16:03:21 <stickster> #startmeeting
16:03:21 <zodbot> Meeting started Thu Oct  1 16:03:21 2009 UTC.  The chair is stickster. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:03:21 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:03:35 <stickster> #info This meeting supersedes the 2009-10-01-16.00 meeting.
16:03:37 <stickster> #topic Roll call
16:03:43 * notting is here
16:03:44 <stickster> Sorry guys, let's do another roll call real quick like.
16:03:45 * stickster 
16:03:46 * mmcgrath here
16:03:57 * spot is here
16:04:01 * poelcat here
16:04:24 * glezos here
16:04:42 * jwb is still here
16:04:54 <stickster> We appear to be missing caillon, mdomsch, dgilmore
16:05:18 * stickster did invites already. Let's proceed so as not to waste time.
16:05:38 * mmcgrath attempts to summon a gilmore
16:05:39 <stickster> We've reserved until UTC 1635 to talk about target audience and related issues.
16:06:26 <stickster> poelcat was going to put together a chronology of our topics around better defining the project and the distro.
16:06:47 * poelcat hopes it wasn't due today? :)
16:06:50 <stickster> I think he may still be working on that, but that doesn't mean we can't have a productive discussion here.
16:06:54 <stickster> poelcat: Nope
16:07:32 <poelcat> i'm not sure how much the chronology will help
16:07:55 <poelcat> i do think we've come to some recent focus on f-a-b in the discussions there
16:08:05 <stickster> I think this comes down to a couple basic questions
16:08:11 * mdomsch is better late than never
16:08:17 <stickster> I'll try to state them here, and encourage people to fix them if they're imprecise
16:09:17 <stickster> What is the target audience for whom Fedora contributors such as developers, packagers, etc. should make decisions about how they produce material for the distro?
16:09:20 <stickster> and
16:09:49 <stickster> Do our policies, processes, etc. support that audience?
16:10:08 * stickster looks for people to add to or fix these questions...
16:10:11 * poelcat was thinking of it slightly different... who is target audience for
16:10:17 <poelcat> 1) fedora project
16:10:20 <poelcat> 2) fedora distro
16:10:43 * mmcgrath agrees with seth from f-a-b that 1) really isn't in contention
16:11:23 <poelcat> okay
16:11:42 <poelcat> i just wanted to be sure there weren't cases where both audiences would not match 1 & 2
16:11:47 <jwb> hm, i forget what seth said
16:11:50 * jwb goes to look it up
16:12:26 <jwb> oh.  heh.  he just said it wasn't in contention
16:12:54 <poelcat> do we have a clear statement on #1 then?
16:13:01 <poelcat> and if so where is it?
16:13:15 <notting> stickster: i think the answer to your second question is 'no', regardless of the answer to your first
16:13:24 * glezos looks too
16:13:47 * stickster notes that poelcat recently took a stab at updating the Overview in this regard.
16:14:09 <mmcgrath> slowest meeting ever :)
16:14:26 <glezos> heh
16:14:26 <stickster> To me, the target audience of the Fedora Project is the set of people who agree with our mission and core values, and want to contribute to FOSS.
16:15:08 <mmcgrath> stickster: I tend to agree, it seems that if someone was looking to get involved with Fedora, at that level, it's clear what they're interested in and getting involved with.
16:15:18 <stickster> I believe that is both what we are now, and what we want to be in terms of the *project*.
16:15:18 <mmcgrath> because the Fedora Project itself has values, etc.
16:15:24 <poelcat> that makes sense, but I wonder if the target audience for the distro is a subset of those people
16:15:46 <notting> poelcat: i think it's the other way around
16:15:51 <mdomsch> stickster, "our users are already the converted"
16:15:54 <mdomsch> ?
16:16:00 <jwb> i don't think so
16:16:01 <poelcat> notting: how so?
16:16:04 <caillon> stickster, so ambassadors don't count?  they may not want to contribute directly, but spread it around so others can.
16:16:07 <stickster> poelcat: I don't think so -- if anything, I think that where we want to be in the future is that the *distro* target is a superset of the Project audience.
16:16:08 <notting> the audience for fedora-the-distro could be a *superset* of the audience of fedora-the-project
16:16:16 <jwb> you can be a user of fedora without necessarily wanting to participate in the project
16:16:17 <glezos> caillon: well, that's a way of contributing.
16:16:20 <jwb> right, what notting said
16:16:33 <mdomsch> stickster, notting, right
16:16:33 <dgilmore> caillon: "want to contribute to FOSS" is very vague
16:16:38 <stickster> caillon: I think glezos is right.
16:17:00 <dgilmore> contributing could just be participating in user forums
16:17:07 <stickster> Let's not debate semantics over what a contribution is. I think if you look at any Fedora team, you'll see contribution happening.
