15:00:56 <pknirsch> #startmeeting Fedora Base Design Working Group (2014-08-01)
15:00:56 <zodbot> Meeting started Fri Aug  1 15:00:56 2014 UTC.  The chair is pknirsch. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:56 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:01:10 <pknirsch> #meetingname  Fedora Base Design Working Group
15:01:10 <zodbot> The meeting name has been set to 'fedora_base_design_working_group'
15:01:18 <pknirsch> Hello and welcome everyone!
15:01:41 <pknirsch> #chair dgilmore haraldh jreznik masta
15:01:41 <zodbot> Current chairs: dgilmore haraldh jreznik masta pknirsch
15:02:06 <pknirsch> who we got today?
15:02:16 * jreznik is here
15:02:21 <haraldh> <-
15:02:43 <moben> <--
15:03:45 <pknirsch> vpavlin and msekleta already mentioned they couldn't be here today, so lets get started
15:03:51 <pknirsch> first up, news from the front!
15:04:02 <pknirsch> #topic Moar status update builrequires cleanup
15:04:31 <moben> I've mostly been working on reproducible builds
15:05:06 <moben> as the whole "rebuild a few times to find superfluous buildreqs" idea only works if we can get those
15:05:46 <moben> the good news: about 85 - 90% of the tentative base package list do reproduce fine
15:06:28 <moben> I've been writing some upstream patches here and there
15:07:41 <moben> I'm not sure yet how to proceed with dead/stubborn upstreams
15:08:01 * dgilmore is here
15:08:12 <jreznik> 85-90 is a good achievement
15:08:27 <pknirsch> yea
15:08:37 <moben> would we carry patches in our (e.g. doc-generating) packages or fix things up during rpm build like for .jars?
15:08:37 <pknirsch> dead/stubborn upstreams are typically the hardest
15:09:19 <moben> would that be a fpc decision or a case-by-case thing if maintainers want to carry a patch for this?
15:09:47 <moben> (assuming I can convince fpc to put reproducible builds in the guidelines)
15:10:12 <pknirsch> it's really up to the maintainers at the end of they day. In the past we've had that very lax actually, but it's of course discouraged as patches should always go upstream when possible
15:12:03 <masta> yo!
15:12:36 <pknirsch> heya
15:13:00 <moben> pknirsch: obviously upstream first is always the priority here :)
15:13:07 <pknirsch> yar
15:14:51 <moben> That was the shortlog from me, I guess
15:15:13 <pknirsch> nice, thanks moben !
15:15:19 <moben> oh, I also hit some that don't do timestamps but other indeterminism
15:15:32 <moben> which is even more annoying
15:17:59 <pknirsch> ugh
15:18:01 <pknirsch> not nice
15:18:32 <moben> yeah. guile generates unique identifiers for macros or whatnot from urandom
15:18:36 <pknirsch> and true, without reproducible builds doing BR removals is going to less useful :)
15:18:42 <pknirsch> O.o
15:18:46 <pknirsch> le f**k
15:19:02 <moben> though I've been in touch with upstream there and they're going to use a hash of the source in the future
15:20:13 <pknirsch> :)
15:20:43 <pknirsch> but that sounds really good. Have you done any tests yet with BR removals for packages where you know you can do reproducible builds?
15:22:13 <moben> just pass '-d' to my script and it will try to reproduce a build with all deps first and error out otherwise :)
15:22:21 <pknirsch> ah! :)
15:22:29 <pknirsch> clever
15:22:38 <moben> maybe that should be the default...
15:23:04 <pknirsch> probably, as that would automatically skip all the packages where you already know that all builds will be different :)
15:23:55 <moben> due to the its long runtime if've not produced many actual removals yet, as I'd rather like to get it working on a larger set of packages first
15:24:28 * pknirsch nods
15:24:31 <pknirsch> sure thing
15:24:43 <moben> then we can just throw more resources at it later :)
15:25:33 <pknirsch> heheh
15:26:25 <pknirsch> alright, lets move on to Flock
15:26:46 <pknirsch> #topic Flock planing/meetup
15:27:24 <pknirsch> mattdm: you around? You've already sent out a quick email to the devel list, anything else to comment on that?
15:27:28 <dgilmore> we should sitdown and be productive in person
15:27:33 <masta> my regrets - had to cancel Flock attendance this year. I guess see you guys next year
15:27:42 <pknirsch> masta: ye, no worries
15:27:43 <mattdm> pknirsch I'm around, yeah. :)
15:27:50 <pknirsch> dgilmore: aye
15:28:23 <mattdm> the devel list message was about http://flock2014.sched.org/event/57c3681879f07c10dff11745aab6f21b
15:28:29 <mattdm> the fedora.next joint session
15:28:46 <mattdm> at devconf earlier this year, pknirsch presented on Base
15:28:58 * pknirsch nods
15:29:02 <mattdm> wondering if phil, you want to do that again, or someone else.
15:30:19 <pknirsch> mattdm: absolutely!
15:30:33 <pknirsch> mattdm: as long as it's not conflicting with my secondary arch talk ofc :)
15:30:47 <mattdm> pknirsch it's a joint session, so it Conflicts With Nothing
15:31:03 <mattdm> (say that in an echoing, ominous voice)
15:31:10 <pknirsch> O.o
15:31:20 <dgilmore> pknirsch: thanks :)
15:31:31 <pknirsch> dgilmore: np :)
15:34:20 <masta> somebody setup a google hangout or something, I'll jump into the session from remote.
15:34:43 <dgilmore> masta: they had hangouts for all the sessions last year
15:34:51 <dgilmore> so it should be happening i think
15:35:13 <pknirsch> nice :)
15:35:18 <jreznik> btw. env & stack guys plan dinner
15:35:39 <jreznik> masta: we are going to have streams
15:36:38 <pknirsch> jreznik: does that mean we can crash that party? ;)))
15:36:40 <pknirsch> ups
15:36:41 <pknirsch> :)
15:37:08 <jreznik> eat all the food they would have
15:37:26 <pknirsch> :)))
15:37:54 <pknirsch> alright, but yea, i'd be happy to do that part for Base again of course
15:41:44 <pknirsch> OK, dgilmore, or mattdm, do you know if there are rooms available for meeting sessions for SIGs/WGs?
15:42:09 <mattdm> pknirsch: yes, there are supposed to be.
15:42:15 <pknirsch> nice, good!
15:43:50 <pknirsch> alright, thats it for Flock then. last up then,
15:43:54 <pknirsch> #topic Open Floor
15:44:48 <pknirsch> Anything else we need to discuss today?
15:48:52 <pknirsch> Doesn't look like it, so lets wrap up for today
15:48:56 <pknirsch> and see you all next week!
15:49:02 <pknirsch> (or most of you ;))
15:49:15 <pknirsch> EoM, have a great weekend everyone!
15:49:23 <pknirsch> #endofmeeting
15:50:13 <pknirsch> #endmeeting