17:00:39 <mattdm> #startmeeting
17:00:39 <zodbot> Meeting started Mon Jun  9 17:00:39 2014 UTC.  The chair is mattdm. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:39 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:00:48 <mattdm> #meetingname board
17:00:48 <zodbot> The meeting name has been set to 'board'
17:00:55 <jwb> hi
17:01:04 <mattdm> #meetingtopic Board and new FPL townhall
17:01:11 <mattdm> #topic Hello!
17:01:13 <mattdm> hi everyone!
17:01:16 <mattdm> hi jwb
17:01:17 <croberts> hi
17:01:20 <Sparks> greetings
17:01:26 <gholms> :D
17:01:34 * nirik is lurking around in the back
17:01:50 <cwickert> mattdm: board meeting?
17:01:55 <cwickert> oops
17:01:58 <mattdm> cwickert indeed :)
17:02:06 * cwickert forgot to scroll down ;)
17:02:22 <mattdm> My idea here is to make this first meeting not very formal -- just open floor and chat.
17:02:46 <mattdm> We've got plenty to do, but I figured it'd be good to start this way.
17:02:57 <mattdm> #topic Open floor
17:03:09 <mattdm> so we might as well get started :)
17:03:23 <mattdm> Hi everyone. I'm Matthew Miller, ye olde new FPL
17:03:27 <mattdm> (or something)
17:03:43 <mattdm> Anyone have anything they'd like to talk about?
17:04:10 <gholms> #info Greetings to Matthew Miller, the new FPL
17:04:17 <jwb> is there anyone here besides the board members?
17:04:21 <croberts> Is the weather nice in Boston?
17:04:25 <mattdm> jwb a very good question
17:04:28 * jds2001 is here :)
17:04:30 <croberts> I am jwb
17:04:33 <gholms> jwb: Well, we have nirik
17:04:41 * nirik waves
17:05:13 <mattdm> croberts: it's traditional boston weather -- we had two weeks of beautiful spring and now it's too hot :)
17:05:19 <jwb> gholms, he's ever present.  he doesn't count ;)
17:05:42 <gholms> mattdm: What got you interested in the FPL hat?  It's quite... large.  :)
17:06:26 * quaid also is lurking
17:06:40 <gholms> jwb: Everywhere and nowhere at the same time ;)
17:06:42 <number80> .hellomynameis hguemar
17:06:43 <zodbot> number80: hguemar 'Haïkel Guémar' <karlthered@gmail.com>
17:06:51 <mattdm> So, I've been working on this fedora.next stuff for a while, first kind of from a cloud sig perspective, and then fesco
17:07:16 <mattdm> and, the idea all along was really that it should be something that involves the whole project
17:07:27 <mattdm> which goes beyond the technical areas overseen by fesco
17:07:44 <mattdm> and beyond the fedora release cycle that fesco is basically tasked with
17:08:17 <mattdm> so, when robyn said she was ready to move on, it seemed like it might be the next logical step
17:08:18 <handsome_pirate> !
17:08:46 <mattdm> and I've always been interested in everything across the project and have been following it broadly, so I thought I might try EVEN MORE of that :)
17:08:48 <mattdm> handsome_pirate: go
17:09:16 <handsome_pirate> I just wanted to say congrats, mattdm
17:09:27 <mattdm> handsome_pirate: thanks!
17:09:29 <handsome_pirate> EOF
17:10:00 <mattdm> any questions or comments from the QA perspective?
17:10:15 <mattdm> that's an area where I mostly have sat on the sidelines and watched you all be awesome
17:10:23 <mattdm> with the occasional +1 karma to something :)
17:10:35 <handsome_pirate> !
17:11:17 <mattdm> handsome_pirate: go
17:11:42 <mattdm> also I think we have few enough people that you can just go ahead and we don't need to follow the formal hand-raising protocol for this meeting
17:12:00 <handsome_pirate> mattdm:  Roger
17:12:20 <handsome_pirate> So, all I can say right now is, QA needs more peeps with all the fedora.next stuff
17:12:27 <handsome_pirate> How do you plan to address that?
