17:59:54 <lamby> #startmeeting 17:59:54 <MeetBot> Meeting started Tue Nov 15 17:59:54 2016 UTC. The chair is lamby. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:59:54 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic. 18:00:06 <lamby> Hello all, welcome to our… 4th IRC meeting. 18:00:13 <h01ger> this year 18:00:15 <h01ger> :) 18:00:17 <lamby> Who is about? :) 18:00:26 <lamby> Our agenda is here: https://pad.riseup.net/p/reproducible-irc-meeting-4 18:00:54 <lamby> Will let the dust settle and then we'll crack on ASAP; I'm sure folks have places to be. 18:01:00 * vagrantc waves 18:01:08 * deki \o 18:01:11 * HW42 18:01:13 * neverpanic 18:01:23 * h01ger is very tired and wants to go afk 18:01:25 <lamby> hey deki, HW42, neverpanic, emaste 18:01:29 <emaste> hello! 18:02:03 <h01ger> hi hi :) 18:02:20 <boklm> hello 18:02:30 <lamby> hey boklm 18:02:42 * lamby has tea. 18:03:04 <h01ger> can we start please? 18:03:17 <spectranaut> hi too! 18:03:28 <lamby> h01ger: Just waiting for the stragglers… for example ^ 18:03:30 <lamby> hey spectranaut 18:03:46 <lamby> siamezzze: o 18:03:58 <lamby> Cool, lets get going 18:04:13 <lamby> #topic tracker for reproducible-builds.org 18:04:34 <lamby> This is a followup to the task tracker discussion from last meeting. 18:04:55 <infinity0> hi 18:04:57 <lamby> Who added this? I'm not sure what it's *specically* about. 18:05:24 <infinity0> added what? 18:05:25 <deki> issue/task tracker? 18:05:25 <vagrantc> there was discussion about simplifying the code needed to run 18:05:36 <h01ger> its unresolved from last meeting 18:05:37 <vagrantc> e.g. all in one script rather than git hooks 18:05:49 <lamby> Who added this to the agenda? What needs to be discussed here. 18:05:59 <h01ger> its unresolved from last meeting 18:06:05 <deki> has anyone else used it since last meeting? 18:06:12 <deki> or opinions about it? 18:06:37 * h01ger wants a webbased on or one where on can commit easily plaintextz in git 18:06:42 <infinity0> i added some things, other people didn't want to use it because it runs code 18:06:44 * vagrantc wonders if future meeting agenda items shouldn't come with a point-person or be skipped 18:06:53 <h01ger> i dont care about decentralised 18:07:00 <emaste> vagrantc: good point. 18:07:23 <lamby> From a distance, and probably therefore from the point-of-view of a newbie to reproducible, it looks complicated, even if it ran no code whatsoever. 18:08:08 <lamby> (it might /not/ be complicated, but perceptions, etc.) 18:08:36 <HW42> I think a repo with .md file would do the job and can be easily generate html. but I would also not realy object to a "classic" browser/server based tracker 18:08:58 <lamby> Does this mean we want to change tech? Or keep at this one for a bit? (They will all have various downsides after all, but switching all the time is also kinda busywork.) 18:09:02 <h01ger> next topic? somebody needs to do this/something and than we can discuss again 18:09:03 <vagrantc> it really depends on how many issues we're likely to track 18:09:36 <HW42> FWIW I think a pseudo packacge in the debian BTS would also work, but IIRC the whole point was to not use the Debian BTS 18:09:50 * h01ger now tracks his urgent todo with local uncommitted modifications of TODO from jenkins.d.n.git ;) i needed *something* 18:10:11 <emaste> :) 18:10:13 <lamby> infinity0: thoughts? eg. fancy just moving to plaintext for now or taking this task (lol) for now? 18:10:25 <infinity0> moving to plaintext is fine by me 18:10:38 <lamby> infinity0: Cool. Okay to action-time you that? 18:10:57 <infinity0> sure 18:11:01 <lamby> #action infinity0 To move tasks.git to plain text 