusing being generally fucked in head (aka being a true believer) with being willing to go for technically worse decisions *when* *it* *comes* *to* *making* *them*. Which would be in gcc-related work in this case. I don't see any evidence of the latter...
Quoted 6 lines · Linus Torvalds
Quoted 3 lines · Alexandre Oliva
1. I asked you why GPLv2 is better, and you said it was because it promoted giving back in kind.
Where I explained that "in kind" was about *software*.
Yes, we'd already established that.
Quoted 9 lines
Quoted 2 lines
2. I asked you what you didn't like about GPLv3, and you said it was Tivoization.
Right. The GPLv3 asks you to give back *money*.
That's like the Microsoft license agreements. I don't like them either.
Oh, and replace "money" with "access to hardware", to make that thing technically correct.
technically, it asks you to pass on (!= give back) access to the software (not to the hardware that contains it).
Quoted 4 lines
Quoted 2 lines
3. Then I argued that, since Tivoization enables tivoizers to remove some motivation for potential developers (= their customers)
That's simply not my *reason* for doing "tit-for-tat".
No. The reason, again, is the portion you snipped out.
Could you try again?
Quoted 4 lines
I just want software back. I think it is *wrong* for me to ask for anything else. It's literally my personal "moral choice": I think the hardware manufacturers need to make their _own_ choices when it comes to _their_ designs.
It's comforting to see that you're not the pure-pragmatics-no-morals character that some people (yourself included) try to paint you as. Thank you for this. I can relate with that. I can easily respect that, as much as I think that (poor attempt at humor follows, no offense really intended) standing up for the freedoms of the poor hardware manufacturers against the evil software developers who want to control the ways they can use to control their customers serves the common good or even your own stated interests.
Quoted 9 lines
But if you actually want to discuss "number of developers" and their motications, I actually have another few arguments for you:
- I just personally think your math is bogus. I think more people think like I do, than people think like you and the FSF does.
But I don't even depend on that. Because:
- I think that *technical*quality* is more important than *quantity*.
This argument fails to make the point you're trying to make. I wrote:
you trade the potential contributions of all those users for the contributions of tivoizers, apparently assuming that all tivoizers would simply move away from the community, taking their future contributions away from your community, rather than moving to a position in which you'd get not only the contributions from the company itself, but also from all their users
and you say "oh, I don't care about quantity, I care about quality", as if this somehow related with the above.
Just do the math. Hypothetically, Linux goes GPLv3, without permission to tivoize. TiVo has to decide among:
- switching to another kernel, no further contributions from them
- sticking with old version, no further usable contributions from them
- switching to ROM, still the same contributions from TiVo
- no more tivoization, contributions from TiVo and users
So, you see, in no case do you get more contributors while at the same time losing TiVo's quality contributions.
The argument is not about quality vs quantity, it's about getting lots of additional contributions along with what's already in place, VS ending up without some quality contributions you get today.
Quoted 1 line
With the GPLv2, you need to give your software modifications back, but the
^^^^ BZZT!
Quoted 1 line
GPLv2 never *ever* makes any technical limitations on the end result.
Actually, just think of how many times you've heard the argument "I can't give you the source code for this driver/firmware/etc under the GPLv2 because the law says so."
Quoted 3 lines
In the GPLv3 world, we have already discussed in this thread how you can follow the GPLv3 by making the TECHNICALLY INFERIOR choice of using a ROM instead of using a flash device.
Yes. This is one option that doesn't bring any benefits to anyone. It maintains the status quo for users and the community, but it loses the ability for the vendor to upgrade, fix or otherwise control the users. Bad for the vendor.
As another option, the vendor can respect users' freedoms, and then everybody wins big. That's the option that anti-tivoization provides economic incentive for vendors to take. Sure, they may still prefer the alternative above, or stick with an older version (which has its costs), or move to different software (which also has its costs), but it's unreasonable to claim that I'm advocating for vendors to move to ROM.
I'm saying they have this option. I'm advocating for them to respect users' freedom. And if that's incompatible with their business model, well, so what? GPLv2 and Free Software in general are incompatible with a number of business models too, and who's complaining? Heck, even using slave work-forces was part of legitimate business models at some point in time.
Quoted 1 line
The GPLv2 requires that you give source code out.
^^^^^^^^ BZZT ;-)
Quoted 4 lines
But if you want to make your hardware in a way that it only runs signed versions, because of some reason like an FCC rule, or banking rule, or just because you damn well want, the GPLv2 doesn't stop that.
And then, the user is stopped from making appropriate technical decisions.
Quoted 2 lines
The GPLv3 doesn't stop it *either*, but the GPLv3 requries that you make the INFERIOR TECHNICAL CHOICE.
This is a lie (by which I don't mean it's malicious).
What it requires is the vendor to decide between making the inferior technical choice and respecting users' freedoms to make their own technical choices.
Quoted 3 lines
b) I think you're simply wrong in your math. I think more people like the middle-ground and not-frothing-at-the-mouth spirit of "open source" over the religious dogma of "free software".
It looks like the math you're talking about is in no way related with anything I've argued about. You seem to be thinking about the number of people who claim to be on the "free software" or "open source" sides, but I can't fathom in what way this is related with whether you get more or less contributions from users as a consequence of users' being permitted to tinker with the free software in their own devices.
Quoted 6 lines
I think Linux has pretty much proved my point. Look at Hurd, then look at Linux. Am I *that* much better than the Hurd developers (yes, of course I am, but let's assume not). Or is it just that my approach of being more _pragmatic_ about things rather than talking about those "four freedoms" all the time was just much easier for people to accept?
I could argue that Hurd took a different approach that proved to be far more difficult, and that the urgency for the development of a Free Software kernel by the GNU project disappeared with the relicensing of Linux under a Free Software license (thanks!). But I guess you'll just dismiss that on whatever reasons move you, and I don't really care about these particular historical issues to spend time discussing them.
Quoted 2 lines
See? Those are three totally different reasons why I think the GPLv2 is the right license for me, and for the kernel.
Ok, the only one that stands is the moral reason. That's a good one, but it contradicts the stated reasons as to why you prefer GPLv2 over GPLv3, stating it was a purely pragmatic decision, based on getting more software contributions back, based on an assumption, that so far lacks any material evidence, that permission for tivoization somehow gets you more of that than getting just as many contributions from the former-tivoizer, plus whatever any of their users decide to contribute.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 1 line · On Monday 18 June 2007 19:31:30 Alexandre Oliva wrote:
On Jun 18, 2007, Linus Torvalds <[email protected]> wrote:
<snip>
Quoted 9 lines
Quoted 3 lines
With the GPLv2, you need to give your software modifications back, but the
^^^^ BZZT!Quoted 1 line
GPLv2 never *ever* makes any technical limitations on the end result.
Actually, just think of how many times you've heard the argument "I can't give you the source code for this driver/firmware/etc under the GPLv2 because the law says so."
Sorry to tell you this, but anyone that makes a modification to GPLv2 covered code and distributes that modification is bound by the license. If a law makes following the license illegal, then they can't use any rights granted by the license. They are breaking the law by refusing to follow the license.
<snip>
Quoted 9 lines
Quoted 1 line
The GPLv2 requires that you give source code out.
^^^^^^^^ BZZT ;-)Quoted 4 lines
But if you want to make your hardware in a way that it only runs signed versions, because of some reason like an FCC rule, or banking rule, or just because you damn well want, the GPLv2 doesn't stop that.
And then, the user is stopped from making appropriate technical decisions.
You marked the "requires" as an error. Technically it is. Practically, however, it is rare for a modification to not fall under the "distribution" part of the license, making the "release the source" requirement active almost all the time.
<snip>
Quoted 10 lines
Quoted 3 lines
b) I think you're simply wrong in your math. I think more people like the middle-ground and not-frothing-at-the-mouth spirit of "open source" over the religious dogma of "free software".
It looks like the math you're talking about is in no way related with anything I've argued about. You seem to be thinking about the number of people who claim to be on the "free software" or "open source" sides, but I can't fathom in what way this is related with whether you get more or less contributions from users as a consequence of users' being permitted to tinker with the free software in their own devices.
"More Developers" (either "Free Software" or "Open Source") == "More Contributions"
That equation is very simple to understand - claiming its wrong is impossible. <snip>
Quoted 4 lines
Quoted 2 lines
See? Those are three totally different reasons why I think the GPLv2 is the right license for me, and for the kernel.
Ok, the only one that stands is the moral reason.
Apparently because you can't admit that a good reason *IS* a good reason when it conflicts with your belief that the FSF is correct. (The same as the "Science can't be right because it conflicts with the bible" I hear from all kinds of Xtians these days)
DRH
PS: I know I've said I'm done with this conversation, but this is like a bad habit. I just couldn't help myself.
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 9 lines · Daniel Hazelton
Quoted 5 lines · On Monday 18 June 2007 19:31:30 Alexandre Oliva wrote:
On Jun 18, 2007, Linus Torvalds <[email protected]> wrote:
Actually, just think of how many times you've heard the argument "I can't give you the source code for this driver/firmware/etc under the GPLv2 because the law says so."
Sorry to tell you this, but anyone that makes a modification to GPLv2 covered code and distributes that modification is bound by the license.
Of course I know that. I'm not the one making those arguments. And then, not all of those pieces of code are indeed moficiations of GPLv2-covered code, so your objection is off target.
Quoted 11 lines
Quoted 10 lines
Quoted 3 lines
b) I think you're simply wrong in your math. I think more people like the middle-ground and not-frothing-at-the-mouth spirit of "open source" over the religious dogma of "free software".