17:13:03 <jwb> the WGs are going to have to step up and help out
17:13:13 * number80 suggests to clone handsome_pirate for that
17:13:20 <mattdm> I think there a lot of areas where we could use more people. I *hope* that we can make the fedora.next framework a way that makes it easier to get people involved.
17:13:34 <handsome_pirate> number80:  Hrm, that would be scary.
17:13:39 <mattdm> because particularly, yeah, we can funnel them in from the workgroups and the associated sigs
17:14:15 <gholms> mattdm: How does that differ from before?
17:14:55 <mattdm> gholms I think before, a lot of the different sigs didn't necessarily see the QA as their responsibility. if you wanted QA, go over to the QA team.
17:15:10 <jwb> gholms, there were no WGs, there was only one main thing QA focused on, and even then QA mostly did install testing
17:15:55 <mattdm> and I think as we get the broader test plans in place (thanks very much to the qa people working on this), it'll be easier for people to find a place to contribute
17:16:07 <mattdm> the whole big thing is intimidating
17:16:37 <jds2001> if you point people (me) at what needs doing, I could step up my involvemnt I think. However, free time is at something of a premium for me (the story of the last few years...)
17:17:06 <mattdm> From the cloud perspective, we're going to have automatic upload of new nightly images to cloud providers hooked up soon, and after that, wired into taskotron
17:17:12 <jds2001> but I'm just one guy....
17:17:12 <mattdm> then people can make small tests and drop them in place
17:17:22 <mattdm> jds2001: interested in reviving bugzappers? :)
17:17:26 <jwb> jds2001, one guy is one more guy than no guy
17:18:38 <jds2001> mattdm: bugzappers is a big job :)
17:18:58 <adamw> ahhhhhh bugzappers
17:19:03 <jds2001> one could literally spend all day doing nothing else :)
17:19:12 <number80> jds2001: you're looking at something like mentoring ?
17:19:26 <jds2001> number80: that was one idea.
17:19:40 <mattdm> I think one area where I've seen a few sparks of interest lately is the Join SIG
17:19:41 <number80> just join the merry QA team meetings and ask for what you can do ;)
17:20:09 <adamw> we do have a reasonably good (we think) join page for QA with quite a few pointers - https://fedoraproject.org/wiki/QA/Join
17:20:14 <gholms> adamw: You highlight on that word, don't you?  :P
17:20:17 <mattdm> A lot of people have really this same question: how can I easily plug into Fedora to make some small contributions
17:20:21 <adamw> gholms: no, just happened to drop by
17:20:46 <adamw> but there are things we need more experienced folks to do to help with that. it would be nice if more maintainers of key bits provided test cases for their bits
17:21:18 <mattdm> adamw: +1 -- I think we need to promote that more.
17:21:20 <adamw> we can help with the implementation of that, but it can get a bit discouraging sitting down as a qa person to write a test plan for some component and more or less pulling it out of your rear end
17:21:26 <adamw> it helps to have...help
17:21:52 <gholms> #help wanted for for writing test plans
17:21:53 <number80> adamw: QA hackfest at Flock ? I'd help for cloud bits
17:22:01 <adamw> number80: solid idea
17:22:14 <mattdm> I like the idea of putting test plans next to packages somehow.
17:22:15 <gholms> #idea QA hackfest at Flock
17:22:22 <jds2001> adamw: i'll be there :)
17:22:33 <gholms> #idea Add test plans to packages... somehow
17:22:37 <mattdm> next to the git repo with taskotron tests
17:22:42 <jds2001> adamw: barring last minute injuries requiring major surgery :D
17:22:50 <adamw> gholms: we have a mechanism for it at present. it's a silly mechanism, but it's there. https://fedoraproject.org/wiki/QA:SOP_package_test_plan_creation
17:22:57 * mattdm knocks on wood, throws salt over shoulder, etc
17:22:57 <jwb> gholms, i like the flock idea, but i worry we'll sit around and wait for that.  we need stuff sooner.