16:17:11 <mdomsch> today, do we feel that fedora-the-distro is targeted mainly at fedora-the-project members?
16:17:16 <mdomsch> it seems that way
16:17:23 <dgilmore> or it could be packaging/evangalisim/writing code
16:17:43 <mdomsch> then, how can we broaden the scope, so that fedora-the-distro can target a wider audience?  And which wider audience?
16:17:46 <caillon> mdomsch, not really.  we're not making it easy for even project members to use the distro
16:17:47 <stickster> mdomsch: I agree, and I do not believe that audience is going to be sufficient to grow Fedora in the future.
16:18:10 <mdomsch> aim small, miss small
16:18:10 <caillon> i guess as a de facto answer, sure, since we're the only ones really using it...
16:18:19 <stickster> caillon: Let's separate the issue of success/failure from the issue of what we're *trying* to do.
16:18:27 <stickster> Otherwise this discussion is going to get very confusing.
16:18:32 <dgilmore> id like to see us target all IT pros.  but maybe thats not wide enough
16:18:57 <poelcat> dgilmore: what is the criteria for an "IT pro" ?
16:19:41 <notting> stickster: well, currently, we're trying at a policy level to avoid endless arguments about stifling maintainers, bureaucracy , rules, etc., etc. so we don't apply restrictions or coherent policy in many areas
16:19:49 <poelcat> stickster: or change the question to "what we want to be" ?
16:19:54 <dgilmore> poelcat: someone who is competant with IT functions. understands basic networking, systems management, etc
16:20:14 <mdomsch> for comparison, ubuntu targets 'humans', moblin targets anyone with $400 in their pocket
16:20:27 <stickster> mdomsch: Neither of which are good definitions either
16:20:30 <mmcgrath> notting: agreed.  It seems that when a conflict or idea comes up, it's impossible to predict the outscome by any stated set of criteria.  This causes pain to everyone.
16:20:31 <mdomsch> agreed
16:20:54 <caillon> notting, which also leads me to believe we're not targetting anyone.
16:21:03 <glezos> I think that we have a long way to go to target *either* end-users who like our philosophy *or* people who want to contribute. I'll get dizzy if I try to think what we need to do to address both.
16:21:15 <glezos> caillon: +1
16:21:44 <notting> mdomsch: to be fair, ubuntu's marketing says they target humans. i suspect internally they have a little clearer picture
16:21:48 <poelcat> notting: which would lead me to believe it fedora is all about "development" and not "end users"
16:22:10 * stickster notes poelcat is saying "is," not "should be"
16:22:19 <dgilmore> notting: i bet they dont
16:22:31 <glezos> poelcat: I bet most non-Fedora users think that.
16:22:42 <mdomsch> notting, not dissing anyone, they have great focus on usability
16:22:54 <mmcgrath> mdomsch: and they have a reputation for that.
16:23:03 <poelcat> so we can spend the rest of the meeting talking about what is broken or we can start talking about WHAT WE WANT TO BE
16:23:16 <stickster> poelcat: I would prefer the latter
16:23:27 <poelcat> and what it will take to get there
16:23:28 <mmcgrath> whereas we have a great focus on new features, being first, etc.
16:23:29 <stickster> Let me point people at an interesting article.  Please read it later when you have a chance.
16:23:31 <stickster> http://www.useit.com/alertbox/participation_inequality.html
16:23:45 <jwb> poelcat, well, fixing what is currently broken is what it will take to get anywhere
16:24:03 <jwb> i agree with caillon on several points.
16:24:14 <stickster> jwb: You don't know what's worth fixing and what's worth abandoning until you know what your desired end state is.
16:24:19 <glezos> Judging from the endless discussions on this, I think we don't have enough data to judge. I'll bring once more the idea of collecting the feedback of the community with a survey or something.
16:24:19 <poelcat> jwb: i think that is too short sighted... we might miss the a bigger vision opportunity if we stop to fix everythign that is broken
16:24:29 <mmcgrath> jwb: which points are caillon's points?
16:24:40 <stickster> glezos: Surveying our existing community to find out where we'd like to be later?
16:24:55 <jwb> stickster, i disagree.  we have a lot of process and technical issues that need fixing either way
16:25:11 <poelcat> jwb: i guess i think the things that are broken can later be idenitified as roadblock in the path to getting to where we really want to be
16:25:15 <jwb> however, i'm happy to talk about what we WANT to be
16:25:33 <glezos> stickster: why not? "What don't you like about Fedora? What areas do you think we need to improve? Do you like the fact that we have one default desktop or would you like a choice in the Installation? [/me ducks]
16:26:10 <jwb> glezos, that will get you a myriad of replies
16:26:23 <jwb> which ones will you focus on?
16:26:45 <glezos> jwb: Multiple choice instead of paragraphs can help, I believe.