18:11:05 <lamby> thakns 18:11:23 <lamby> :) 18:11:25 <lamby> #topic DDPO & json 18:11:33 <h01ger> next, nothing happened 18:11:38 <lamby> h01ger: hm? 18:12:15 <h01ger> DDPO & json - no changes since last week, the todo+status is on the list 18:12:24 <lamby> cool, thanks 18:12:25 <h01ger> DDPO & json - no changes since last week, the todo+status is in the last meetings log 18:12:51 <lamby> #topic Upcoming events/presentations 18:12:59 <lamby> first, libreplanet 18:13:03 <vagrantc> i seem to be collecting events 18:13:13 <lamby> vagrantc, you were going to talk with spectranaut? How's that going? 18:13:30 <vagrantc> spectranaut: was there anything more you wanted to add? 18:13:41 <vagrantc> spectranaut: last i checked in with you you just said to write it up 18:13:43 <spectranaut> nope I'm not submitting! 18:13:53 <spectranaut> think I said.. 18:13:55 <vagrantc> i've got till tomorrow to do that 18:13:59 <lamby> Aw. 18:14:25 <lamby> Okaley, well nothing to co-ordinate, so… 18:14:31 <vagrantc> with all three of these, if anyone would be available for some review, that'd be great 18:14:31 <infinity0> ok done, switched to plain text 18:14:31 <lamby> Next. Scale 18:14:42 <lamby> vagrantc: yours ^ 18:14:43 <lamby> * ? 18:14:51 <vagrantc> scale is due *today* 18:14:51 <emaste> I will probably submit a Reproducible FreeBSD talk to AsiaBSDCon 2017 , Tokyo Mar 9-12 18:15:09 <vagrantc> scale: i've submitted a talk, but probably should fleh out the long description more 18:15:17 <vagrantc> hep greatly appreciated 18:15:19 <vagrantc> help 18:15:26 <emaste> I will probably submit a Reproducible FreeBSD talk to AsiaBSDCon 2017 , Tokyo Mar 9-12 18:15:28 * vagrantc can't even type it's so urgent 18:15:35 <lamby> I believe these are agenda items when they require liasing with people. Otherwise, just go ahead and submit! 18:15:43 <lamby> ie. is there anything to discuss about SCALE? 18:15:55 <vagrantc> i could really use a hand writing up a longer description 18:16:10 <lamby> vagrantc: Can you send a link, etc to the ML? 18:16:26 <vagrantc> maybe i'll start a pad or soemthing, and link to the list 18:16:30 <lamby> Cool 18:16:45 <lamby> #action vagrantc to send a link to SCALE (etc.) draft to the ML for help 18:16:48 <spectranaut> I'd look at it to, vagrant -- related I'm giving a talk to the local users group on reproducible builds tomorrow :) 18:16:52 <lamby> Same with ELC. 18:17:03 <spectranaut> local linux users group* 18:17:15 <vagrantc> sounds good 18:17:16 <lamby> Anything else needing discussion on submitting talk stuff? 18:17:28 <vagrantc> figured i'd let people know about the events i was looking at to coordinate 18:17:35 <lamby> *nod* 18:17:50 <lamby> Cool. 18:17:53 <HW42> vagrantc: maybe add them to the website or blog 18:18:16 <vagrantc> i should probably use the list to mention which talks i'm thinking about 18:18:16 <lamby> #topic Summit updates 18:18:24 <vagrantc> don't want to add to website/blog till they've been confirmed 18:18:27 <lamby> h01ger: fancy providing an update? ^ :) 18:18:43 <HW42> vagrantc: I meant the events, not you talks 18:18:50 <HW42> *your 18:19:03 <vagrantc> HW42: well, if there's nothing r-b related, why mention? 18:19:11 <HW42> point 18:19:23 <h01ger> there not really much news 18:19:48 <h01ger> if someone still wants to come, please do talk to me, we still have free seats 18:20:01 <lamby> Neat. Did you reping people who did not yet reply? 18:20:06 * h01ger looks much forward to the summit and see you (all!) there! ;-) 18:20:14 * vagrantc too 18:20:28 <lamby> Dhole - come to the summit :3 18:20:36 <HW42> h01ger: did you asked the Yocto people? 