It looks like the math you're talking about is in no way related with anything I've argued about. You seem to be thinking about the number of people who claim to be on the "free software" or "open source" sides, but I can't fathom in what way this is related with whether you get more or less contributions from users as a consequence of users' being permitted to tinker with the free software in their own devices.
Quoted 4 lines
"More Developers" (either "Free Software" or "Open Source") == "More Contributions"
That equation is very simple to understand - claiming its wrong is impossible.
YES! Thank you! This is exactly the point I'm trying to make.
Now can you please explain this to Linus in terms that his brain won't dismiss as "coming from a fundamentalist"?
Quoted 2 lines
Apparently because you can't admit that a good reason *IS* a good reason when it conflicts with your belief that the FSF is correct.
No, seriously. Linus is disputing the equation above, dismissing my various attempts to show it to him, whenever it appears in teh context of tivoization, apparently because it doesn't match his moral belief that tivoization ought to be permitted on his moral grounds.
Quoted 2 lines
PS: I know I've said I'm done with this conversation, but this is like a bad habit. I just couldn't help myself.
You've helped me a lot while at that. Thanks!
I hope this helps others fundamentalist anti-fundamentalists :-) see reason too.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 8 lines · On Monday 18 June 2007 22:06:57 Alexandre Oliva wrote:
Quoted 7 lines · Daniel Hazelton
Quoted 5 lines · On Monday 18 June 2007 19:31:30 Alexandre Oliva wrote:
On Jun 18, 2007, Linus Torvalds <[email protected]> wrote:
Actually, just think of how many times you've heard the argument "I can't give you the source code for this driver/firmware/etc under the GPLv2 because the law says so."
Quoted 6 lines
Quoted 2 lines
Sorry to tell you this, but anyone that makes a modification to GPLv2 covered code and distributes that modification is bound by the license.
Of course I know that. I'm not the one making those arguments. And then, not all of those pieces of code are indeed moficiations of GPLv2-covered code, so your objection is off target.
I had a parsing error with your statement. My mind made the jump to "they are doing it with GPLv2 code" - meaculpa.
Quoted 14 lines
Quoted 14 lines
Quoted 10 lines
Quoted 3 lines
b) I think you're simply wrong in your math. I think more people like the middle-ground and not-frothing-at-the-mouth spirit of "open source" over the religious dogma of "free software".
It looks like the math you're talking about is in no way related with anything I've argued about. You seem to be thinking about the number of people who claim to be on the "free software" or "open source" sides, but I can't fathom in what way this is related with whether you get more or less contributions from users as a consequence of users' being permitted to tinker with the free software in their own devices.
"More Developers" (either "Free Software" or "Open Source") == "More Contributions"
Quoted 7 lines
Quoted 2 lines
That equation is very simple to understand - claiming its wrong is impossible.
YES! Thank you! This is exactly the point I'm trying to make.
Now can you please explain this to Linus in terms that his brain won't dismiss as "coming from a fundamentalist"?
No need. Linus already understands the equation, and also the secondary fact that most home users are not developers. However, companies like TiVO do employ developers. This is why the equation works.
Quoted 7 lines
Quoted 2 lines
Apparently because you can't admit that a good reason *IS* a good reason when it conflicts with your belief that the FSF is correct.
No, seriously. Linus is disputing the equation above, dismissing my various attempts to show it to him, whenever it appears in teh context of tivoization, apparently because it doesn't match his moral belief that tivoization ought to be permitted on his moral grounds.
Actually you are in error here. You are saying "More home users == More Developers" when the ratio of home users to developers isn't all that high. (small set of facts: "Hacker" == "Developer" (in most cases, where the term, as defined in the Jargon File, can actually be applied), "Home User" * 0.10 (ie: 10%) == "Developer" (approximately, and the correlation may be lower). "TiVO" == "Developers" (note the plural - they do employ more than one person for development))
So "TiVO", even though they are walking all over the freedoms you love, means more *guaranteed* developers than the potential pool from the users of their boxes. (the pool of potential developers among the millions of TiVO users is actually miniscule, despite the size of the sample)
However, you do make a good argument. But when you look at the statistics[1] they don't hold water.
Quoted 7 lines
Quoted 2 lines
PS: I know I've said I'm done with this conversation, but this is like a bad habit. I just couldn't help myself.
You've helped me a lot while at that. Thanks!
I hope this helps others fundamentalist anti-fundamentalists :-) see reason too.
I love that phrase!
But seriously, all I did was stop trying to give fully reasoned counters (complete with examples) and state the simple truth.
DRH
[1] "There are three types of lies - lies, damn lies and statistics" - Attribution uncertain (Benjamin Disraeli, Mark Twain and Alfred Marshal are all said to have issued this famous quotation)
PS: I've beaten the addiction! This post was to clarify some things that were either misunderstood or stood a chance of being twisted to mean something other than what I intended. (not that anyone did the latter)
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 7 lines · Daniel Hazelton
Actually you are in error here. You are saying "More home users == More Developers" when the ratio of home users to developers isn't all that high. (small set of facts: "Hacker" == "Developer" (in most cases, where the term, as defined in the Jargon File, can actually be applied), "Home User" * 0.10 (ie: 10%) == "Developer" (approximately, and the correlation may be lower). "TiVO" == "Developers" (note the plural - they do employ more than one person for development))
As I wrote in another e-mail, why makes this proportion different for the other conditions determined by the GPL?
I.e., what is it that makes this particular condition so allegedly harmful for bringing in more developers and contributions, when compared with the requirements on passing on source code, licensing necessary patents, not suing other users over patent infringement in the software, not invoking anti-circumvention laws, not entering discriminatory agreements?
Quoted 7 lines
So "TiVO", even though they are walking all over the freedoms you love, means more *guaranteed* developers than the potential pool from the users of their boxes. (the pool of potential developers among the millions of TiVO users is actually miniscule, despite the size of the sample)
However, you do make a good argument. But when you look at the statistics[1] they don't hold water.
Err... I have no idea of the actual user base of TiVo, but if it's really in the millions, and your 10% figure above is right, this makes for hundreds of thousands of hackers that could be scratching their itches and improving Linux on TiVo boxes.
How many thousand employees does TiVo have working on Linux?
(I realize a full-time employee is a lot more than a Joe Random Hacker, but still, I'm keeping a ratio of 100:1 to make up for that)
Quoted 1 line
PS: I've beaten the addiction!
Good for you!
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 8 lines · Alexandre Oliva
Quoted 3 lines
"More Developers" (either "Free Software" or "Open Source") == "More Contributions"
No, seriously. Linus is disputing the equation above, dismissing my various attempts to show it to him, whenever it appears in teh context of tivoization, apparently because it doesn't match his moral belief that tivoization ought to be permitted on his moral grounds.
No. Linus is not AT ALL disputing the equation above.
But you are too f*cking stupid to admit that I *accepted* the
- "More developers" == "More contributions" == good
equation, but I was claiming that your *other* part was totally broken.
You try to claim that the GPLv3 causes "More developers", and that, my idiotic penpal, is just crazy talk that you made up.
But since you cannot follow a logical argument, and cannot make one up on your own, you instead make up some *other* argument, and try (like above) to try to say that I made that claim.
The GPLv2 is the one that allows more developers.
The GPLv2 is the one that is acceptable to more people.
Face it, the "open source" crowd is the *bigger* crowd. The FSF crowd is vocal and opinionated, but it's largely made up of people who _talk_ more than they actually _code_.
Hot air doesn't make the world go round. Real code does.
Look at the kernel developers who claim that the GPLv2 is better. Not just me. Then look at the people who actually GET THINGS DONE.
There's a big overlap there.
Now, look at the people who try to sell the GPLv3 as the best thing since sliced bread. How many of those are people who actually get things *done*?
I haven't really seen a single one. Last I did the statistic, I asked the top ~25-30 kernel developers about their opinion. NOT A SINGLE ONE preferred the GPLv3.
So I have actual *numbers* on my side. What do you have, except for a history of not actually understanding my arguments?
Linus
Quoted 2 lines · Linus Torvalds
So I have actual *numbers* on my side. What do you have, except for a history of not actually understanding my arguments?
Why do I suddenly have an image of Palin as Ximenez doing the answer?
Quoted 3 lines · Linus Torvalds
The GPLv2 is the one that allows more developers.
The GPLv2 is the one that is acceptable to more people.
Based on my understanding that the anti-tivoization provisions are *the* objectionable issue about GPLv3 for those of you who dislike GPLv3, this is circular reasoning:
anti-tivoization is bad
=> we reject licenses with it
=> there are fewer developers willing to develop with such licenses => anti-tivoization is bad
Quoted 1 line
Face it, the "open source" crowd is the *bigger* crowd.
I really don't know about that. I can believe it may be so in LKML.
Quoted 3 lines
I haven't really seen a single one. Last I did the statistic, I asked the top ~25-30 kernel developers about their opinion. NOT A SINGLE ONE preferred the GPLv3.
Wow, that's a really big sample among all Free Software and Open Source developers out there. And not even a little bit biased at that.
Quoted 2 lines
So I have actual *numbers* on my side. What do you have, except for a history of not actually understanding my arguments?
Which is worse, not understanding or repeatedly snipping out and addressing unrelated points?
Let's please try again.
I'll try to keep it simple, since you can't seem to be able to grasp the entire argument, and keep disregarding essential parts, disputing unrelated points and jumping to the conclusions that you've disputed the point I was trying to make.
I'll present it in parts, as an attempt to stop you from making this mistake, that I'm sure is not intentional.
The first part is in this e-mail.