16:26:55 <mdomsch> I want to be several things.  A distro usable on desktops.  A distro usable on servers. The premier platform that can be extended to use cases we can't even conceive of.
16:26:59 <caillon> stickster, i also disagree with that statement....  we aren't really at the state where we can tell what needs fixing because of other stuff being too broken to get it into the hands of new users whose opinion matters most
16:27:05 <notting> mdomsch: that sounds good...
16:27:11 * poelcat wonders if we could get some feedback from the other people attending the meeting on what they want Fedora to be?
16:27:16 * stickster points back at his link earlier. That diagram and what the surrounding article says about participation and contribution is important to me in terms of our future targets.
16:27:21 <poelcat> other people == non-board members
16:27:22 <notting> glezos: that brings to mind the comment from henry ford about his users telling him that what they wanted was a faster horse
16:27:47 <glezos> notting: that's definitely something the reviewer of the feedback should have in mind.
16:28:00 <stickster> caillon: I suppose it's hard to argue vague generalities in this case. I was hoping not to spend the discussion time talking about individual things that need fixing, though they absolutely are there (and there are quite a few).
16:28:18 <caillon> fair enough point
16:29:05 <notting> so. fedora-the-distro is an open platform for people to contribute to, package for, etc.
16:29:12 <jwb> i would like Fedora to be consistent, regardless of who uses it
16:29:28 <notting> i don't want to lock down what people *can* do with it. but i think we need to somewhat constrain what *we* produce it for. if that makes sense.
16:29:35 <jwb> notting, yes
16:29:39 <caillon> right
16:29:45 <mmcgrath> jwb: that's kind of my thing too.  I have a preference in this area, but I'm fine with changing it to something else.
16:30:00 <mmcgrath> I just want it to be something :)
16:30:29 <poelcat> do we have a list of what those possible "somethings" coudl be?
16:30:47 <mmcgrath> poelcat: well we could just ramble some off, mdomsch mentioned some earlier
16:30:47 <poelcat> would it be possible to do that here or move that to f-a-b?
16:30:50 <mmcgrath> we could target netbooks
16:30:54 <mmcgrath> educational institutions
16:31:00 <mmcgrath> Experienced Linux Users
16:31:01 <notting> to point at an example, while i do not want to prevent people using Fedora on servers, i suspect that that should not be the primary focus, as there are better alternatives such as rhel or similar rebuilds in that area
16:31:03 <mmcgrath> New Linux Users
16:31:15 <poelcat> hard core developers
16:31:18 <poelcat> beginning developers
16:31:20 <mmcgrath> it can be whatever we want it to be, but the bigger it gets, the more confusing it gets.
16:31:39 <jwb> as an example, we have a common phrase that 'rawhide eats babies'
16:32:00 <jwb> if that's the position and attitude of Fedora the distro, then it doesn't matter who or what you are targeting
16:32:22 <poelcat> jwb: fedora targets killers ;-)
16:32:24 * mmcgrath doesn't think it is the attitude of Fedora.
16:32:24 <stickster> "It doesn't matter how broken this gets."
16:32:25 <caillon> and the more we target specific things such as "experienced users" the more we are "locking down" what newer users can do with it
16:32:37 <stickster> caillon made a good point about this on the private Board list.
16:32:38 <mmcgrath> also I don't think we should focus on the bugginess of the current state of Fedora.
16:32:44 <jwb> stickster, yes
16:32:52 <stickster> mmcgrath: We're trying to separate that out in this discussion, correct.
16:32:57 <jwb> mmcgrath, it is the attitude
16:33:02 <jwb> you see it all the time
16:33:29 <mdomsch> jwb, for many of mmcgrath's use cases above, a higher focus on quality may be a prequisite
16:33:30 <mmcgrath> caillon: which is exactly why we shouldn't be targeting new users.
16:33:55 <jwb> new users, or even experience users, express surprise and people reply with 'rawhide is dangerous'.  they express it for _stable_ releases and you hear silence
16:34:09 <caillon> experienced users don't even want to use rawhide
16:34:12 <jwb> mdomsch, yes
16:34:14 <caillon> people who commit code to rawhide don't want to use it
16:34:18 <mmcgrath> If we do want to target the 'any user desktop'.  Then we need to slow down. give QA time to work the bugs out, make sure things are working.  provide upgrade paths when config files change, etc.
16:34:20 <dgilmore> jwb: ive not heard anyone say "rawhide eats babies" in a long time
16:34:23 <stickster> What do we want to put in people's hands when they download Fedora 15 or 16? What do we want them to experience then and in the further future?
16:34:26 <mmcgrath> but that, to me, all gets away from Fedora's core values.
16:34:43 * stickster asked a rhetorical question there, we'll have to consider that on-list.
