Those ruddy punctuation fixes
The current hubub was caused by a build breakage, right? And that build breakage was caused by a spellfix outside of a comment.
Simple rules to avoid that problem:
1. don't fix anything not in a comment. (spell-fix.pl doesn't.)
2. build after you fix
3. don't fix any comments in code you can't build
4. don't fix any punctuation at all
The earlier hubub was caused by people worried about improper fixes (removing puns, changing meaning, etc.) The simple expedient of having several literate people (sensitized to all the concerns raised in this thread) review the change should get rid of most of that risk.
So I don't think the spelling police have to beat a total retreat. They just have to follow the above rules of thumb.
Or is the hubub really about something else? Like people just hate the whole idea, and wish it would go away, and don't really want to discuss it rationally?
- Dan
Signature · 3 lines
Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045
Quoted 3 lines · Dan Kegel
Or is the hubub really about something else? Like people just hate the whole idea, and wish it would go away, and don't really want to discuss it rationally?
Spelling fixes are annoying ways to break patches that provide no user visible value. It also detracts from the character of the
code: who wants "drain brammage" to be replaced with "brain dammage"?
-ben
Signature · 1 line
Junk email? <a href="mailto:[email protected]">[email protected]</a>
Quoted 8 lines · Benjamin LaHaise
Quoted 3 lines · Dan Kegel
Or is the hubub really about something else? Like people just hate the whole idea, and wish it would go away, and don't really want to discuss it rationally?
Spelling fixes are annoying ways to break patches that provide no user visible value.
Indeed, breaking patches with spelling fixes is a big mistake, and should never be done.
Quoted 1 line
It also detracts from the character of the code: who wants "drain brammage" to be replaced with "brain dammage"?
Nobody. That's why I maintain a spelling police code of conduct at
http://www.kegel.com/kerspell/
along with my spellchecking tools. I explicitly say "Do not fix puns!" and "Have several literate people review your changes!"
- Dan
Signature · 3 lines
Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045
Quoted 8 lines · Benjamin LaHaise
Quoted 3 lines · Dan Kegel
Or is the hubub really about something else? Like people just hate the whole idea, and wish it would go away, and don't really want to discuss it rationally?
Spelling fixes are annoying ways to break patches that provide no user visible value. It also detracts from the character of the
code: who wants "drain brammage" to be replaced with "brain dammage"?
There is a difference btween plain wrong and misleading text, printing wrong things to the console, text that confuses non first language speakers and accidentally removing humour.
Someone broke a little bit of asm and it got fixed. Someone broke the pnp layer a lot longer before and its still not fixed. You have regular releases, stuff gets broken now and then.
People are actually *doing* things, give them a break.
Quoted 4 lines · Alan Cox
Quoted 3 lines
Spelling fixes are annoying ways to break patches that provide no user visible value. It also detracts from the character of the
code: who wants "drain brammage" to be replaced with "brain dammage"?People are actually *doing* things, give them a break.
There's a shitload of patches in the 2.4 commit archives that mostly still apply. With each iteration of spelling fixes, it becomes more and more work to weed through these to find out if things are really applied or not. In the beginning I used Rusty's 'isapplied' script. Its now pretty much useless, requiring manual inspection of code on every diff.
Spelling fixes aren't going to get us to 2.6.0 any faster. There are _dozens_ of known problems, and I'll take patches fixing real problems over spelling fixes any day.
Dave
Quoted 6 lines · Dave Jones
There's a shitload of patches in the 2.4 commit archives that mostly still apply. With each iteration of spelling fixes, it becomes more and more work to weed through these to find out if things are really applied or not. In the beginning I used Rusty's 'isapplied' script. Its now pretty much useless, requiring manual inspection of code on every diff.
True enough. What areas is 2.5 paticularly behind on right now, I've noticed the radio drivers are bit out here and there, what else is badly behind ?
On Fri, 7 Mar 2003 23:52:47 -0100 Dave Jones <[email protected]> wrote:
| On Fri, Mar 07, 2003 at 11:15:59PM +0000, Alan Cox wrote: | | > > Spelling fixes are annoying ways to break patches that provide no | > > user visible value. It also detracts from the character of the | > > code: who wants "drain brammage" to be replaced with "brain dammage"? | > People are actually *doing* things, give them a break. | | There's a shitload of patches in the 2.4 commit archives that | mostly still apply. With each iteration of spelling fixes, | it becomes more and more work to weed through these to find | out if things are really applied or not. In the beginning I | used Rusty's 'isapplied' script. Its now pretty much useless, | requiring manual inspection of code on every diff. | | Spelling fixes aren't going to get us to 2.6.0 any faster. | There are _dozens_ of known problems, and I'll take patches | fixing real problems over spelling fixes any day.
What/where is the "2.4 commit archives"?
Do you want/need any help with this?
--
~Randy
Quoted 1 line · Dave Jones
Spelling fixes aren't going to get us to 2.6.0 any faster.
One problem with spelling fixes is that there's never a good moment, nor a good procedure to do them.
Quoted 2 lines
There are _dozens_ of known problems, and I'll take patches fixing real problems over spelling fixes any day.
I guess it would be nice if Linus would give some advance warning when he's going to accept speelnig fixes. That way, people with "real work" could schedule a few weeks at the beach, while the spelling police does its grisly job.
Of course, we all know that this will never happen ;-)
- Werner
Signature · 3 lines
_________________________________________________________________________ / Werner Almesberger, Buenos Aires, Argentina [email protected] / /_http://www.almesberger.net/____________________________________________/
Quoted 8 lines · Werner Almesberger
One problem with spelling fixes is that there's never a good moment, nor a good procedure to do them.
Quoted 2 lines
There are _dozens_ of known problems, and I'll take patches fixing real problems over spelling fixes any day.
Quoted 6 lines
I guess it would be nice if Linus would give some advance warning when he's going to accept speelnig fixes. That way, people with "real work" could schedule a few weeks at the beach, while the spelling police does its grisly job.
Of course, we all know that this will never happen ;-)
The next best thing might be for the spelling police
to maintain an archive of all the accepted spellfix patches. That would let a sufficiently motivated person
(say, a spelling policeman) to offer a patch update service with a reasonable certainty that he could
bring arbitrary patches up to date with respect to
the spellfixes in the main tree.
I've outlined a proposal for this at http://www.kegel.com/kerspell
This spellfix business is way more work than is reasonable. The people who spend the time making it work while taking care of all these issues (avoiding pun removal, avoiding breakage, avoiding wrong corrections, patch updating) are clearly on acid, but I salute them. Me, I just take notes and watch :-)
- Dan
Signature · 3 lines
Dan Kegel http://www.kegel.com http://counter.li.org/cgi-bin/runscript/display-person.cgi?user=78045
Quoted 2 lines · Dan Kegel
The next best thing might be for the spelling police to maintain an archive of all the accepted spellfix patches.
At the end of the day, bulk fixes may just be the least evil.
Alternatives like passing all spelling fixes individually through the appropriate channels for each subsystem would still cause conflicts, and also burden Linus with lots of re-iterations of the same theme, coming from many directions and spread over weeks.
A bit of coordination would be useful, though. E.g. I suppose it would hurt people like Dave Jones (with his 2.4 patch queue) a lot less if the spelling changes would happen while their queue is reasonably short.
Quoted 1 line
This spellfix business is way more work than is reasonable.
It will be easier the next time ;-)
- Werner
Signature · 3 lines
_________________________________________________________________________ / Werner Almesberger, Buenos Aires, Argentina [email protected] / /_http://www.almesberger.net/____________________________________________/