Dispute this:
non-tivoized hardware => users can scratch their itches => more contributions from these users
tivoized hardware => users can't scratch their itches => fewer contributions from these users
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 5 lines · On Tuesday 19 June 2007 01:51:19 Alexandre Oliva wrote:
Quoted 3 lines · Linus Torvalds
The GPLv2 is the one that allows more developers.
The GPLv2 is the one that is acceptable to more people.
Quoted 8 lines
Based on my understanding that the anti-tivoization provisions are *the* objectionable issue about GPLv3 for those of you who dislike GPLv3, this is circular reasoning:
anti-tivoization is bad
=> we reject licenses with it
=> there are fewer developers willing to develop with such licenses => anti-tivoization is bad
The logic is close to:
=> License forbids X
=> developer has requirement for X in license, can't add to project => License forbidding X is bad
When it comes to TiVO the reason the developers "required" the "tivoization" was because the company itself demanded it. The reason: providers of the content their device works with demanded it.
Quoted 3 lines
Quoted 1 line
Face it, the "open source" crowd is the *bigger* crowd.
I really don't know about that. I can believe it may be so in LKML.
Actually, it is. Every "Free Software" developer that I personally know could care less about the FSF's motives - until they impact them. Since they don't care what the FSF does, publishes, etc... they cannot be termed "Free Software" developers (using the definition of the term "Free Software" provided by the FSF). In fact, almost all of them will state either: "I work on Open Source software" or "I develop FOSS stuff".
Quoted 7 lines
Quoted 3 lines
I haven't really seen a single one. Last I did the statistic, I asked the top ~25-30 kernel developers about their opinion. NOT A SINGLE ONE preferred the GPLv3.
Wow, that's a really big sample among all Free Software and Open Source developers out there. And not even a little bit biased at that.
Yes, the sample could be considered "biased" - jst as a sample taken among the GCC developers could be considered "biased" towards the other end of the spectrum.
Quoted 5 lines
Quoted 2 lines
So I have actual *numbers* on my side. What do you have, except for a history of not actually understanding my arguments?
Which is worse, not understanding or repeatedly snipping out and addressing unrelated points?
It's time to quote a very ancient source: "Don't point out the speck in your neighbors eye when you cannot see the log in your own"
In other words - you've done the same and more.
Quoted 14 lines
Let's please try again.
I'll try to keep it simple, since you can't seem to be able to grasp the entire argument, and keep disregarding essential parts, disputing unrelated points and jumping to the conclusions that you've disputed the point I was trying to make.
I'll present it in parts, as an attempt to stop you from making this mistake, that I'm sure is not intentional.
The first part is in this e-mail.
Quoted 7 lines
Dispute this:
non-tivoized hardware => users can scratch their itches => more contributions from these users
tivoized hardware => users can't scratch their itches => fewer contributions from these users
Linus doesn't have to. Statistically the number of people that will even think of modifying the code running on a "tivoized" device is minute - at most 5% of the users of such a device. Of those people the ones with the skill to actually do the work is an even smaller number - figure 2.5 to 3% of them. Of those with the skill, probably about 10% of them are actually *good* enough at it for their changes to be useful. Of that number, figure that only 25%, at most, will contribute the changes back.
Apply that to a sample case:
"tivoized" device total users: 1,000,000
people that think about modifying: 50,000 (5%)
people with skill: 1500 (3%)
people who are good enough for the changes to be useful: 150 (10%) those who will contribute them back: 38 (25%)
Now a normal companies software department is anywhere from 10 to 50 people. Large companies can employ more - IBM employs hundreds.
What you are arguing is that people should abandon a firm set of developers that is proven to be large for the potential at adding, at most, 38 developers per million users of the device. If that number was more than 1000 per device I'd agree. The numbers don't support your argument.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 15 lines · Daniel Hazelton
Quoted 5 lines · On Tuesday 19 June 2007 01:51:19 Alexandre Oliva wrote:
Quoted 3 lines · Linus Torvalds
The GPLv2 is the one that allows more developers.
The GPLv2 is the one that is acceptable to more people.
Quoted 8 lines
Based on my understanding that the anti-tivoization provisions are *the* objectionable issue about GPLv3 for those of you who dislike GPLv3, this is circular reasoning:
anti-tivoization is bad
=> we reject licenses with it
=> there are fewer developers willing to develop with such licenses => anti-tivoization is bad
Quoted 5 lines
The logic is close to:
=> License forbids X
=> developer has requirement for X in license, can't add to project => License forbidding X is bad
I'm not sure it was clear that '=>' was meant as logical implication. Read it as "therefore".
It's actually funny that what your inference sequence (in spite of the missing initial operand) rings so true about my impressions about some of the reactions I'm getting here.
GPLv3 forbids tivoization, therefore developer has requirement for tivoization in the license, therefore GPLv3 forbidding tivoization is bad.
:-)
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
I haven't really seen a single one. Last I did the statistic, I asked the top ~25-30 kernel developers about their opinion. NOT A SINGLE ONE preferred the GPLv3.
Wow, that's a really big sample among all Free Software and Open Source developers out there. And not even a little bit biased at that.
Sorry that I missed the <irony> markers.
Quoted 3 lines
Yes, the sample could be considered "biased" - jst as a sample taken among the GCC developers could be considered "biased" towards the other end of the spectrum.
FWIW, I haven't taken such a sample, because I know my network of contacts would likely make it statistically useless. I'd not try to make an argument based on that.
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
So I have actual *numbers* on my side. What do you have, except for a history of not actually understanding my arguments?
Which is worse, not understanding or repeatedly snipping out and addressing unrelated points?
In other words - you've done the same and more.
I've honestly tried not to. I believe Linus has, too. Many of us have talked past each other, a lot.
That was actually the point behind breaking up the argument in small pieces.
If Linus hadn't got the whole argument, a number of times, before, this might be described as dishonest, but since he did, and he can refer back to those messages, he can know where I'm going.
Quoted 9 lines
Quoted 7 lines
Dispute this:
non-tivoized hardware => users can scratch their itches => more contributions from these users
tivoized hardware => users can't scratch their itches => fewer contributions from these users
Linus doesn't have to.
Of course he doesn't. But he will. Because he's always right, and he wants to show that. That this is a bait and he knows it won't stop him. He knows there's no hook, because he knows where I'm going with the argument. But it's going to be interesting to watch.
Quoted 2 lines
Statistically the number of people that will even think of modifying the code running on a "tivoized" device is minute
Wait a minute, these figures you made up are for the tivoized hardware (no changes allowed to the GPLed software in it), or for the non-tivoized hardware (changes allowed to the GPLed software in it)?
Quoted 1 line
those who will contribute them back: 38 (25%)
Regardless of what you meant, this is 38 developers *on top* of however many the company pays to work on that, unless you're jumping the gun and spoiling the multi-part argument.
Quoted 1 line
What you are arguing is that people should abandon
I'm not arguing any such thing. Where's any such argument above?
At this point, I'm only comparing a tivoized device with a non-tivoized device. Nothing but it.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 29 lines · On Tuesday 19 June 2007 02:44:32 Alexandre Oliva wrote:
Quoted 15 lines · Daniel Hazelton
Quoted 5 lines · On Tuesday 19 June 2007 01:51:19 Alexandre Oliva wrote:
Quoted 3 lines · Linus Torvalds
The GPLv2 is the one that allows more developers.
The GPLv2 is the one that is acceptable to more people.
Quoted 8 lines
Based on my understanding that the anti-tivoization provisions are *the* objectionable issue about GPLv3 for those of you who dislike GPLv3, this is circular reasoning:
anti-tivoization is bad
=> we reject licenses with it
=> there are fewer developers willing to develop with such licenses => anti-tivoization is badQuoted 5 lines
The logic is close to:
=> License forbids X
=> developer has requirement for X in license, can't add to project => License forbidding X is badI'm not sure it was clear that '=>' was meant as logical implication. Read it as "therefore".
It's actually funny that what your inference sequence (in spite of the missing initial operand) rings so true about my impressions about some of the reactions I'm getting here.
Quoted 5 lines
GPLv3 forbids tivoization, therefore developer has requirement for tivoization in the license, therefore GPLv3 forbidding tivoization is bad.
:-)
However, my argument is straight logic, nothing "circular" about it. :) Replacing "X" in my logic path above with "tivoization" and "license" with "GPLv3", as you've done, does produce a valid chain of logic.
Quoted 10 lines
Quoted 7 lines
Quoted 7 lines
Quoted 3 lines
I haven't really seen a single one. Last I did the statistic, I asked the top ~25-30 kernel developers about their opinion. NOT A SINGLE ONE preferred the GPLv3.
Wow, that's a really big sample among all Free Software and Open Source developers out there. And not even a little bit biased at that.
Sorry that I missed the <irony> markers.
Quoted 7 lines
Quoted 3 lines
Yes, the sample could be considered "biased" - jst as a sample taken among the GCC developers could be considered "biased" towards the other end of the spectrum.
FWIW, I haven't taken such a sample, because I know my network of contacts would likely make it statistically useless. I'd not try to make an argument based on that.
FWIW the Linux Kernel shouldn't be as homogeneous a population as it is. I'd expect it with an FSF run project, because they require copyright assignment in order to participate, but with a project like Linux, where everyone maintains the copyright to their contributions, should be a hell of a lot less homogeneous than Linus' numbers make it seem.
<snip>
Quoted 6 lines
Quoted 2 lines
Statistically the number of people that will even think of modifying the code running on a "tivoized" device is minute
Wait a minute, these figures you made up are for the tivoized hardware (no changes allowed to the GPLed software in it), or for the non-tivoized hardware (changes allowed to the GPLed software in it)?