16:35:08 * stickster notes we're at time
16:35:26 <notting> and we've gotten nowhere? :)
16:35:26 <stickster> We don't have an answer ready-made here. That's not really surprising to anyone, I'm sure.
16:35:27 <caillon> and i think that's one of the reasons why it's in such a horrible state, really
16:35:27 <mmcgrath> stickster: I still want Fedora to be a lab where the brightest of minds come together to shape what other distors will look like in a couple of years.
16:35:40 <mmcgrath> notting: yep
16:35:41 <caillon> and that has a trickle down effect to the rest of what we produce
16:35:43 <stickster> Let's see if we can bring some of the same energy here on IRC to an actual discussion on FAB.
16:35:46 <jwb> wait, no
16:35:49 <jwb> we've gotten somewhere
16:36:28 <jwb> there are some of us expressing the opinion that we're attacking this a bit prematurely because there are prerequisites needed in all cases
16:36:35 <poelcat> everyone pile on here and we can keep this going outside of the meeting https://fedoraproject.org/wiki/Fedora_is_for
16:36:49 <notting> poelcat: wiki? not f-a-b?
16:36:59 <jwb> f-a-b please
16:37:10 * poelcat hates mailing lists for these types of things
16:37:10 * stickster thinks those are both useful tools
16:37:15 <poelcat> because things get lost and rat holed
16:37:38 <poelcat> discussion can take place on the Talk page?
16:37:38 <jwb> wikis are not good places for discussion...
16:37:43 <jwb> ugh
16:37:47 <poelcat> :)
16:37:49 <poelcat> sorry
16:37:52 * stickster is going to turn this discussion over to Q&A in 60 seconds.
16:37:53 <caillon> neither is irc :D
16:37:55 * mmcgrath is all fab, all day long
16:37:58 <stickster> jwb: There is a Talk page for discussions.
16:38:00 <jwb> caillon, heh :)
16:38:05 <stickster> but I'm open to either
16:38:14 * poelcat was looking to generate a list of *ideas* not discussion
16:38:19 <poelcat> that can happen on f-a-b
16:38:35 <stickster> OK, we're going to Q&A now, everyone.
16:38:45 <stickster> #topic Community feedback, Q&A
16:38:52 <notting> of course, the last few messages on this topic went to -board, not -fab
16:38:56 <stickster> skvidal: Go ahead
16:38:58 <skvidal> question for the board: https://www.redhat.com/archives/fedora-desktop-list/2009-September/msg00106.html <-- I'd love to hear what the board thinks about Wickert's comments here with regard to community interaction from the gnome desktop team
16:39:52 <poelcat> skvidal: what type of feedback are you looking for?
16:40:21 <skvidal> I think that his questions raise a good point and is relevant to the "Who is fedora distro for" discussion
16:40:49 <skvidal> where is our desktop sig discussing things and why does it all seem internalized?
16:41:08 <mmcgrath> stickster: are we on Q&A or skvidal's question?
16:41:09 <skvidal> I know the advantages of echo chambers when it comes to thinking your ideas are good - but it doesn't help the distro
16:41:14 <glezos> skvidal: My impression is that they raise a question for "Who is really deciding for the Fedora distro bits?"
16:41:18 <skvidal> mmcgrath: I was the first question :)
16:41:22 <jwb> mmcgrath, skvidal's question _is_ Q&A?
16:41:23 <mmcgrath> :)
16:41:23 <stickster> mmcgrath: skvidal is a community question-asker :-)
16:41:33 * skvidal feels like he doesn't count
16:41:36 <skvidal> *sniff*
16:41:49 <stickster> skvidal: I think it's detrimental to the community to surprise them with things
16:41:52 <mmcgrath> skvidal: just checking which to do first, but I guess they're the same :)
16:42:09 <caillon> apparently, i need to catch up on that thread
16:42:29 <dgilmore> skvidal: i think the desktop sig does a poor job of working with others. and it feels very much internal and impossible to be involved from outside.
16:42:33 <stickster> Not everyone is going to agree with every change, but 100% consensus is not required for changes. But surprising people with changes is not acceptable.
16:42:37 <dgilmore> skvidal: you count
16:42:37 <mmcgrath> skvidal: I think the tone there is completely out of line.
16:42:49 <jwb> mmcgrath, on who's part?
16:42:54 * spot would like to see the Desktop group be more involved and transparent with the Fedora Community
16:43:00 <poelcat> i seems reasonable to me that maybe the design team could have provided some guidance or been consulted
16:43:14 <skvidal> as an add on: https://fedoraproject.org/wiki/Desktop/Whiteboards/UpdateExperience
16:43:20 <skvidal> "We currently provide our users with a very poor experience when updating their operating system. Actually, it sucks in a variety of ways. "
16:43:27 <stickster> There are plenty of other developers, packagers, and teams that make decisions on what they want to do. But they should all be communicating it in advance.