17:23:09 <adamw> (basically, the test cases go in a mediawiki category with a name in a set format. bodhi is the thing that consumes that at present, but there's no law against anything else doing it.)
17:23:19 <gholms> jwb: Indeed.
17:23:51 <mattdm> https://lists.fedoraproject.org/pipermail/devel/2014-January/194856.html
17:23:54 <number80> jwb: at least, if nothing is done until august, we can kickstart it there
17:24:00 <mattdm> ^ conversation dejavu
17:24:19 <jwb> number80, heh, yes.
17:24:24 <mattdm> wow I can't belive that was back in january
17:24:31 <adamw> =)
17:24:51 <gholms> jwb: If something does happen before then, it'll still be great to have lots of people working on it there.  ;)
17:25:08 <jwb> yes, absolutely
17:26:47 <mattdm> I am totally at risk of going into a long bunch of questions about how the tasktron per-package test location design ended up (or at least where it is sitting)
17:26:59 <mattdm> but instead maybe I'll make a note of that and look at it later :)
17:27:19 <gholms> Heh heh
17:27:23 <tflink> per-package test location design?
17:27:40 <mattdm> yeah there was talk about git submodules....
17:27:51 <mattdm> (see, here we go)
17:27:56 <mattdm> :)
17:28:00 * yn1v is lurking
17:28:02 <gholms> D:
17:28:07 * cwickert listens up
17:28:08 <tflink> oh, I think I know what you're talking about
17:28:16 <jwb> we might want to see if others have other topics first
17:28:18 * jds2001 thinks it would be wonderful for maintainers to submit tests which can be automatically run as a result of a build (maybe via fedmsg?)
17:28:25 <adamw> the Package release criteria / validation testing stuff i'm working on atm should give us at least a reasonable set of (manual, for now) testing that covers the core functions of each product. i'm just hoping it doesn't wind up producing a reaction of 'oh god that's way more than we realized.'
17:28:38 <adamw> jds2001: that is the plan for taskotron
17:28:55 <jwb> fwiw, the kernel is already doing this
17:29:08 <mattdm> adamw: that reaction might not be terribly bad as long as we can get positive results in the aftermath
17:29:09 <gholms> adamw: At this point is it even worth writing stuff for autoqa?
17:29:13 <adamw> jds2001: but the *first* plan for taskotron is just to re-implement autoqa (only better), and that's all we're committed to having in place for f21 alpha; we do need to be moving on manual testing as well, and we will always need some degree of manual testing as well as auto
17:29:20 <adamw> gholms:  no, not at all. autoqa is dead.
17:29:30 <adamw> (long live taskotron!0
17:29:39 <gholms> :)
17:30:04 * cwickert wonders if he is online
17:30:07 * cwickert is
17:30:09 <jwb> cwickert, you are
17:30:12 <jds2001> cwickert: you are :)
17:30:18 <tflink> gholms: the current plan is to decommission autoqa production in early July. staging is already in deep hibernation :)
17:31:06 * gholms clearly needs to read the QA lists more
17:31:32 <tflink> I'm not sure the details on that have been communicated well yet - I was waiting for stuff that we figured out at the bodhi2/taskotron FAD last week
17:31:45 <mattdm> so several lessons here: 1) more qa help needed, 2) taskotron is awesome, and 3) qa people are definitely involved and visible and show up at meetings. :)
17:32:26 <mattdm> Are there some non-QA topics people would like to talk about?
17:32:38 * gholms raises hand
17:32:40 <adamw> NO LET US TALK ABOUT QA ALL THE TIME
17:32:43 <mattdm> gholms: go
17:32:45 <gholms> Heh
17:32:55 <mattdm> adamw: it's an _option_, just trying to make some space :)
17:32:59 * jds2001 wants to talk about bacon! :D
17:33:09 <gholms> What should the board be doing to help with fedora.next that it hasn't been doing?
17:33:48 <gholms> jds2001: Sounds like a good thing to close the meeting with.  :)
17:34:13 <cwickert> ?