Actually, any generic "TiVO"-like hardware - whether it is tivoized or not. Admittedly the numbers are significantly different for PC's (and other types of general purpose computing devices).
Quoted 5 lines
Quoted 1 line
those who will contribute them back: 38 (25%)
Regardless of what you meant, this is 38 developers *on top* of however many the company pays to work on that, unless you're jumping the gun and spoiling the multi-part argument.
38ppm is a fairly small amount, regardless.
Quoted 6 lines
Quoted 1 line
What you are arguing is that people should abandon
I'm not arguing any such thing. Where's any such argument above?
At this point, I'm only comparing a tivoized device with a non-tivoized device. Nothing but it.
You've been making the argument the entire time you've been arguing that the "anti-tivoization" language in the GPLv3 is necessary. I think I'd rather see a guaranteed increase of developers - even if it is only 10 - rather than hoping that the potential pool of 38 actually follows through. Wouldn't you?
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 8 lines · Daniel Hazelton
Quoted 3 lines · On Tuesday 19 June 2007 02:44:32 Alexandre Oliva wrote:
GPLv3 forbids tivoization, therefore developer has requirement for tivoization in the license, therefore GPLv3 forbidding tivoization is bad.
However, my argument is straight logic, nothing "circular" about it. :) Replacing "X" in my logic path above with "tivoization" and "license" with "GPLv3", as you've done, does produce a valid chain of logic.
Yes. Isn't it funny though that tivoization became necessary as a consequence of GPLv3 forbidding it?
Quoted 2 lines
FWIW the Linux Kernel shouldn't be as homogeneous a population as it is.
Nah. Communities tend to form around similar values. Linus started the community.
Quoted 5 lines
Quoted 3 lines
Wait a minute, these figures you made up are for the tivoized hardware (no changes allowed to the GPLed software in it), or for the non-tivoized hardware (changes allowed to the GPLed software in it)?
Actually, any generic "TiVO"-like hardware - whether it is tivoized or not.
So your claim is that a user's possibility to scratch her own itches makes no difference whatsoever as to their amount of contributions she is likely to make?
Am I the only one who thinks this is utter nonsense?
Quoted 7 lines
Quoted 5 lines
Quoted 1 line
those who will contribute them back: 38 (25%)
Regardless of what you meant, this is 38 developers *on top* of however many the company pays to work on that, unless you're jumping the gun and spoiling the multi-part argument.
38ppm is a fairly small amount, regardless.
Yes. And your estimates are way too low too, FWIW. Any reason why you changed your mind as to the 10% before?
Quoted 9 lines
Quoted 6 lines
Quoted 1 line
What you are arguing is that people should abandon
I'm not arguing any such thing. Where's any such argument above?
At this point, I'm only comparing a tivoized device with a non-tivoized device. Nothing but it.
You've been making the argument the entire time you've been arguing that the "anti-tivoization" language in the GPLv3 is necessary.
And then I decided that, since the argument wasn't getting through, I had to break it into pieces.
The piece I've presented so far has no abandonment whatsoever. It's a comparison between two different situations, to evaluate which of them brings more contributions from users, regardless of the contributions from the vendor, that are assumed to be the same, since there's no material difference as far as the vendor is concerned (as in, vendor has no reason to tivoize)
So your arguments bear zero relationship with the piece I've proposed. Can you see that?
Quoted 3 lines
I think I'd rather see a guaranteed increase of developers - even if it is only 10 - rather than hoping that the potential pool of 38 actually follows through. Wouldn't you?
Yes. How does this relate with the piece of the argument I've proposed so far, or the whole argument I've posted before?
Answer: It doesn't. At all. You're just showing you didn't understand the argument. Which shows why I have to explain it piece by piece. Which suggests you shouldn't try to jump to conclusions.
Once again, now with clearer starting conditions (not intended to match TiVo in any way, BTW; don't get into that distraction)
Vendor doesn't care about tivoizing, their business works the same either way.
Vendor's employees will contribute the same, one way or another, so their contributions are out of the equation.
Users get source code in either case, and they can modify it and share it. They're in no way stopped from becoming part of the community.
Given these conditions:
In a tivoized device, users will be unable to scratch their itches. This doesn't stop them from contributing to the project, but they may lack self-interest motivation to contribute, because they won't be able to use their modifications in the device they own.
In a non-tivoized device, users can scratch their itches. They can contribute just as much as they would in a tivoized device, but since they can use the changes they make to make their own devices work better for them, this works as a motivator for them to make changes, and perhaps to contribute them. Therefore, they will tend to contribute more.
Can you point out any flaw in this reasoning, or can we admit it as true?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 6 lines · On Tuesday 19 June 2007 04:04:52 Alexandre Oliva wrote:
Quoted 5 lines · Daniel Hazelton
Quoted 3 lines · On Tuesday 19 June 2007 02:44:32 Alexandre Oliva wrote:
GPLv3 forbids tivoization, therefore developer has requirement for tivoization in the license, therefore GPLv3 forbidding tivoization is bad.
Quoted 6 lines
Quoted 3 lines
However, my argument is straight logic, nothing "circular" about it. :) Replacing "X" in my logic path above with "tivoization" and "license" with "GPLv3", as you've done, does produce a valid chain of logic.
Yes. Isn't it funny though that tivoization became necessary as a consequence of GPLv3 forbidding it?
-ELOGIC
It didn't become necessary as a result of the GPLv3 forbidding it. As I pointed out in text that was cut to keep the post short, there could be any number of reasons why "tivoization" is needed by the manufacturer. Other people have also pointed that out. This whole bit was to point out that you were inferring circular logic where none existed.
<snip>
Quoted 10 lines
Quoted 6 lines
Quoted 3 lines
Wait a minute, these figures you made up are for the tivoized hardware (no changes allowed to the GPLed software in it), or for the non-tivoized hardware (changes allowed to the GPLed software in it)?
Actually, any generic "TiVO"-like hardware - whether it is tivoized or not.
So your claim is that a user's possibility to scratch her own itches makes no difference whatsoever as to their amount of contributions she is likely to make?
Exactly.
Quoted 4 lines
Am I the only one who thinks this is utter nonsense?
Quoted 2 lines
Quoted 2 lines
Quoted 1 line
those who will contribute them back: 38 (25%)
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines
Regardless of what you meant, this is 38 developers *on top* of however many the company pays to work on that, unless you're jumping the gun and spoiling the multi-part argument.
38ppm is a fairly small amount, regardless.
Yes. And your estimates are way too low too, FWIW. Any reason why you changed your mind as to the 10% before?
That 10% was, IIRC, a reference to the potential number of "Hackers" that would own a TiVO. On thinking about it I realized that the number of hackers owning a TiVO would be vanishingly small because of "tivoization". So in this new set of numbers I dropped it entirely.
... crap I am tempted to respond to nastily has been cut ...
Quoted 10 lines
Quoted 3 lines
I think I'd rather see a guaranteed increase of developers - even if it is only 10 - rather than hoping that the potential pool of 38 actually follows through. Wouldn't you?
Yes. How does this relate with the piece of the argument I've proposed so far, or the whole argument I've posted before?
Answer: It doesn't. At all. You're just showing you didn't understand the argument. Which shows why I have to explain it piece by piece. Which suggests you shouldn't try to jump to conclusions.
Wrong. Nobody here needs a "piece by piece" explanation. So, in the belief that you were intelligent enough to understand that, I was providing proof that refutes your argument entirely.
With a situation as complex as what exists you can't split the argument into two and claim that, since "Argument A" is true in the "split" argument that it is true when the argument isn't split. This holds true for almost all real-world situations.
Now, I am not enjoying the discussion anymore. I've asked once before - remove me from the CC list.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 7 lines · Daniel Hazelton
Quoted 6 lines · On Tuesday 19 June 2007 04:04:52 Alexandre Oliva wrote:
Quoted 5 lines · Daniel Hazelton
Quoted 3 lines · On Tuesday 19 June 2007 02:44:32 Alexandre Oliva wrote:
GPLv3 forbids tivoization, therefore developer has requirement for tivoization in the license, therefore GPLv3 forbidding tivoization is bad.
Quoted 8 lines
Quoted 6 lines
Quoted 3 lines
However, my argument is straight logic, nothing "circular" about it. :) Replacing "X" in my logic path above with "tivoization" and "license" with "GPLv3", as you've done, does produce a valid chain of logic.
Yes. Isn't it funny though that tivoization became necessary as a consequence of GPLv3 forbidding it?
-ELOGIC
I see. Try 'modprobe logic', it worked for me years ago ;-) :-D
Quoted 1 line
It didn't become necessary as a result of the GPLv3 forbidding it.
Which is why I said it was funny, because your inference chain stated *exactly* (with an implied "for the developers") that it did.
Do you understand what an inference chain is? A => B, as in A implies B, which can also be read as A therefore B if A is known to hold.
Quoted 2 lines
there could be any number of reasons why "tivoization" is needed by the manufacturer.
This claim is false.
Tivoization is when hardware manufacturer takes copyleft software and blocks updates by the user of the hardware.
No single law so far has shown an example that even resembled to mandate copyleft software, and no contract could possibly establish a condition like this.
Therefore, this claim is false.
Quoted 2 lines
This whole bit was to point out that you were inferring circular logic where none existed.
There *is* circular logic is in place.
The initial premise of this fallacy is that anti-tivoization is bad for the project.
This is used to conclude that licenses with such provisions should be rejected.