16:43:39 <mmcgrath> jwb: really all of it, butI can see why it spun out of control
16:43:41 <stickster> skvidal: You have to go to the end of the line!
16:43:46 <glezos> spot: +1
16:43:46 <skvidal> stickster: that's fine
16:43:48 <mmcgrath> but to say 'there is no magic design team' is...  bad.
16:44:06 <notting> skvidal: taking 'update experience' as the sum total of the firehose of random things at random times with random reasons... it's 100% correct
16:44:16 * mdomsch listened to oxf13's presentation to ctyler's class last week
16:44:28 <mdomsch> as we approach beta, landing major changes shouldn't be possible
16:44:38 <poelcat> i think this points to our need to have a more wholistic approach to planning our releases
16:44:41 <skvidal> notting: saying it 'sucks' is just a tiny bit of hyperbole, I suspect - if only b/c ours sucks  - only insofar as EVERYTHING else sucks more :)
16:44:54 * stickster assumes skvidal got what he wanted from the first question and moves on to the second.
16:44:57 <skvidal> sure
16:44:59 <skvidal> thanks
16:45:02 <notting> skvidal: ... no, in terms of predictable updates, rhel is much better
16:45:11 <mdomsch> the NFSv4-by-default change wanted to happen at the last moment too
16:45:21 <skvidal> notting: fair enough
16:45:23 <caillon> (i think this sort of feeds into my point from the private board list about not really having any guidelines for code changes or updates)
16:45:26 <mdomsch> I appreciate people are working hard to improve the distros
16:45:33 <jwb> mdomsch, that was due to confusion, not due to internal only communication
16:45:36 <notting> my concern with the theme change is mainly timing
16:45:48 <stickster> I also think our update experience is bad. We too often have things broken, and though tools like PackageKit mitigate the brokenness to an extent, it's not acceptable for even non-naive Fedora users
16:45:57 <mdomsch> jwb, granted; how can we reduce confusion then?
16:46:01 <glezos> poelcat: I think we need some serious help from you in getting some valuable outcome from this meeting... :)
16:46:11 <poelcat> when i raised the issue of landing the theme change so late it was dismissed as low risk
16:46:11 <jwb> mdomsch, i think rel-eng is working that on the fedora-devel list
16:46:45 <jwb> poelcat, critical-path is what is mostly focused on now
16:46:49 <jwb> a theme is not critical path
16:47:02 <jwb> (from a rel-eng/QA standpoint)
16:47:04 <poelcat> glezos: it can't just be me... it has to be an attitude across the project and for the most part as notting noted, most people just want to be able to do their thing
16:47:30 <poelcat> you can't have a polished cohesive outcome with anarchy :)
16:47:36 * poelcat overstates things
16:47:57 <stickster> poelcat: Someone made an apt comparison to an eighteen wheeler with every wheel pointed in a different direction. I think that was overstating the problem, but imagine you have four wheels out of alignment. That's a pretty bumpy ride.
16:48:02 <jwb> so we're starting to drift into those problems i was talking about...
16:48:09 <poelcat> :)
16:48:12 <mmcgrath> poelcat: still, there is a good arguement that what we want is closer to a polished cohesive outcome, and where we are is closer to anarchy
16:48:44 <poelcat> mmcgrath: i've always felt like i was in the minority for holding that view
16:49:07 <mmcgrath> poelcat: I think people just don't like talking about it.
16:49:09 <poelcat> we should probably take another question
16:49:18 <stickster> poelcat: I haven't seen any others as of yet that are Board questions.
16:49:30 <stickster> If I missed one, someone should shout at me in the other channel
16:49:38 <glezos> FWIW, being clear about goals, having early-on discussions in the open, taking into account all the stakeholders' opinion, and other variables, BEFORE announcing a decision, is A Good Thing (tm) in open source... and shouldn't be confused with anarchy.
16:49:45 <glezos> (I'm not implying this has happened in this discussion)
16:50:06 <mdomsch> glezos, it's the opposite
16:50:55 <stickster> I think that page is on the right track as far as realizing and scoping the problem of updates, which is a major source of unhappiness in Fedora-land.
16:51:08 <glezos> mdomsch: Unforutnately, I've met people who confuse them and refuse to have a real open development ecosystem.
16:51:37 <stickster> But what I worry about is that by fixing that in the absence of guiding principles, we're just prolonging the pain
16:51:54 * stickster notes that this question is way easier to answer in a community that is basically dictated to, in the end, by a SBDFL
16:51:59 <poelcat> stickster: agreed
16:52:15 <notting> 's'? does that mean i can appoint myself? :)
16:52:30 <stickster> notting: No, you would be "NB"
16:52:36 <stickster> jwb: You can ask your question in here, go ahead
16:52:49 <jwb> does the Board have an interest in examining our current policies and procedures to see if they are really accomplishing the kind of productive environment we want to foster?