18:20:42 <neverpanic> HW42: See mail on the list 18:20:45 <h01ger> lamby: i'm tired. there not really much news. i need rest. now. 18:20:51 <h01ger> HW42: yes 18:20:58 * emaste has been completely swamped, but will come 18:21:02 <HW42> h01ger: nice 18:21:10 <HW42> neverpanic: the mail was the reason to ask 18:21:17 <lamby> h01ger: I'm not chastising you, merely curious. 18:21:31 <HW42> neverpanic: but yeah now I read that they talked to h01ger 18:21:39 <HW42> nvm 18:22:01 <lamby> Cool. 18:22:15 <lamby> #topic disorderfs 18:22:21 <lamby> (Who added this?) 18:22:34 <h01ger> its disabled because unmounting caused problems 18:22:36 <lamby> Background is that we disabled it a while ago as it was causing some FTBFS. We should probably just fix it. 18:22:47 <h01ger> which led to disk filled up and thus ftbfs 18:22:56 <deki> but then we no longer have 91% :> 18:23:06 * h01ger added this (FWIW) 18:23:13 <neverpanic> Is there a ticket descibing what the problem is? 18:23:22 <h01ger> neverpanic: conviniently no 18:23:41 <h01ger> my idea is to enable it on i386 only, so that the impact of the ftbfs will be low 18:23:49 <h01ger> patches doing that much welcome 18:23:54 <deki> on i386/armhf there are fileordering variations anyway 18:23:58 <deki> even without disorderfs 18:24:06 <lamby> deki: how so? 18:24:24 <deki> different kernels/filesystems or so 18:24:25 <h01ger> see jenkins.d.n.git/hosts/profitbricks-build2-i386/etc/pbuilder/rebuild-hooks/D01_modify_environment 18:24:30 <lamby> h01ger: Sounds like a plan. 18:24:45 <h01ger> deki: i believe only on armhf, or also on i386? 18:24:51 <deki> h01ger: also on i386 18:24:55 <h01ger> nice 18:25:04 <h01ger> ah 18:25:23 <lamby> In that case, it's tempting just turn it on everywhere… 18:25:39 <lamby> Simpler patch too; more likely to get done, etc. :3 18:25:40 <h01ger> armhf+i386 has /srv/workspace on ext3 while its on tmpfs on amd64 18:25:49 <deki> ah :) 18:25:55 <h01ger> lamby: no. amd64 builds too fast 18:26:08 <lamby> I blame profitbricks. 18:26:11 <h01ger> so better enable it on i386 and migate the impace^w^w^wfind a fix there 18:26:17 <h01ger> lamby: no funny 18:26:21 <lamby> Okay. 18:26:51 <lamby> I'd like to #action something here. "Someone" should write a patch isn't that great. 18:26:54 <h01ger> why do you blame nice people out of the blue for things they have nothing to do with? 18:27:07 <vagrantc> ext4 on armhf, fwiw 18:27:12 <lamby> h01ger: Eh?It was a joke… 18:27:23 <h01ger> lamby: sigh 18:27:38 <lamby> I don't understand, sorry. 18:27:40 <h01ger> people might actually read the log and not get that 18:27:48 <vagrantc> well, now it's explicit 18:27:57 <h01ger> yup 18:27:59 <h01ger> haha 18:28:05 <emaste> profitbricks is too good to us 18:28:32 * vagrantc guesses the joke was they're so good to us providing such fast builders that we have problems resulting from such fast builds 18:28:44 <h01ger> (enabling disorderfs is on my urgent todolist, so we can move on…) 18:29:07 <h01ger> vagrantc: oh, thanks. now i get the joke. (i said was tired) 18:29:11 <lamby> h01ger: Can I #action you on that? 18:29:17 <h01ger> no 18:29:27 <lamby> Okay, let's move on. 18:29:36 <lamby> #topic Logo 18:29:52 <lamby> I added this as a placeholder reminder to try and think of more logos and/or sources of more logos. 18:29:56 <lamby> That's all really. 18:30:08 <vagrantc> should we put out a call for logos? 18:30:32 <lamby> vagrantc: I have to various people. If there's a more general way of doing that, sure, go ahead. 18:30:33 <emaste> I think we should 18:30:37 <deki> has it been posted on twitter yet? 18:30:49 <lamby> deki: Good point 18:30:52 <h01ger> whats the logo url again? 