This is then used to conclude that there are fewer developers who would develop under such licenses.
Which is then used to conclude that anti-tivozation is bad for the project.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 13 lines · On Tuesday 19 June 2007 13:06:17 Alexandre Oliva wrote:
Quoted 12 lines · Daniel Hazelton
Quoted 11 lines · On Tuesday 19 June 2007 04:04:52 Alexandre Oliva wrote:
Quoted 9 lines · Daniel Hazelton
Quoted 3 lines · On Tuesday 19 June 2007 02:44:32 Alexandre Oliva wrote:
GPLv3 forbids tivoization, therefore developer has requirement for tivoization in the license, therefore GPLv3 forbidding tivoization is bad.
However, my argument is straight logic, nothing "circular" about it. :) Replacing "X" in my logic path above with "tivoization" and "license" with "GPLv3", as you've done, does produce a valid chain of logic.
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines
Yes. Isn't it funny though that tivoization became necessary as a consequence of GPLv3 forbidding it?
-ELOGICI see. Try 'modprobe logic', it worked for me years ago ;-) :-D
Try using real logic rather than the logic of your religion.
Quoted 7 lines
Quoted 1 line
It didn't become necessary as a result of the GPLv3 forbidding it.
Which is why I said it was funny, because your inference chain stated *exactly* (with an implied "for the developers") that it did.
Do you understand what an inference chain is? A => B, as in A implies B, which can also be read as A therefore B if A is known to hold.
Okay, since you want it in a specific language rather than as a set of bullet points (which is what I used the => for)
Company X has requirement for restriction Y
=> License on product Z disallows restriction Y
=> Product Z loses Company X and the exposure use in their product gives => License on product Z is bad for the product
Understandable now?
Quoted 7 lines
Quoted 2 lines
there could be any number of reasons why "tivoization" is needed by the manufacturer.
This claim is false.
Tivoization is when hardware manufacturer takes copyleft software and blocks updates by the user of the hardware.
By that definition you are correct.
Quoted 3 lines
No single law so far has shown an example that even resembled to mandate copyleft software, and no contract could possibly establish a condition like this.
No argument here. What I was stating is that are legal (and other reasons) why a company might have to lock down their software in a process similar to "Tivoization".
Quoted 1 line
Therefore, this claim is false.
Only when you define a term as specifically as you have done for "Tivoization". I should, perhaps, have used a different term - it would then have been patently true. Though, at that point, you would likely have argued that it wasn't "tivoization"
Quoted 8 lines
Quoted 2 lines
This whole bit was to point out that you were inferring circular logic where none existed.
There *is* circular logic is in place.
The initial premise of this fallacy is that anti-tivoization is bad for the project.
Quoted 8 lines
This is used to conclude that licenses with such provisions should be rejected.
This is then used to conclude that there are fewer developers who would develop under such licenses.
Which is then used to conclude that anti-tivozation is bad for the project.
Your view of the logic is, in this case, flawed. It's more along the lines of what I've detailed above. Now, please, go away. You aren't doing your "religion" any good. In fact, you are damaging it - repeatedly.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Quoted 6 lines · Daniel Hazelton
Company X has requirement for restriction Y
=> License on product Z disallows restriction Y
=> Product Z loses Company X and the exposure use in their product gives => License on product Z is bad for the productUnderstandable now?
Well, considering that I've made this claim myself as part of my complete argument in a number of times I've presented it, yes, this is understandable and correct. This is indeed one of the cases.
That said, very few companies have a scrict *requirement* for keeping the ability to modify the software on the customer's computer while denying this ability to the customer.
So this case you're discussing is the least common case. It could nearly be dismissed, rather than being the dominating topic in the discussion, as it's been so far.
Quoted 3 lines
What I was stating is that are legal (and other reasons) why a company might have to lock down their software in a process similar to "Tivoization".
Ok. Most of these can be addressed (with inconvenience) with ROM. Others are business reasons, and for these, the increased cost and inconvenience of the alternatives may shift them to an unlock situation.
Quoted 4 lines
Quoted 1 line
Therefore, this claim is false.
Only when you define a term as specifically as you have done for "Tivoization".
It's not my definition. This was from Wikipedia.
Quoted 2 lines
I should, perhaps, have used a different term - it would then have been patently true.
Depends on what the different term was.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 6 lines · Daniel Hazelton
Quoted 3 lines · On Tuesday 19 June 2007 04:04:52 Alexandre Oliva wrote:
So your claim is that a user's possibility to scratch her own itches makes no difference whatsoever as to their amount of contributions she is likely to make?
Exactly.
Hmm, interesting... Are you doing this just to make it more interesting, or did you never take Open Source 101? ;-) Or economy, game theory, ecology, politics, for that matter?
Do you realize how many contradictions arise from this claim?
A few examples for you. Let me know if you don't understand why these contradict this claim, I'll be happy to explain it to you.
How many newbies have patches accepted that they didn't test?
At which point in becoming a Linux developer does one leave behind this altruistic attitude and becomes moved by self interest only?
How many people you know got to know Linux in their TiVos, took the Linux sources that TiVo distributes and built it for their own PC?
Quoted 6 lines
Quoted 2 lines
Yes. And your estimates are way too low too, FWIW. Any reason why you changed your mind as to the 10% before?
That 10% was, IIRC, a reference to the potential number of "Hackers" that would own a TiVO. On thinking about it I realized that the number of hackers owning a TiVO would be vanishingly small because of "tivoization".
Aah, ok.
So, you lowered the estimate for the case of a tivoized device, but then claimed it was the same for an otherwise-identical non-tivoized device.
Seriously, try 'modprobe logic', you'll like it ;-)
Quoted 1 line
Wrong. Nobody here needs a "piece by piece" explanation.
Then why do you keep making claims that are not related with either the part of the argument I'm posing or the full argument I've already presented?
Quoted 2 lines
So, in the belief that you were intelligent enough to understand that, I was providing proof that refutes your argument entirely.
The only proof you provided was that you didn't understand the argument.
Quoted 4 lines
With a situation as complex as what exists you can't split the argument into two and claim that, since "Argument A" is true in the "split" argument that it is true when the argument isn't split. This holds true for almost all real-world situations.
Which proves you don't understand how logical proofs work.
Let me explain it to you.
First, you need to establish initial premises. Whether or not they resemble any similarity with anything else you're thinking of is irrelevant. They might even be known to be false, in proof by contradiction.
Then, you apply logical inference rules to the initial premises, and establish consequences of the initial premises.
One of these consequences may be what you are trying to prove. Or, you may come to a contradiction, and prove that initial premises are self-contradictory. Or you may come to no useful conclusion whatsoever.
Quoted 1 line
Now, I am not enjoying the discussion anymore.
Understandable.
Quoted 1 line
I've asked once before - remove me from the CC list.
I'll try to remember to do that.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 10 lines · Alexandre Oliva
Once again, now with clearer starting conditions (not intended to match TiVo in any way, BTW; don't get into that distraction)
Vendor doesn't care about tivoizing, their business works the same either way.
Vendor's employees will contribute the same, one way or another, so their contributions are out of the equation.
no, this is what you are missing.
with tivoizing, the vendors employees are working on linux and contributing
without tivoizing and with GPLv3 license involved the vendors employees are working on a propriatary OS and are contributing nothing back
Quoted 2 lines
Users get source code in either case, and they can modify it and share it. They're in no way stopped from becoming part of the community.
not if the GPLv3 achieves it's objectives of forcing the vendor to stop useing opensource software.
Quoted 16 lines
Given these conditions:
In a tivoized device, users will be unable to scratch their itches. This doesn't stop them from contributing to the project, but they may lack self-interest motivation to contribute, because they won't be able to use their modifications in the device they own.
In a non-tivoized device, users can scratch their itches. They can contribute just as much as they would in a tivoized device, but since they can use the changes they make to make their own devices work better for them, this works as a motivator for them to make changes, and perhaps to contribute them. Therefore, they will tend to contribute more.
Quoted 2 lines
Can you point out any flaw in this reasoning, or can we admit it as true?
in a tivoized device, users have access to the source code and can understand how things work, incoporate improvements into other projects, etc.
In addition, users who bypass the lockdown restrictions can modify the software on that device.
in a non-tivoized device users have a black-box and have to reverse engineer everything becouse the vendor releases no source at all.
David Lang
Quoted 6 lines · On Jun 19, 2007, [email protected] wrote:
Quoted 5 lines · Alexandre Oliva
Once again, now with clearer starting conditions (not intended to match TiVo in any way, BTW; don't get into that distraction)
Quoted 7 lines
Quoted 5 lines
Vendor doesn't care about tivoizing, their business works the same either way.
Vendor's employees will contribute the same, one way or another, so their contributions are out of the equation.
no, this is what you are missing.
No, this is just one of many possible situations.
You're just thinking of a different scenario, in which tivoization is important for the tivoizer. We'll get to that.
Please be patient. Don't get the illusion that you're disputing or weakening the argument with these statements. Disputing premises of a logical inference is pointless as far as disputing a larger argument that this particular lemma would be a part of.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 18 lines · Alexandre Oliva
Quoted 6 lines · On Jun 19, 2007, [email protected] wrote:
Quoted 5 lines · Alexandre Oliva
Once again, now with clearer starting conditions (not intended to match TiVo in any way, BTW; don't get into that distraction)
Quoted 7 lines
Quoted 5 lines
Vendor doesn't care about tivoizing, their business works the same either way.
Vendor's employees will contribute the same, one way or another, so their contributions are out of the equation.
no, this is what you are missing.
No, this is just one of many possible situations.