16:52:53 <caillon> YES
16:52:56 <caillon> (or at least i do)
16:53:00 <stickster> Absolutely.
16:53:00 <notting> jwb: absolutely
16:53:10 <jwb> if people want examples i could probably come up with some
16:53:17 <mdomsch> we've just been stuck trying to define the environment we want to foster :-)
16:53:33 <jwb> mdomsch, i would start with 'less broken' and work from there ;)
16:53:36 <poelcat> mdomsch: exactly
16:53:45 <jwb> ok, examples
16:53:56 <jwb> our release cycle is a 6 month death march
16:53:56 <poelcat> i guess there are two (probably more) approaches
16:54:11 <jwb> if it were 9 months, would it allow for more polish, testing, and quality?
16:54:13 <stickster> jwb: I do have a concern that we don't kick FESCo out of the picture in doing that. But in the end, FESCo is a delegate of the Board. We should invite/consult with those folks, and in the end the Board makes the call if there's not consensus.
16:54:17 * spot wonders if we can do this without hyperbole
16:54:18 <poelcat> 1) we pick 5 things we want to make less broken in F13 (the release not the person ;-)
16:54:30 <jwb> spot, i'm trying
16:54:36 <poelcat> 2) we set a bigger vision and work towards it.. .fixing broken stuff if it is in the way
16:54:56 <spot> jwb: afaik, no one has died in the course of making a Fedora release.
16:55:17 <caillon> jwb, only if we didn't move out the feature freeze.  e.g. had 3 extra months of frozen tundra
16:55:24 <poelcat> which approach do people want to take?
16:55:49 <jwb> poelcat, i prefer 1
16:55:51 <mmcgrath> wait, are we still talking about the thing we stopped talking about 20 minutes ago?
16:56:00 <jwb> mmcgrath, sort of
16:56:07 <poelcat> mmcgrath: there was nothing else to discuss ;-)
16:56:07 <caillon> mmcgrath, we didn't stop, just delayed it ;)
16:56:07 <stickster> mmcgrath: We ran out of questions, so it's not a huge problem.
16:56:17 <mmcgrath> ah
16:57:06 <stickster> At the risk of sounding too brutally honest, the evolution of Fedora starting with FC1 -> today has made 2 *very very* difficult.
16:57:21 <jwb> poelcat, the problem with a grand vision, in my opinion is that 1) it's quite daunting, 2) we have turnover that leads to reinterpretation of visions, 3) it seems very limiting
16:58:03 <mmcgrath> jwb: the problem with not having a grand vision is far worse though.
16:58:13 <jwb> really?
16:58:14 <glezos> poelcat: I think 1. is the only way to do things with volunteer communities. Unless if we're talking about paid people to work on development which ends up on our media.
16:58:27 <jwb> mmcgrath, so what is the grand vision we've been using for the past 12 releases?
16:58:39 <poelcat> jwb: IMHO we don't have one
16:58:41 <jwb> glezos, i tend to agree
16:58:44 <mdomsch> jwb, if I understand you, you are concerned that our quality isn't high enough, and believe that the 6-month cadence is too fast to do a proper job of QA.  yes?
16:58:53 <stickster> It's possible to do vision and top-down direction, but that omelette requires a lot of egg-breaking. In other words, some people -- and it's impossible to know how many until you do it -- will be completely opposed and simply drop out. And most of those people we do want to keep around.
16:58:53 <mmcgrath> jwb: there hasn't been one, look at where we are.
16:59:07 <jwb> mdomsch, well, yes.  but that is an example, not a definition of the problem
16:59:12 * poelcat has a hard stop at top of the hour... will read log later
16:59:20 <mmcgrath> people fighting on the lists more than ever because they all disagree with what fedora is.
16:59:44 <jwb> mmcgrath, the distro or the project?
16:59:58 * spot is also going to have to bail at 1
17:00:00 <stickster> Let's focus on the distro.
17:00:03 <jwb> right
17:00:04 <mmcgrath> people fighting in the project because they disagree with what the distro is.
17:00:22 <stickster> Do Board members think that the problems we are having today are due to not having a singular direction for Fedora?
17:00:32 <stickster> Simple yes/no.
17:00:35 <notting> yes.
17:00:37 <mmcgrath> yes
17:00:37 <notting> well
17:00:42 <notting> 'the problems' is a little vague.
17:00:47 <mmcgrath> at least some of them.
17:00:49 <dgilmore> no
17:00:54 <jwb> mmcgrath, as far as i can see, the increase in fighting is because we're trying to define the defacto vision, not because we don't have one
17:01:19 <mmcgrath> jwb: on fedora-devel?