18:30:58 <emaste> I posted it to twitter 18:31:01 <h01ger> https://wiki.debian.org/ReproducibleBuilds/Logo 18:31:22 <emaste> but, we should pick a closing date and make it an explicit call for logos 18:31:35 <lamby> emaste: That idea was rejected last meeting FYI 18:32:00 <lamby> Folks can edit the wiki and send out calls separate to this meeting, so unless we want to open the logo deadline discussion again (?). 18:32:18 * h01ger would really prefer we wait for a logo we like 18:32:24 <h01ger> instead of working with a deadline here 18:32:39 <vagrantc> it could be a soft deadline for proposals, even if we don't *decide* on one 18:32:45 <emaste> I thought it was a pre-summit deadline that was rejected, not a deadline in general 18:33:05 <vagrantc> i think submissions are more likely to come with a deadline, at least from the general public 18:33:22 <emaste> right 18:33:36 <emaste> it doesn't have to be a deadline for us to choose, just a deadline for us to solicit 18:33:43 <h01ger> and then we are forced to decide because we set a deadline 18:33:52 <lamby> Nah, we could always re-solicit. ) 18:33:55 <h01ger> and then we'll pick something bad because we said we would 18:34:23 <emaste> no, if nothing good gets submitted we don't pick any 18:34:28 <h01ger> we did well for 2y without a logo 18:34:45 <h01ger> emaste: i dont believe that will happen. some people will like crap and pick that. 18:34:50 <h01ger> and say, hey, its great 18:35:10 <vagrantc> and some people will reject perfectly good and useable logos and never pick anything ... 18:35:14 <h01ger> sorry to be an elitist artist here ;-p 18:35:16 <vagrantc> so, whatever 18:35:39 <emaste> indeed, so no deadline then. but let's be explicit that we're soliciting submissions 18:36:02 <lamby> emaste: explicit where? 18:36:15 <emaste> on the wiki 18:36:20 <lamby> cool, i'll add a big banner 18:36:35 * vagrantc proposes a deadline of 2092-12-31 18:36:38 <lamby> #action lamby to make it very explicit we're solititing submissions (w/no deadline) 18:36:40 <h01ger> or, we go with logo v0.1 and make it explicit that this is a first version of the logo ;) 18:36:54 * h01ger trying to be less of an elitist artist and more fun 18:37:11 <lamby> Let's move on. 18:37:13 <emaste> lamby: i.e., how could someone who's not a wiki.debian.org account holder submit a logo 18:37:14 <vagrantc> cheers 18:37:16 <emaste> yes 18:37:19 <lamby> #topic debrebuild 18:37:25 <vagrantc> emaste: get someone to proxy submit 18:37:26 <lamby> (Who added this?) 18:37:28 <HW42> I 18:37:31 <h01ger> emaste: mail it to us. i'll happily add it to the wiki… 18:37:45 <lamby> HW42: Cool, go ahead :) 18:37:53 <HW42> I added this, josch made a nice tool currently called debrebuild 18:38:07 <HW42> it's an improved version of srebuild 18:38:07 <emaste> h01ger: right, I mean let's mention that (emailing) on the wiki 18:38:14 <lamby> emaste: (ack, will do) 18:38:18 <h01ger> emaste: please go ahead 18:38:30 <HW42> currently it lives in the linked mail thread. 18:38:35 <h01ger> HW42: where's the git repo? is it in sid? ah. 18:38:38 <h01ger> HW42: link? 18:38:49 <h01ger> https://lists.alioth.debian.org/pipermail/reproducible-builds/Week-of-Mon-20161107/007577.html 18:39:11 <HW42> it was my suggestion to make it not sbuild specific 18:39:33 <HW42> but now we need either a Maintainer or josch will merge it back into sbuild 18:39:46 <h01ger> HW42: josch liked the suggestion (AIUI) but was not captable to work on it 18:39:52 <h01ger> HW42: how about you? ;) 18:40:31 <HW42> I certainly can help, but IIUC the main point was a Debian maintainer 18:40:44 <deki> can't it be added to devscripts or so? 18:40:57 <h01ger> HW42: how about you become one? or as a first stop, forward that mail to the bts, eg as a bug against devscripts? 