Quoted 7 lines
You're just thinking of a different scenario, in which tivoization is important for the tivoizer. We'll get to that.
Please be patient. Don't get the illusion that you're disputing or weakening the argument with these statements. Disputing premises of a logical inference is pointless as far as disputing a larger argument that this particular lemma would be a part of.
if a company doesn't care about tivoizing then they won't do it, it takes time and money to tivoize some product and it will cause headaches for the company.
their reasons for wanting to tivoize a product may be faulty, but they think that the reasons are valid or they wouldn't go to the effort.
David Lang
Quoted 6 lines · On Jun 19, 2007, [email protected] wrote:
if a company doesn't care about tivoizing then they won't do it, it takes time and money to tivoize some product and it will cause headaches for the company.
their reasons for wanting to tivoize a product may be faulty, but they think that the reasons are valid or they wouldn't go to the effort.
Absolutely right. And we'll get to that. Please just be patient.
In fact, how much the company cares about tivoizing is completely irrelevant to that point. I shouldn't even have included it, but I did because I thought it would be useful as a boundary condition.
So just disregard that.
Is there agreement that, comparing tivoized and non-tivoized hardware, we get'd more contributions if the hardware is not tivoized, because users can scratch their own itches, than we would for tivoized hardware?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 17 lines · Alexandre Oliva
Subject: Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
Quoted 6 lines · On Jun 19, 2007, [email protected] wrote:
if a company doesn't care about tivoizing then they won't do it, it takes time and money to tivoize some product and it will cause headaches for the company.
their reasons for wanting to tivoize a product may be faulty, but they think that the reasons are valid or they wouldn't go to the effort.
Absolutely right. And we'll get to that. Please just be patient.
In fact, how much the company cares about tivoizing is completely irrelevant to that point. I shouldn't even have included it, but I did because I thought it would be useful as a boundary condition.
Quoted 6 lines
So just disregard that.
Is there agreement that, comparing tivoized and non-tivoized hardware, we get'd more contributions if the hardware is not tivoized, because users can scratch their own itches, than we would for tivoized hardware?
if you also make the assumption that the company won't use propriatary software instead then I think you would get agreement. but the disagrement is over this exact assumption. you assume that these companies will use non-tivoized products if you make it hard to use the software covered by the GPL, most other people are saying that they disagree and the result would be fewer companies useing software covered by GPL instead.
David Lang
Quoted 2 lines · On Jun 19, 2007, [email protected] wrote:
if you also make the assumption that the company won't use propriatary software instead then I think you would get agreement.
Ah, good point. When I posed the one of the two cases of the inicial scenario as "no tivoization", I meant Free Software without constraints.
Quoted 5 lines
but the disagrement is over this exact assumption. you assume that these companies will use non-tivoized products if you make it hard to use the software covered by the GPL, most other people are saying that they disagree and the result would be fewer companies useing software covered by GPL instead.
I understand that. And what I'm saying is that, even if fewer such companies use GPLed software, you may still be better off, out of additional contributions you'll get from customers of companies that switch from tivoization to unconstrained Free Software, because of the additional costs of the alternatives.
And no, I can't prove it, but it's good that at least the argument is no longer completely disregarded while something else is disputed.
Now that you guys at least understand what the argument is, you can figure out the solution by yourselves.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 21 lines · Alexandre Oliva
Quoted 2 lines · On Jun 19, 2007, [email protected] wrote:
if you also make the assumption that the company won't use propriatary software instead then I think you would get agreement.
Ah, good point. When I posed the one of the two cases of the inicial scenario as "no tivoization", I meant Free Software without constraints.
Quoted 5 lines
but the disagrement is over this exact assumption. you assume that these companies will use non-tivoized products if you make it hard to use the software covered by the GPL, most other people are saying that they disagree and the result would be fewer companies useing software covered by GPL instead.
I understand that. And what I'm saying is that, even if fewer such companies use GPLed software, you may still be better off, out of additional contributions you'll get from customers of companies that switch from tivoization to unconstrained Free Software, because of the additional costs of the alternatives.
Quoted 6 lines
And no, I can't prove it, but it's good that at least the argument is no longer completely disregarded while something else is disputed.
Now that you guys at least understand what the argument is, you can figure out the solution by yourselves.
good, if you are no longer going to claim that your opinion on this unprovable point is the Truth (not the capitol) hopefully you can accept that a lot of very smart people are convinced that you are wrong on this point and not just 'confused'
David Lang
Quoted 9 lines · Alexandre Oliva
Yes. How does this relate with the piece of the argument I've proposed so far, or the whole argument I've posted before?
Answer: It doesn't. At all. You're just showing you didn't understand the argument. Which shows why I have to explain it piece by piece. Which suggests you shouldn't try to jump to conclusions.
Quoted 6 lines
Once again, now with clearer starting conditions (not intended to match TiVo in any way, BTW; don't get into that distraction)
Vendor doesn't care about tivoizing, their business works the same either way.
Not true. A PVR that can record pay per view and encrypted digital channels and other such things (as mandated by MPAA or whoever is unfortunately in charge of such stupid thigns), is much easier to sell to customers, and as a result they have to prevent things that the content providers want prevented (never mind that such prevention is futile because there is no DRM that can't be broken, only made more difficult to break). So their business does not work anywhere near as well if they can't sell devices that do what their customers want them to do (which is record stuff from TV, not play tetris, even though tetris might be fun too). Now if they don't do what the content providers want, they have to sell an inferior PVR, while a competitor that does what the content providers demand (in order to provide what customers demand), then they won't sell very much. They could go with another non GPL os and software of course, and spend way more moeny on it (or potentially use BSD or something) and deal with the problem that way. Either way GPL software gets no contributions under the GPL v3 if the company wants to be competitive in that business.
Quoted 2 lines
Vendor's employees will contribute the same, one way or another, so their contributions are out of the equation.
No, because they are no longer using GPL software.
Quoted 2 lines
Users get source code in either case, and they can modify it and share it. They're in no way stopped from becoming part of the community.
The users don't get any code now, because the vendor has no requirement to release it, or may even have no permission to realse it. And end users are for the most part incapable of contributing anyhow.
Quoted 6 lines
Given these conditions:
In a tivoized device, users will be unable to scratch their itches. This doesn't stop them from contributing to the project, but they may lack self-interest motivation to contribute, because they won't be able to use their modifications in the device they own.
I bought some hardware and built a box to run mythtv. I can only record stuff from regular analog cable, because I can't buy any hardware that supports digital encrypted cable. If I bought a locked down PVR instead, then I would be able to record that, but I wouldn't be able to play around. To me playing around was more important than recording encrypted channels, but to most customers, being able to record everything is what is important.
Quoted 6 lines
In a non-tivoized device, users can scratch their itches. They can contribute just as much as they would in a tivoized device, but since they can use the changes they make to make their own devices work better for them, this works as a motivator for them to make changes, and perhaps to contribute them. Therefore, they will tend to contribute more.
Their device can't physically do some of the thigns they wanted it to do though, since those features disappeared when the tivoization was removed due to licensing issues with the content makers.
Quoted 2 lines
Can you point out any flaw in this reasoning, or can we admit it as true?
Certainly fails to be true.
Who is to blame? The content makers for requiring lock down of devices that can access their content? Consumers for demanding to be able to use that content and hence buying devices that can use it? The company that made a device that did what consumers demanded by implementing it the way required by the content makers?
--
Len Sorensen
Quoted 10 lines · On Jun 20, 2007, [email protected] (Lennart Sorensen) wrote:
Quoted 6 lines · Alexandre Oliva
Once again, now with clearer starting conditions (not intended to match TiVo in any way, BTW; don't get into that distraction)
Vendor doesn't care about tivoizing, their business works the same either way.
Not true. A PVR that can record pay per view and encrypted digital channels
You see the "not intended to match TiVo" above?
Do you see that it's pointless to dispute antecedents of a logical inference rule, if you don't know what role it plays in the full argument?
Consider that this could be a proof by contradiction to realize how pointless your objection is, no matter how true the point you state is. It bears no relationship with the argument at hand, and you said so yourself, by disputing the assumptions of the inference, rather than its conclusions.
Assumptions that were not even used to arrive at the conclusions, BTW.
Quoted 4 lines
Quoted 2 lines
Can you point out any flaw in this reasoning, or can we admit it as true?
Certainly fails to be true.
Once you change the conditions to twist whatever else you want, then you arrive at different conclusions. What's the surprise here?
What you're doing is like, after getting an argument like this:
assumptions:
1+1 = 2
2+1 = 3provable consequence:
1+1+1 = 3
responding:
no, no, that's wrong! the right argument is:
assumptions:
3+4 = 7
2+1 = 3provable consequence:
2+1+4 = 7
therefore the answer is 5, not 3!
You see how illogical this is?
It doesn't matter whether your argument is correct. It just doesn't dispute the proposition at hand.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 16 lines · Daniel Hazelton
Quoted 7 lines
Dispute this:
non-tivoized hardware => users can scratch their itches => more contributions from these users
tivoized hardware => users can't scratch their itches => fewer contributions from these users
Linus doesn't have to. Statistically the number of people that will even think of modifying the code running on a "tivoized" device is minute - at most 5% of the users of such a device. Of those people the ones with the skill to actually do the work is an even smaller number - figure 2.5 to 3% of them. Of those with the skill, probably about 10% of them are actually *good* enough at it for their changes to be useful. Of that number, figure that only 25%, at most, will contribute the changes back.
