Return-Path: X-Greylist: delayed 471 seconds by postgrey-1.34 at layers.openembedded.org; Tue, 17 Mar 2020 02:46:50 UTC Received: from adige.pair.com (adige.pair.com [209.68.5.114]) by mail.openembedded.org (Postfix) with ESMTP id 367CB60C66 for ; Tue, 17 Mar 2020 02:46:50 +0000 (UTC) Received: from adige.pair.com (localhost [127.0.0.1]) by adige.pair.com (Postfix) with ESMTP id D1246959CE; Mon, 16 Mar 2020 22:39:00 -0400 (EDT) Received: from linc.localnet (unknown [IPv6:2406:e003:130a:e501:f188:4b83:8582:1e57]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by adige.pair.com (Postfix) with ESMTPSA id 0B1DE959AA; Mon, 16 Mar 2020 22:38:59 -0400 (EDT) From: Paul Eggleton To: tsc@lists.openembedded.org Date: Tue, 17 Mar 2020 15:38:56 +1300 Message-ID: <22665987.Wx5bFTl31r@linc> MIME-Version: 1.0 Cc: openembedded-core@openembedded.org, openembedded-devel@lists.openembedded.org Subject: OpenEmbedded TSC Meeting Minutes 2020-03-17 X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 17 Mar 2020 02:46:51 -0000 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="UTF-8" OpenEmbedded Technical Steering Committee (TSC) Meeting Minutes 2020-03-17 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Meeting was held in #oe-tsc on Freenode; channel access public. Present: =2D Richard Purdie (RP) =2D Joshua Watt (JPEW) =2D Bruce Ashfield (zeddii) =2D Paul Eggleton (bluelightning) Apologies: =2D Martin Jansa (JaMa) Summary: * Need to think about ways to encourage contributions - Make contributions more visible? (see also last meeting's discussion) - Include contributor names in release notes - Response emails to patches / when patch merged? - Ideas / discussion around any barriers to contributing and removing the= m very welcome =46ull meeting long =2D---------------- [11:01] I presume we are meeting now? [11:02] That's the plan [11:03] JaMa doesn't appear to be online, and RP is currently away [11:03] I forgot to send out a reminder today [11:04] I have just pinged RP - I could also text him [11:10] Hi, I'm here now, sorry I'm late [11:10] no worries [11:10] do we have zeddii? [11:11] Not sure. He was on earlier today [11:15] Is there an agenda? [11:15] I must admit I got M3 built and kind of went and "collapsed" ;= =2D) [11:16] I'm not aware of anything [11:16] I did have one item the YP TSC asked me to raise [11:16] Sorry for not officially adding it to the wiki etc [11:16] here. sorry, I was supposed to be on vacation this week, s= o I removed the meeting from my calendar. [11:17] Basically, it was to ask the OC TSC to consider the resourcing= situation we have and to consider what we could do about it. [11:17] zeddii: No problem, we just started [11:17] There are various bodies that are thinking about it (YP TSC, Y= P board, OE board) but the OE TSC probably also needs to give it some thoug= ht [11:19] RP: Ya, make sense. [11:20] Are there any identified reasons why people don't contribute? [11:22] JPEW: In general the pattern seems to be employers expecting s= omeone else to do the core work [11:22] Some things don't have clear return on investment too [11:22] agreed. and agreed. [11:22] which things? [11:23] anything that is mostly refactoring I would imagine [11:23] JPEW: build failure triage (SWAT), bug triage [11:23] things where its day to day monitoring without direct line of = sight to a problem they themselves see [11:23] and even things like why a reference kernel, when they don= =E2=80=99t use it directly. [11:23] zeddii: right [11:24] also, new feature development like hashequiv [11:24] or as bluelightning says, refactoring or new architectural work [11:25] yup. [11:28] I assume that I either got disconnected, or we are all dee= p in thought. [11:28] zeddii: The latter :) [11:29] I've spent a lot of time thinking about it and I'm not sure wh= at we can do [11:29] the problem is that only OSVs really care about =E2=80=9C= credibility=E2=80=9D in the community. if you know what I mean. [11:29] Aside from employeer expectations, any other reasons why peo= ple don't participate? [11:29] We have made some changes like newcomer bugs. I also need to s= end out some kind of summary of the challenges to the wider community. I'm = still working on that [11:30] JPEW: there is probably a knowledge barrier too [11:30] I could volunteer some of my time to help someone that act= ually picked up a new comer bug, but I think that=E2=80=99s probably just a= drop in the ocean. [11:30] JPEW: even some of our regular contributors think of bitbake a= s a blackbox, you're one of the few who doesn't [11:30] zeddii: Not necessarily if you can help someone get interest= ed, it pays dividends :) [11:31] zeddii: lots of small things do grow over time, I am trying to= encourage people. Not all will work out but its great to see when people d= o grow [11:32] that=E2=80=99s what I was wondering, are none of them bein= g picked up, since newcomers don=E2=80=99t know where to start. maybe pairi= ng them up with someone to office advice can help. but maybe that has alre= ady been tried (and yes, I know that pulls more time from people that don= =E2=80=99t have it, etc). [11:32] I=E2=80=99m throwing my hat into the ring for that. but th= at implies someone is looking for that kind of help. [11:32] zeddii: That's a good idea. I'd be willing to advise someone= fixing a bug also (newcomer or otherwise) [11:32] zeddii: we're seeing poor take up despite what I thought was a= good spectrum of bugs there [11:33] Perhaps more of a visibility problem? [11:33] I would note that the we lack YP advocacy people atm too :( [11:33] visibility is definitely part of it [11:34] It's hard being on the inside looking out and trying to figu= re