18:41:07 <lamby> HW42: Why do we need a Debian maintainer specifically? To upload it…? 18:41:52 <HW42> so if I understand you correctly your are fine to put it in a git repo and we will later look who uploads a package? 18:41:53 <vagrantc> if someone needs upload rights, surely there's a handful of people in this group ablle to sponsor uploads :) 18:42:08 <HW42> IIUC the point was to get it fast into the Debain acrhive 18:42:20 <lamby> HW42: Yep. I'm very happy to upload/sponsor stuff… 18:42:33 <vagrantc> devscripts seems a plausible candidate, then 18:42:47 <HW42> ok, I will respond to joschs mail with this 18:42:52 <lamby> HW42: great, thanks! 18:42:53 <HW42> (after all he did the work) 18:42:54 <vagrantc> if you don't want an entirely separate package 18:43:00 <h01ger> HW42: as vagrantc said, i'd be more than glad to sponsor packages you say are good for the archive! (where sponsor means review _etc_ and then sign+upload ;) 18:43:22 <lamby> #action HW42 to respond to joschs mail re. sponsorship of a new package and/or adding to devscripts or similar 18:43:51 <lamby> Thanks, will be nice to see the ball rolling on that 18:44:09 <lamby> #topic separate mailing list reproducible-bugs@ for bugs? 18:44:20 <lamby> Someone want to claim this? :) 18:44:26 * h01ger would like that 18:44:42 <deki> for what bugs? 18:45:12 <vagrantc> x-debbugs-cc? maintainer (e.g. diffoscope)? 18:45:32 <lamby> It would be nice if whoever added this would make the case for it. Any takers? 18:46:04 <infinity0> oh this was me 18:46:12 <lamby> I'm guessing the issue is excessive x-debbugs-cc from patches/ftbfs is annoying. They could go to another list, presumably ending in -bugs. 18:46:18 <lamby> infinity0: hi. go ahead. 18:46:26 <infinity0> yeah, it's just making discussions take more time to follow and find 18:46:34 <infinity0> discussion threads, i mean 18:47:07 <lamby> Okay. Do you have a proposal? 18:47:08 <infinity0> easier to get an overview of the outstanding discussions if they're not split view-wise by 20 bugs to small packages in between them 18:47:12 <HW42> is there an easy way to also get the follow up mails to this ML 18:47:42 <infinity0> yes, we create a reproducible-bugs@lists.alioth.debian.org and put new bugs there. or i guess it's easier to move the discussions to another list (since the BTS already contains the current ML) 18:48:01 <h01ger> the "proposal" part of this topic is actually quite hard. the "yes, the idea is good" part is easy 18:48:13 <lamby> We can easily move the usertags to a new list, it would be harder to move the discussions. 18:48:20 <h01ger> reproducible-bugs@lists.alioth.debian.org sounds like a good name. 18:48:21 <deki> infinity0: you don't use a mail client with threading? :) 18:48:26 * h01ger nods lamby 18:48:40 <lamby> The problem is mostly the x-debbugs-cc; for that we just need to a) create the list and b) change the template(s) 18:48:48 <infinity0> i do but the overview list looks like "discussion A \n bug 1 \n bug 2 \n ... \n bug 20 \n discussion B" 18:48:55 <deki> ah ok 18:49:17 <h01ger> deki: while i agree in theory, in practice not everybody is good at filtering. our list is too full 18:49:23 <infinity0> it's not urgent, we can think about it some more. atm i am highlighting the "important" threads with local colours 18:49:36 <lamby> I think we all agree its an issue 18:49:39 <HW42> when we have moved this to r-bugs@ is there an easy way that the BTS sends also the follow-up mails 18:49:52 <HW42> I'm always wondering how many bugs get responses 18:50:07 <lamby> HW42: Followup mails go to the submitter anyway, so that's an orthogonal issue. 