Quoted 6 lines
Apply that to a sample case:
"tivoized" device total users: 1,000,000
people that think about modifying: 50,000 (5%)
people with skill: 1500 (3%)
people who are good enough for the changes to be useful: 150 (10%) those who will contribute them back: 38 (25%)
based on my experiance looking at the software released for tivos, I think you are over-estimating these numbers. if there are more then a dozen people producing things that are good enough to be useful and releasing their results as opensource software I would be surprised.
and for all that the FSF is claiming that tivos can't being modified it's really not that hard to change.
David Lang
Quoted 4 lines · On Jun 19, 2007, [email protected] wrote:
based on my experiance looking at the software released for tivos, I think you are over-estimating these numbers. if there are more then a dozen people producing things that are good enough to be useful and releasing their results as opensource software I would be surprised.
Yup.
And how many more would there be should it not be tivoized? More hackers would buy the devices, a number of them with the explicit intent and interest in modifying the software in it.
You're losing all that.
Quoted 2 lines
and for all that the FSF is claiming that tivos can't being modified it's really not that hard to change.
But is it legal?
How many would contribute changes to a list where there are TiVo people watching, which might expose these contributors to liabilities?
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 7 lines · Alexandre Oliva
Quoted 4 lines · On Jun 19, 2007, [email protected] wrote:
based on my experiance looking at the software released for tivos, I think you are over-estimating these numbers. if there are more then a dozen people producing things that are good enough to be useful and releasing their results as opensource software I would be surprised.
Quoted 7 lines
Yup.
And how many more would there be should it not be tivoized? More hackers would buy the devices, a number of them with the explicit intent and interest in modifying the software in it.
You're losing all that.
based on the knowledge shown by these users you aren't loosing much.
remember, not all tivo models are locked down, and most of those that are locked down can be unlocked pretty easily. it's only a couple models that have required a soldering iron.
the xbox hacks show that people will produce ways to bypass hardware restrictions that are easy for less techinical people to use (and people less technical then this aren't gong to be contributing code anyway)
as a result of watching the hacker groups I can safely say that the lockdown has not blocked many users. it has slowed modification of the hacks to new types of hardware, but not for very long.
Quoted 4 lines
Quoted 2 lines
and for all that the FSF is claiming that tivos can't being modified it's really not that hard to change.
But is it legal?
that depends on what lawyer you ask, and what they think of the applicability of the DMCA.
Quoted 2 lines
How many would contribute changes to a list where there are TiVo people watching, which might expose these contributors to liabilities?
in the case of Tivo, there are many public boards that talk about hacking tivos, and tivo employees login and contribute to them. just about all of these boards ban specific topics. for example:
getting service without paying for it is banned on every board I personally look at
transferring video off of the tivo is banned on some boards, but not on others.
David Lang
Quoted 4 lines · On Jun 19, 2007, [email protected] wrote:
Quoted 1 line · Alexandre Oliva
You're losing all that.
based on the knowledge shown by these users you aren't loosing much.
Remember, the sample is biased, the hackers who'd like to hack it are less likely to buy it, and some might be only making private changes to avoid legal hassles.
And then, it doesn't have to be all that much. Since you're talking specifically about TiVo customers, how much has TiVo effectly contributed in terms of code? (It was pointed out before, I know, but it's important to keep this in perspective, lest people lose sight of what's at stake)
Quoted 1 line
remember, not all tivo models are locked down,
Only the earliest that you can't find for sale any more, right?
Quoted 3 lines
as a result of watching the hacker groups I can safely say that the lockdown has not blocked many users. it has slowed modification of the hacks to new types of hardware, but not for very long.
Well, then... What's the point *for* tivoization, again? To slow down the contributions? And that's good because...?
Quoted 2 lines
in the case of Tivo, there are many public boards that talk about hacking tivos, and tivo employees login and contribute to them.
Nice!
Thanks for the info, this was very enlightening.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 11 lines · Alexandre Oliva
Quoted 4 lines · On Jun 19, 2007, [email protected] wrote:
Quoted 1 line · Alexandre Oliva
You're losing all that.
based on the knowledge shown by these users you aren't loosing much.
Remember, the sample is biased, the hackers who'd like to hack it are less likely to buy it, and some might be only making private changes to avoid legal hassles.
Quoted 5 lines
And then, it doesn't have to be all that much. Since you're talking specifically about TiVo customers, how much has TiVo effectly contributed in terms of code? (It was pointed out before, I know, but it's important to keep this in perspective, lest people lose sight of what's at stake)
they contributed all changes they made to GPL code, exactly as required.
in addition they have provided the full build environments, not just listing the versions of the compilers, etc.
Quoted 10 lines
Quoted 1 line
remember, not all tivo models are locked down,
Only the earliest that you can't find for sale any more, right?
Quoted 3 lines
as a result of watching the hacker groups I can safely say that the lockdown has not blocked many users. it has slowed modification of the hacks to new types of hardware, but not for very long.
Well, then... What's the point *for* tivoization, again? To slow down the contributions? And that's good because...?
no, the point is that while tivoization is not nessasarily the best thing it's far better then the company useing propriatary code.
delayed contributions are better then no contributions.
David Lang
Quoted 13 lines · On Tuesday 19 June 2007 19:49:24 [email protected] wrote:
Quoted 11 lines · Alexandre Oliva
Quoted 1 line · On Jun 19, 2007, [email protected] wrote:
remember, not all tivo models are locked down,
Only the earliest that you can't find for sale any more, right?
Quoted 3 lines
as a result of watching the hacker groups I can safely say that the lockdown has not blocked many users. it has slowed modification of the hacks to new types of hardware, but not for very long.
Well, then... What's the point *for* tivoization, again? To slow down the contributions? And that's good because...?
Quoted 6 lines
no, the point is that while tivoization is not nessasarily the best thing it's far better then the company useing propriatary code.
delayed contributions are better then no contributions.
David Lang
This logic has been proven already. Apple used KHTML and KJS as the backend systems for Safari. While Safari was in development they held onto all their changes and modifications. When they released Safari, they contributed it all back, making both things better. Fact: KHTML and KJS are also the core of Apples "WebKit" and "WebCore" technologies - both KHTML and KJS are Open Source projects, making up a part of KDE.
DRH
Signature · 1 line
Dialup is like pissing through a pipette. Slow and excruciatingly painful.
Hi Alexandre,
Quoted 7 lines · Alexandre Oliva
Dispute this:
non-tivoized hardware => users can scratch their itches => more contributions from these users
tivoized hardware => users can't scratch their itches => fewer contributions from these users
Maybe, but in what numbers? It's not like every user out there is able to scratch their itches in the first place. Besides, people who want to hack their hardware either work around the restrictions (like Linux on xbox) or buy hardware that doesn't have any.
In any case, I hope we call quits on this thread. People have already explained to you that there are certain devices (ATMs, medical equipment) where you absolutely want "tivoization". Furthermore, Linus has repeatedly explained to you why he (and bunch of other _kernel_ hackers -- this is a thread on LKML, remember) thinks it's stupid for a software license to restrict hardware design choices. If you don't see the point, fine, but lets stop wasting each others time, ok?
Quoted 11 lines · Pekka Enberg
Hi Alexandre,
Quoted 7 lines · Alexandre Oliva
Dispute this:
non-tivoized hardware => users can scratch their itches => more contributions from these users
tivoized hardware => users can't scratch their itches => fewer contributions from these users
Maybe, but in what numbers?
We'll get to that.
Quoted 4 lines
Furthermore, Linus has repeatedly explained to you why he (and bunch of other _kernel_ hackers -- this is a thread on LKML, remember) thinks it's stupid for a software license to restrict hardware design choices.
Most of these explanations carried the assumption that this would lower the amount of contributions he'd get. And this is precisely the circular logic in their arguments, and this is why this argument is intended to show the initial assumption is false. But hey, maybe I'm wrong, and it won't show that at all, and I'll be proved wrong, and then there's going to be one more very vocal defensor of your ideals.
Now, if people want to keep on fooling themselves, I guess they can, as long as they don't hurt others in the process. If they try to fool others, I feel it is my moral duty to intervene.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Hi Alexandre,
Quoted 4 lines · At some point in time, I wrote:
Quoted 4 lines
Furthermore, Linus has repeatedly explained to you why he (and bunch of other _kernel_ hackers -- this is a thread on LKML, remember) thinks it's stupid for a software license to restrict hardware design choices.
Quoted 2 lines · Alexandre Oliva
Most of these explanations carried the assumption that this would lower the amount of contributions he'd get.
Then may I suggest you go back and re-read those threads because that's not at all what I am reading.
Quoted 3 lines · Alexandre Oliva
Now, if people want to keep on fooling themselves, I guess they can, as long as they don't hurt others in the process. If they try to fool others, I feel it is my moral duty to intervene.
Sure. I just wanted to point out that your "moral duty" really comes off as "trolling" at this end. So could we please move this discussion to maybe "[email protected]" or some other more appropriate list and stop flooding LKML?
Pekka
Quoted 3 lines · Alexandre Oliva
technically, it asks you to pass on (!= give back) access to the software (not to the hardware that contains it).
That "technically" is just another way of saying "if you look cross-eyed at it, and don't look too closely".
Quoted 3 lines
No. The reason, again, is the portion you snipped out.
Could you try again?
No.
Not worth my time. You have shown yourself unable to learn and understand.
Just live with it. The GPLv2 is better for the kernel.
Linus
Quoted 11 lines · Alexandre Oliva
Quoted 8 lines · Linus Torvalds
I just want software back. I think it is *wrong* for me to ask for anything else. It's literally my personal "moral choice": I think the hardware manufacturers need to make their _own_ choices when it comes to _their_ designs.