out how visible you are sometimes [11:34] the newcomer doc on the YP wiki is pretty good I th= ink, but that assumes people find that [11:35] I did just get to it by googling "yocto project con= tribute" [11:36] Hmm, is it pretty easy to identify new contributers on the M= L? [11:37] JPEW: for a person or for a machine? [11:38] person (email address) [11:38] I=E2=80=99m thinking more of companies that use OE, not en= d users. since getting contributions from each category is quite a differen= t thing. [11:40] zeddii: Sure, but having people in the company that are invo= lved anyway helps [11:40] sometimes :eyeroll: [11:40] JPEW: I mean its easy for a computer to look at an email and s= ee if they've had a patch merged before [11:42] RP: Right. Would it be bad form to send them a message "welc= ome to the community" message? I've not been doing OSS very long, so I don'= t know all the edicate [11:42] Thanks for your patch! Here's some other ways you can get in= volved... [11:42] JPEW: if they were new to OE but an experienced OSS contributo= r it may seem patronising :/ [11:42] I think it might be nice - you could include "if yo= u don't get a reply in x days here's what to do" [11:43] I wouldn't find it patronising if I was joining a n= ew project, others may differ possibly [11:43] I know I've struggled when joining other projects, = people sort of expect you to know what to do / where to look and that's not= always easy even if you're experienced with OSS, every project is different [11:44] yah. is it a get started issue, credit issue, or lack of m= apping to ROI issue .. hard to say, since we don=E2=80=99t even have that i= nfo :( [11:45] zeddii: Hard to operate effectively without data [11:45] I don't have an objection to it [11:45] When do the other TSC's meet [11:46] By the time they sent a contribution they're already part way = there though, we may need to find new contributors as well [11:46] JPEW: you mean the YP TSC? [11:46] Ya [11:47] zeddii: the "data" I have says all the above [11:47] I jokingly say that with all of the users of OE, and a fai= rly narrow set of contributors, I assume that means that they have no issue= s. which I can=E2=80=99t imagine is the case. [11:47] JPEW: YP TSC meets after the engineering sync tomorrow [11:47] zeddii: most of my data is reading the mailing lists and socia= l media posts from users, new and experienced [11:49] OK. Maybe we can draft up a skeleton e-mail and all look it = over (maybe the YP TSC too) [11:49] non-corporate contributions are hard to capture, since the= y are likely elsewhere for many high level applications. [11:50] JPEW: another thing may be a "your patch has merged" email fro= m patchwork [11:50] RP: Ya that would be good too [11:50] I mean, it would be nice if we went to openembedded.org an= d saw some logo=E2=80=99s, but I suppose that collides with YP very quickly= =2E and you get a logo if you have a patch in the release ;) [11:51] I also have no objection to that but it needs someone else to = own and drive it [11:51] that @gmail.com company could get a big logo! [11:51] patch merged reports was something that was suppose= d to be added to patchwork a long time ago, unfortunately due to various re= asons it never got done :/ [11:51] bluelightning: one idea that did get talked about but probably= ironically, not with you was a list of contributors to the release [11:52] hmm, yes, I think maybe it did get mentioned somewh= ere [11:52] I thought it was a good idea, since then large and small c= ould see their name in =E2=80=9Cprint=E2=80=9D along with a release. [11:52] (ironically as bluelightning ends up writing the release notes) [11:52] but again, time. I can help with that when the time comes. [11:53] at least as far as commits go there are various git= stats analysis tools that might be helpful (and don't involve us writing a= dditional code) [11:53] zeddii: the time for that one is now, release is in weeks! [11:53] although yes it would be good to have the info actu= ally in the notes... that should be pretty easy for me to do [11:53] I mean, I get an email from the kernel once and a while tr= ying to figure out who I work for, since @gmail.com =E2=80=A6. so there=E2= =80=99s some mappings that can be done on that front as well. [11:54] bluelightning: you can integrate if someone helps too! :) [11:54] bluelightning: if there=E2=80=99s something you think I ca= n do for that, I can help. [11:54] zeddii: thanks - I'll keep that in mind [11:55] I=E2=80=99ll go look at the other release notes to see if = I can think of anything else that might be interesting to add. [11:55] i.e. if a release mentioned some steps forward in a vertic= al that interested some companies, they might be interested to help more in= the future, etc. hard to say if its even possible to do that, but its an i= dea. [11:58] There might be more we could do at conferences and such also= =2E.. although thats not useful in the short term [11:59] some kind of metrics/lists focused on "first contributions fro= m" might be interesting [12:03] Any other topics? [12:03] Not from me [12:03] I'll put this one on the list for next month so we can see i= f there are any more ideas in the meantime [12:04] And I'll try to draft some example e-mails [12:04] If people do have ideas, please do feel free to share. I think= the project has some challenges ahead (as does the world at large atm :( ) [12:06] OK, supper time for me. [12:06] Bye all [12:07] I'll send out the minutes [12:08] Thanks. Don't forget to link on the wiki: https://www.openem= bedded.org/wiki/TSC [12:08] ah yes thanks for the reminder, I think I missed th= at the last couple of times [12:09] Thanks all!