17:01:19 <jwb> stickster, which problems?
17:01:38 * mmcgrath hasn't seen anyone trying to define a defacto vision there.
17:01:56 <stickster> Confused contributors, arguing developers, broken updates, baby-eating development branch
17:02:07 <jwb> well, i'm having a hard time addressing your comment when you just mention 'fighting'.  fighting about what?
17:02:15 <caillon> stickster, yes, there are problems because of that
17:02:32 <jwb> stickster, yes, there are problems because of that
17:02:32 <dgilmore> stickster: i think alot of people while working within the singular direction will still do what they want even if its contrary to the direction.  because there small bit doesnt matter if it differs.
17:02:37 <mmcgrath> jwb: they fight about new ideas, different implementations and things because they're all doing it for what they think is the best fit for $UNDEFINED.
17:02:54 <jwb> mmcgrath, no.  not undefined.  best fit for THEM
17:02:57 <mmcgrath> should we use presto or not by default?  well what would $UNDEFINED think is best?
17:03:06 <jwb> no, THEM
17:03:06 <caillon> stickster, though i believe that we could fix our processes substantially and minimize most of those problems
17:03:18 <dgilmore> stickster: alot of the issues i think we have are a result of growth and would happen regardless
17:03:36 <jwb> caillon, i agree with the caveat that the process fixes cannot be optional
17:03:48 <mmcgrath> jwb: and that's the problem.  We're not uniting towards a common goal.  We're seeing who can shout the longest and loudest.  Because until $UNDEFINED is $DEFINED, neither one is the best answer.
17:03:53 <mdomsch> the presto question is just one example of "what I need is not what you need", for different values of I and you.
17:04:13 <mdomsch> and there are technical solutions, which shouting doesn't help solve
17:04:22 <stickster> The fixes to problems like updates, stability, etc. have to be community-powered. We can't just push requirements on people and expect that they will want to stay involved.
17:04:40 <jwb> stickster, you also can't expect everyone to just magically agree on an approach
17:04:49 <stickster> jwb: I agree.
17:04:52 <jwb> we've done this balancing act for a very long time
17:04:55 <jwb> it is not working
17:04:58 <caillon> which means we need to fix the community processes so that we can have the 90% minglers.
17:04:59 <mmcgrath> jwb: no doubt.  but that doesn't mean we shouldn't have an approach.
17:05:05 <mdomsch> we need something that isn't all-or-nothing
17:05:16 <caillon> er, lurkers
17:05:26 <stickster> We have to lay out the problem precisely, show how a proposed fix will take care of it, take comments and consider the ones that can work within that goal, and then implement.
17:05:36 <stickster> Just like everything else, be willing to risk a failure
17:06:03 <mdomsch> take yum-presto for example.  a download method that knew what kind of bandwidth it was using, and could opt to switch to one or the other automatically
17:06:13 <jwb> stickster, that's fine, but the end result must not be optional
17:06:23 <caillon> i'll use mozilla as an example again.  they had the processes nice and orderly back in the days before mozilla 1.0, and the community grew, way before they had the grand firefox vision.
17:06:32 <stickster> jwb: I agree. At the end, the implementation, *is*.
17:06:40 <jwb> because all it takes is one or two developers not following the defined and agreed upon process, and it renders it useless
17:06:58 <jwb> this also applies to the development cycle btw
17:07:00 <caillon> getting everyone on the same page as to the quality and consistency of what we're trying to achieve will help foster the direction
17:07:03 <stickster> And we also need to know with the fix, how to show some time after implementation whether it succeeded or not.
17:07:22 <caillon> right now, it's basically every packager for themselves
17:07:26 <notting> stickster: community-powered, maybe. but we have to be willing to let contributors go if they don't agree, rather than let them stay around and disrupt (that sounds bad, but i don't see a way around it)
17:07:33 <caillon> and that's hurting everyone involved
17:07:52 <jwb> notting, we can't force them to leave, but i agree with what you're saying
17:08:09 <notting> jwb: ... we probably shouldn't. but we *could*.
17:08:09 * caillon too
17:08:31 <mmcgrath> Maybe we'll get ESR back!
17:08:35 * mmcgrath ducks
17:08:37 * stickster notes that there are actually things we could fix that would make it easier for people to pursue their own goals if they clash.
17:09:08 <stickster> For instance, if we had better support for the personal koji repos (talked about by many, implemented by none), people could break what they want in their own trees.
17:09:35 <caillon> and even *gasp* test their builds before unleashing hell on the masses
17:09:35 <jwb> stickster, they can do that already without relying on us to provide it
17:10:08 <mmcgrath> stickster: sort of taking the git methodology to koji repos?
17:10:14 <jwb> but sure.  kopers.  go for it
17:10:15 <caillon> jwb, for one off packages, yes.  for larger changes that need chain builds, it's... difficult.