- I think that *technical*quality* is more important than *quantity*.
This argument fails to make the point you're trying to make.
No, it has been countered many times and you are simply proving your ignorance.
Quoted 9 lines
you trade the potential contributions of all those users for the contributions of tivoizers, apparently assuming that all tivoizers would simply move away from the community, taking their future contributions away from your community, rather than moving to a position in which you'd get not only the contributions from the company itself, but also from all their users
and you say "oh, I don't care about quantity, I care about quality", as if this somehow related with the above.
Being strictly pragmatic - what makes you think TiVoland is some fledgling grounds of /expert kernel developers/ that are otherwise deprived of contributing, unless they can illegally modify their TiVo?
Under GPLv2, we have the kernel modifications and can include them in our software. If you don't agree with TiVo, purchase an open product. Download their kernel source and use it on your open product. Pure consumerism and capitalism at work.
The GPLv3 is a solution in search of a problem. Worse, it creates problems outside the grasp of your understanding.
Quoted 4 lines
Just do the math. Hypothetically, Linux goes GPLv3, without permission to tivoize. TiVo has to decide among:
- switching to another kernel, no further contributions from them
Bad for us, bad for users.
Quoted 1 line
- sticking with old version, no further usable contributions from them
Bad for us, bad for users.
Quoted 1 line
- switching to ROM, still the same contributions from TiVo
Bad for users.
Quoted 1 line
- no more tivoization, contributions from TiVo and users
Bad for us, bad for users. Legitimate laws and practices require that certain devices not be modified by end users. Therefore TiVo fails and contributions cease.
Quoted 2 lines
So, you see, in no case do you get more contributors while at the same time losing TiVo's quality contributions.
No. Outside of this FSF {academic,religious} diatribe, in the real world, things aren't as you see them. The fact is that these people can get the code and contribute. It just won't run on TiVo. So don't buy TiVo but use the code. Problem solved, free software in action.
Quoted 8 lines
Quoted 3 lines
In the GPLv3 world, we have already discussed in this thread how you can follow the GPLv3 by making the TECHNICALLY INFERIOR choice of using a ROM instead of using a flash device.
Yes. This is one option that doesn't bring any benefits to anyone. It maintains the status quo for users and the community, but it loses the ability for the vendor to upgrade, fix or otherwise control the users. Bad for the vendor.
And users. Don't spin the facts.
You are advocating things which hurt the end user, which the "Spirit" should seek to help. The GPLv3 is here to stroke the FSF ego because they don't like how somebody has legally found a way to use free software in a way they don't agree with. Sort of like dynamite being used violently...
Quoted 3 lines · Kevin Bowling
Legitimate laws and practices require that certain devices not be modified by end users. Therefore TiVo fails and contributions cease.
I've never denied this possibility.
But how about all the other devices that are being tivoized that do NOT require this?
Are people just blind to this possibility? Or does it really not exist, and I'm the first who ever though of it?
Quoted 6 lines
Quoted 4 lines
Yes. This is one option that doesn't bring any benefits to anyone. It maintains the status quo for users and the community, but it loses the ability for the vendor to upgrade, fix or otherwise control the users. Bad for the vendor.
And users. Don't spin the facts.
How is it good that the vendor can downgrade the software behind the user's back?
Oh, yeah, right, it could upgrade it too.
Quoted 1 line
You are advocating things which hurt the end user,
Actually, no. I'm advocating for respect for users' freedoms. Whoever choose not to do that gets slightly hurt in the process, as an incentive for respecting users' freedoms. And yes, when the users' freedoms are not respected (the cases you mentioned), that's bad for the user, no doubt. And bad for the community as well.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 9 lines · Alexandre Oliva
Quoted 3 lines · Linus Torvalds
In the GPLv3 world, we have already discussed in this thread how you can follow the GPLv3 by making the TECHNICALLY INFERIOR choice of using a ROM instead of using a flash device.
Yes. This is one option that doesn't bring any benefits to anyone. It maintains the status quo for users and the community, but it loses the ability for the vendor to upgrade, fix or otherwise control the users. Bad for the vendor.
Not really, Tivo could simply sell you a box without any installed software. The actual software is mailed to you on a credit card sized ROM when you activate service. When they want to (or need to) update the software they send out a new ROM card, maybe yearly as part of the service subscription renewal.
The box could even be sold by third party vendors, I think they may even have started off that way, my old Series 1 had a big Philips logo on it. So now we make sure that this hardware refuses to boot any unsigned code, but it wasn't shipped containing GPLv3 software, so it's license terms simply does not apply.
The software is shipped on a ROM card which can no longer be modified by the manufacturer or any third party, so it would seem to comply with the GPLv3. I can even imagine that the hardware is really general purpose but the ROM is encrypted so that only the BIOS/bootloader can unlock it.
So the GPLv3 seems to fall short on actually preventing tivoization. It just requires an extra layer of indirection, ship hardware seperately from software.
Jan
Quoted 12 lines
The box could even be sold by third party vendors, I think they may even have started off that way, my old Series 1 had a big Philips logo on it. So now we make sure that this hardware refuses to boot any unsigned code, but it wasn't shipped containing GPLv3 software, so it's license terms simply does not apply.
The software is shipped on a ROM card which can no longer be modified by the manufacturer or any third party, so it would seem to comply with the GPLv3. I can even imagine that the hardware is really general purpose but the ROM is encrypted so that only the BIOS/bootloader can unlock it.
Quoted 5 lines
So the GPLv3 seems to fall short on actually preventing tivoization. It just requires an extra layer of indirection, ship hardware seperately from software.
Jan
If this flaw is still the latest GPLv3 drafts, it should be fixed. It's a simple, technical error that can easily be rectified.
I don't have the latest draft handy, but my recollection is that it talks about software that is "transferred along with" hardware. What it should say is object code that contains activation logic for hardware.
That is, if I ship the Linux kernel (assuming for the moment the Linux kernel was under GPLv3) with special activation logic to run on platforms X, Y, and Z, then the source code should have to include how to make my changes have that same special activation logic for those same platforms.
If the GPLv3 doesn't do that, it's broken.
Note that I'm not opining on whether this is a good thing or a bad thing. But people who choose the GPLv3 because they want to prohibit tivoization of their software should in fact get that prohibition.
DS
Quoted 2 lines · Jan Harkes
Not really, Tivo could simply sell you a box without any installed software.
Yes.
Quoted 2 lines
The actual software is mailed to you on a credit card sized ROM when you activate service.
If that's a separate transaction, then yes, I believe it would not be convered under the terms of GPLv3, but IANAL. There might be some catch about intent, and also about contractual obligations in the hardware sale to provide the software (coupons anyone? :-) or some such.
But then, that you sell specialized devices with say only MIT-licensed software shouldn't stop you from selling CD-ROMs with GPLed software, even if those CD-ROMs could possibly run on that device.
The GPLv3 won't remove every way in which people who want/need to stop the user from making changes to the software could accomplishing this (ROM). It will just make this a bit more inconvenient, such that vendors that have the option respect users' freedoms, and those that find it too inconvenient respect the wishes of users who don't want their software turned non-free.
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}Quoted 2 lines · Alexandre Oliva
Quoted 2 lines
The actual software is mailed to you on a credit card sized ROM when you activate service.
...
Quoted 6 lines
The GPLv3 won't remove every way in which people who want/need to stop the user from making changes to the software could accomplishing this (ROM). It will just make this a bit more inconvenient, such that vendors that have the option respect users' freedoms, and those that find it too inconvenient respect the wishes of users who don't want their software turned non-free.
I am trying to read that last sentence and it just doesn't seem to make any sense.
Or are you saying that all that anti-tivoization language that adds complex requirements which change depending on the market some device happens to be sold in and which will most likely make GPLv3 software unusable for various applications ranging from medical equipment to financial transaction systems (and probably others) is there to just make it a _bit_ more inconvenient for vendors to implement a tivo-like scheme?
So what exactly is the point of all this then?
Jan
Quoted 11 lines · Jan Harkes
Quoted 2 lines · Alexandre Oliva
Quoted 2 lines
The actual software is mailed to you on a credit card sized ROM when you activate service.
...
Quoted 6 lines
The GPLv3 won't remove every way in which people who want/need to stop the user from making changes to the software could accomplishing this (ROM). It will just make this a bit more inconvenient, such that vendors that have the option respect users' freedoms, and those that find it too inconvenient respect the wishes of users who don't want their software turned non-free.
Quoted 3 lines
Or are you saying that all that anti-tivoization language that adds complex requirements which change depending on the market some device happens to be sold in
You allude to the definition of User Product, not geographies, I suppose.
Quoted 3 lines
and which will most likely make GPLv3 software unusable for various applications ranging from medical equipment to financial transaction systems (and probably others)
Not unusable, except perhaps for the one example about credit card terminals presented so far.
Quoted 2 lines
is there to just make it a _bit_ more inconvenient for vendors to implement a tivo-like scheme?
I'm not sure they find it to be "just a bit".
Point is to keep Free Software Free freedoms, and ROM doesn't make it non-Free, so this provision is a means to ensure the compliance with the wishes of users who want their software to not be used in ways that make it non-Free.
As it so happens, this also places economic pressure on vendors who tivoize, such that they either face more costly solutions, or enable users to tinker with the software as well. And if neither ROM nor permission are an option for the vendor, well, too bad, the author gets to decide how his software is to be used, right?
Quoted 1 line
So what exactly is the point of all this then?
Keeping Free Software Free.
(and, as a consequence that many of you may welcome, keeping open-source software open source)
Signature · 4 lines
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}