18:50:09 <deki> is there even an interest in the x-debbugs-cc? maybe if it's too noisy we can also just drop that 18:50:14 <h01ger> HW42: thats a deficiency of the bts, submitters are not automatically subscribed… 18:50:22 <h01ger> deki: i love them 18:50:27 <h01ger> (and read them) 18:50:34 <lamby> I'm very interested in x-debbugs-cc, but… yeah. Swamping the others. 18:50:54 <lamby> How about I action myself to a) create the list and b) change the template(s) to point x-debbugs-cc to that? 18:50:56 <h01ger> switching to reproducible-bugs@lists.alioth.debian.org seems like it could work? 18:51:05 <h01ger> lamby++ 18:51:10 <deki> sounds good 18:51:12 <lamby> That would appear to solve 99%, if not all of the issue. 18:51:12 <HW42> +1 18:51:19 <lamby> infinity0: what do you think? 18:51:24 <infinity0> cool, yeah thanks! 18:51:27 <h01ger> lamby: and c.) move all the bugs 18:51:37 <h01ger> and d.) send patches for j.d.n.git :) 18:51:41 <deki> why move the bugs? 18:51:42 <HW42> lamby: c) write a short mail, so people update their personal template 18:51:48 <infinity0> does that include the existing bugs? i could probably write a script for the user-tags, but not sure about the CC-headers 18:51:52 <h01ger> deki: the usertags 18:52:03 <deki> h01ger: can't we keep the usertags as they are? 18:52:05 <lamby> #action lamby to create new -bugs list, etc., update usertags/templates, etc. 18:52:20 <deki> usertags don't create mails 18:52:21 <lamby> Actually, yeah, no need to move usertags. 18:52:23 <h01ger> which reminds me, i have a related topic 18:52:33 <h01ger> wrong user tagged bugs 18:52:40 <lamby> h01ger: Please keep for "any other business" :) 18:52:41 <h01ger> lamby: topic please 18:52:48 <h01ger> it would fit perfectly 18:52:52 <h01ger> and its short 18:53:00 <lamby> (Sure, just 1 moment) 18:53:34 <lamby> Do we want to change the usertag? I'm not sure we do. Just want to change the -cc 18:53:41 <h01ger> current usertagged bugs are visible here: 18:53:47 * deki agrees with only changing -cc 18:54:00 <h01ger> lamby: right, we dont want to, i just see i was wrong. 18:54:03 <h01ger> https://udd.debian.org/cgi-bin/bts-usertags.cgi?user=reproducible-builds%40lists.alioth.debian.org 18:54:09 <h01ger> are all usertagged bugs 18:54:13 <lamby> deki: Cool, okay. Well, we can always move them later. 18:54:20 <h01ger> https://udd.debian.org/cgi-bin/bts-usertags.cgi?user=reproducible-builds%40lists.debian.org is a mistyped bug 18:54:27 <lamby> #action (lamby to not update usertags) 18:54:31 <h01ger> https://udd.debian.org/cgi-bin/bts-usertags.cgi?user=reproducible-builds%40alioth.debian.org are three more mistyped bugs 18:54:39 <h01ger> https://udd.debian.org/cgi-bin/bts-usertags.cgi?user=Reproducible-builds%40lists.alioth.debian.org another one 18:54:41 <lamby> #topic wrong user tagged bugs 18:54:44 <lamby> h01ger: go ahead 18:54:47 <h01ger> can some one please clean that up= 18:54:48 <h01ger> ? 18:55:10 <deki> i will do it 18:55:11 * h01ger is already done with this little subtopic and closes 4 tabs here and says "thank you", whoever picks this 18:55:13 <h01ger> deki: danke 18:55:29 <lamby> #action deki to clean up usertags 18:55:46 <h01ger> https://udd.debian.org/cgi-bin/bts-usertags.cgi?browse=users is awesome, btw 18:56:02 <lamby> Any other business? Mention it and I'll #topic it up… 18:56:37 <h01ger> #topic aob 18:57:05 <h01ger> lamby: ^ 18:57:05 <vagrantc> looks like meetbot only respects lamby 18:57:23 <lamby> Well, seems like there is no AOB so let's just call it a night :) 18:57:26 <lamby> #endmeeting