17:34:29 <gholms> I'm mainly interested in hearing what people think *ought* to be happening, what's stalled, etc.
17:35:00 <gholms> -> cwickert
17:35:11 <cwickert> What is the current scope of Fedora.next? I mean we had basically two different proposals, one was "streams" the other rings.
17:35:25 <cwickert> so far it looks like we only do streams a.k.a. products
17:35:42 <adamw> the 'base' and 'env and stacks' are the sort-of-ring-y bits, i think.
17:35:52 <jwb> theoretically base is doing some of the ring concept.
17:36:02 <jwb> in practice, there hasn't been a whole lot yet
17:36:02 <cwickert> but what about the idea of the rings with different update and release criteria? are we still planing this for the future?
17:36:20 <handsome_pirate> cwickert:  I would severly dislike that
17:36:27 <handsome_pirate> cwickert:  Lots of extra burden on QA
17:36:31 <jds2001> handsome_pirate: why?
17:36:44 <jds2001> handsome_pirate: only for the inner rings, I would think.
17:36:46 <handsome_pirate> Maintaining different sets of criteria
17:36:50 <mattdm> cwickert: to me, Fedora.next is a pretty big umbrella term for looking ahead and adjusting our steps accordingly. So, I use it to include a lot of things.
17:37:02 <mattdm> Including for example delaying the F21 release to give time for Taskotron work
17:37:50 <cwickert> I know we are not finished by F21, but what more can we expect from Fedora.next?
17:38:12 <mattdm> And yeah, the base and env-and-stacks are both part of the "ring" concept and as of yet not particularly ambitious
17:38:31 <mattdm> I'd like to see more from both of those, because I still think they're very important to a more flexible future
17:39:09 <jonar> .fas  jfarcher
17:39:10 <zodbot> jonar: jfarcher 'Jonathan Archer' <jon@rosslug.org.uk>
17:39:15 <mattdm> And I'm going to be working more with the people involved in them to help that happen... that's part of broadening my perspective from the cloud sig outward
17:39:48 <jwb> i think, in part, the lack of output from the base group is driven by the lack of differentiation from the products
17:40:05 <jwb> everything is using the same core set of things still, all on the same schedule
17:40:21 <jwb> if fedora.next migrates away from that, then Base will need to start thinking hard about what that means, etc
17:40:51 <number80> It's fine, if the first release of Fedora.Next is not as ambitious as we'd like it to be
17:41:10 <quaid> adamw: for developing new QA tools and processes, I wonder if there is potential synergy with CentOS Project? While existing QA there has been all around the distro, we're requiring SIGs to do their own QA too (as per how much their community needs), and lots of the tooling needs are likely to be similar to Fedora's (cf. we're going to use Koji for the new community builders the SIGs use, for example)
17:41:13 <number80> it's quite a change, and we need to make it successful, then improve of what we learned
17:41:15 <cwickert> ok, then lets revisit this at FLOCK or after the release
17:41:21 <jds2001> jwb: not sure why - let me try and explain. Base as it exists today would continue serving it's existing function.
17:41:27 * cwickert does not like the idea of different release cycles
17:41:38 <mattdm> I'd also like to see some focus on quality improvments in the base for its own sake.
17:41:44 <cwickert> ´1
17:41:44 <adamw> quaid: i have absolutely no idea how centos works at all, but i'm not at all opposed to the idea...
17:41:46 <cwickert> +1
17:41:48 * quaid thinks most of the relevant people from CentOS will be at Flock as well
17:41:48 <mattdm> number80 +1
17:42:14 <jds2001> jwb: but nothing says (I don't think) that another group can't start up, following the same release criteria, etc, and provide some other base.
17:42:16 <mattdm> quaid I'm definitely for collaboration where we can benefit from shared tools and processes
17:42:20 <jwb> jds2001, base as it exists today is ~2000 packages.  nobody in Base actually wants it to be that large, but there's no major impetus to reduce it other than "just because"
17:42:37 <jwb> jds2001, so when you have a base of most of the major functionality ... :)
17:43:01 <jds2001> jwb: i was thinking of base as something much lower level
17:43:20 <jds2001> jwb: kernel/glibc/systemd and friends.
17:43:36 <jwb> jds2001, so is most of the base wg, but on the technical side that isn't really feasible because of package dependencies and buildrequires
17:44:55 <jwb> quaid, it would be a good discussion to have between the two projects i think
17:44:56 <mattdm> jwb I think, ultimately, the benefit is that it's less expensive to maintain -- including qa
17:44:58 <jds2001> jwb: building on top of base1 vs. base2 should work, no? Provided that you have the right provides in place, of course.
17:45:14 <jds2001> it's quite a mess, but technically *feasible*
17:45:24 <jwb> jds2001, sure, but right now i'm saying base 1 is so large there's no room for base 2
17:45:29 <jwb> or no need
17:45:32 <jds2001> whether it's what we *want* remains to be seen
17:45:49 <jwb> mattdm, sure.  like i said, there's benefit but no _impetus_
17:46:09 <mattdm> And that if we draw a small circle of packages, we can make changes/improvements across that set in a unified way
17:46:12 <jwb> mattdm, "oh.  server and workstation are all going to ship most of this anyway?  well..."
17:46:21 <mattdm> jwb and not glamorous. yeah.
17:46:26 <jwb> mattdm, how is that no critpath?
17:46:29 <jwb> s/no/not
17:46:53 <jwb> answer: it's not critpath because of self-hosting, but still the idea already exists
17:47:22 <mattdm> jwb it is defintely a close relative of critpath
17:47:39 <jwb> mattdm, oddly enough, i see Cloud size reduction efforts as possibly the biggest driver here for base.  except that excludes installer, which the base WG considered important
17:47:58 <mattdm> but mostly critpath is used for testing, not things like, say, difference in package quality guidelines
17:48:31 <jwb> which is a shame
17:49:06 <mattdm> so, the *idea* is that the base can be a framework to make it not a shame
17:49:09 <jwb> however... this is all very good discussion, but it's very technical in nature.  at a higher level, i'd like to see the board reaching out and coordinating among the various groups in a more active manner
17:49:23 <number80> jwb: +1
17:49:27 <mattdm> but it does, as you say, need that impetus
17:49:36 <jwb> marketing and ambassadors for promotion.  branding discussions with all the WGs.  etc.
17:50:04 <mattdm> jwb yes. in fact, this is one of my first thoughts in response to gholms' question back there
17:50:24 <mattdm> It's pretty exciting to see the qa and docs and design stuff come together around fedora.next
17:50:25 <jwb> right.  just trying to steer us back out of "fesco/qa" territory here :)
17:50:35 <gholms> :)
17:50:39 <mattdm> jwb +1 yes thanks
17:50:49 <cwickert> jwb: :)
17:51:03 <mattdm> but we also need to find out what help marketing and ambassadors need so they can be involved
17:51:20 <jwb> sidebar though: i think it would be WONDERFUL to have fesco, qa, ambassadors, etc show up in a board meeting once a month or so and give us updates.
17:51:25 <mattdm> and other than the fudcons at BU a decade ago and the fedora magazine now, those aren't areas I've been very involved with
17:51:44 <mattdm> jwb yes! I think we should probably schedule that, in fact
17:52:00 <mattdm> rather than just having meetings be reactions to tickets
17:52:06 <jwb> indeed.
17:52:11 <gholms> #idea Involve FESCo/QA/ambassadors/etc in board meetings to help communicate/unblock/etc
17:52:50 <number80> +1 for a monthly state of fedora
17:53:18 <mattdm> I know a lot of the other board members *are* very active with ambassadors, so I'm looking for a lot of help in that area
17:53:24 <jwb> yes
17:53:55 <yn1v> state of fedora, I like the meeting name, sounds good
17:54:02 <gholms> mattdm: Can you help out on the FESCo side?
17:54:11 <mattdm> gholms absolutely!
17:54:16 <jwb> speaking of that...
17:54:28 <jwb> mattdm, so you're planning on finishing your FESCo term?
17:55:30 <gholms> Hehe
17:56:24 * yn1v needs to leave
17:56:30 <gholms> That's another interesting question:  between FESCo, the cloud WG, and the board, what will you drop on the floor first if you have to?
17:56:31 <number80> yn1v: bye !
17:56:53 <mattdm> jwb Yes, at least this term. I'll evaluate at the next election and see how I feel about running for that again. :)
17:56:57 <jwb> gholms, yeah.  my question was an obtuse way of asking "how's that time management thing working for you"
17:57:06 <mattdm> gholms: also, my family is on that list :)
17:57:08 <jwb> mattdm, with all due respect, you are crazy.
17:57:21 <jwb> maybe like a fox, but still crazy ;)
17:57:22 <gholms> mattdm: You're insane.  :P
17:57:24 <jds2001> mattdm: im afraid you'll burn out
17:57:34 <mattdm> but, seriously, I think the cloud WG is the part that I'm going to dial back on the most significantly.
17:58:00 <mattdm> David Gay (oddshocks) is on as an intern this summer working on some of the things like cloud image uploads and will be picking up some of the other slack
17:58:30 <mattdm> and the community people (including *cough* some in this room) have really picked up to the point where I feel like I'm not really needed so much (in a good way!)
17:58:34 <gholms> mattdm: rbergeron started by axing sleep and then dialed back on cloud, too.  :)
17:58:37 <number80> :)
17:58:48 <mattdm> *and* Red Hat will be posting a new cloud engineer position to fill in my space
17:59:02 <mattdm> gholms: hah. yes.
17:59:26 * cwickert needs to leave now
17:59:29 <gholms> mattdm: That isn't bad; if anything it means you probably aren't even crazier.
17:59:41 <gholms> I need to leave in a couple minutes.
17:59:46 <mattdm> And, in seriousness, if FESCo starts to feel like it's too overwhelming, I'll step back from that too.
17:59:55 <mattdm> We are at the hour mark
18:00:23 <number80> so the next public board meeting will be in two weeks ?
18:00:25 <mattdm> next meeting: more _business_ :)
18:00:30 <cwickert> :)
18:00:32 <mattdm> number80 yeah. same place, same time.
18:00:41 <number80> great
18:00:49 <cwickert> thanks everybody for their ideas and comments! bye
18:01:08 <number80> cwickert: bye !
18:01:14 <mattdm> And I'm hoping we can get some stuff done in the meantime, of course. :)
18:01:30 <mattdm> So, I'm gonna end the meeting
18:01:41 <gholms> #info Next meeting is in two weeks at the same time/place
18:01:45 <mattdm> but I will be around in the various channels for more discussion
18:01:50 <mattdm> thanks gholms for all the infos :)
18:01:59 <mattdm> and thanks everyone for questions and discussion
18:02:00 * gholms has one more
18:02:25 <gholms> #info Don't be afraid to join #fedora-advisory-board and chat, too
18:02:26 <mattdm> gholms: ok go
18:02:33 <gholms> Thanks.  :)
18:02:39 <mattdm> oh also:
18:03:02 <mattdm> #info request submitted to create an open board trac https://fedorahosted.org/fedora-infrastructure/ticket/4405
18:03:20 <mattdm> #info and to rename the mailing list and irc channel to not be confusing
18:04:41 <randomuser> mattdm, I really appreciate the things like this meeting that you're doing to make Fedora governance more approachable. Thanks.
18:04:53 <adamw> +1
18:05:02 <mattdm> :) thanks
18:05:12 * mattdm still waiting on gholms to type
18:05:44 <jds2001> he's writing a novel, I'm sure :D
18:06:08 <jwb> i think he already did his last info?
18:06:27 <jwb> it was right after he said he had one more...
18:06:27 <mattdm> oh one more info, not one more question
18:06:27 <mattdm> ahahaha.
18:06:39 * mattdm has awesome reading comprehension skillz
18:07:09 <mattdm> in that case....
18:07:14 <mattdm> #endmeeting