17:10:36 <stickster> OK, we're going down a tool rabbit-hole here.
17:11:00 <jwb> caillon, a test of importance? :)
17:11:03 <mdomsch> it's also not our job to solve everyone's possible problems
17:11:13 <jwb> mdomsch, agreed
17:11:24 <mdomsch> people said doing FTBFS tests couldn't be done because we didn't have the resources
17:11:37 <caillon> but i think it should be our job to make it so they don't have as many problems to begin with
17:11:52 <mdomsch> people said rel-eng couldn't do automated rebuilds because the format of the Version and Release tags was too arbitrary
17:12:18 <mdomsch> some seemingly large problems really aren't that difficult to solve, given sufficient focus
17:12:31 <mdomsch> and we tend to get wrapped around the axel over such
17:12:51 <jwb> mdomsch, same with nightly livecds, etc
17:13:01 <mdomsch> jwb, sure
17:13:06 <mmcgrath> and jwb's koji install :)
17:13:13 <jwb> hm?
17:13:38 <jwb> oh.  let's not go down that road...
17:13:42 <mmcgrath> jwb: "it couldn't be done".  but you got it up and going!
17:13:46 * mmcgrath only kidding.
17:13:58 <mdomsch> on a related note - I'm _really_ glad with how the domain trademark licensing discussion has gone
17:14:08 <mdomsch> that was one point where people were really wrapped around the axle
17:14:16 <mdomsch> and seemingly impossible to resolve
17:14:21 <mmcgrath> mdomsch: yeah, that went in circles several times
17:14:29 <stickster> mdomsch: And I think the key was getting counsel to move the discussion to an open forum
17:14:38 <mdomsch> stickster, sure - which was unprecidented
17:14:43 * stickster doesn't want to slide off the current topic, though.
17:14:49 <jwb> mdomsch, ok, so i think in summary you could say this: We need to be less afraid of toe-stepping and more willing to get our hands dirty
17:14:50 <glezos> back. Sorry.
17:14:54 * glezos reads back.
17:14:56 <mdomsch> no, just using it as another example
17:14:56 <caillon> stickster, what's the current topic? :)
17:15:02 <mdomsch> jwb, yep
17:15:18 <jwb> so, i like poelcat's idea of 5 things
17:15:43 <jwb> come up with 5, don't worry about who's supposed to be 'responsible' for it, and start working
17:15:49 * stickster is willing to go with 5 things to see how we can impact F13
17:16:21 * mmcgrath will do that as well but doesn't think it is on target for the discussion we've had today.
17:16:33 <jwb> why not?
17:16:44 <jwb> i see it as evolutionary
17:16:46 <mmcgrath> they're at different levels.
17:16:51 <mmcgrath> one is high level, one is low.
17:16:54 <jwb> yes, but not disjoint
17:17:14 <mmcgrath> I guess the stuff I'm thinking about is more in the F14 F15 timeframe.
17:17:23 <jwb> maybe we'll get insight or definition by fixing some of these things.  maybe we'll just make a better distro.  i don't see either as a bad thing
17:17:49 <mmcgrath> For F13 I'd like to see more people help the QA team get their goals done.  They have lots of good ideas and can use more people to help.
17:17:57 <mmcgrath> but that doesn't come anywhere close to answering what we're talking about today.
17:18:02 <stickster> We can set these up as two separate discussions, because I still think the question of "where do we want to be in F15/F16 timeframe?" is important to answer.
17:18:28 <mmcgrath> jwb: but you're still talking about working towards $UNDEFINED
17:18:34 * stickster thinks the conversational energy here has flagged.
17:18:39 <mmcgrath> lets work (lets drive).  where?
17:18:55 <jwb> i always sucked at answering the 'where will you be in 5 years' question in school.  it seemed impossible to answer to me
17:19:05 <jwb> the best thing i could say was 'not here'
17:19:10 <jwb> or 'here'
17:19:59 <mmcgrath> stickster: we goign to endmeeting?
17:20:04 <stickster> caillon: Sorry, didn't mean to ignore you. I think we have two topics going. One is the high-level topic of "What do we want the Fedora distro to be achieving in the F<n> (15, 16?) timeframe?"
17:20:29 <stickster> caillon: The other is "What are five low-level problems we absolutely want to fix before F13 to improve the distro?"
17:20:44 <stickster> mmcgrath: Yes, in a moment
17:20:47 <jwb> where low-level is low to the Board
17:21:01 <jwb> not necessarily things like "i want bug 5948959594 fixed"
17:21:08 <stickster> jwb: Correct, thanks
17:21:14 <notting> seems nice. although i worry about the board's lack of directable resources
17:21:25 <stickster> notting: Excuses, excuses...
17:21:35 <stickster> Let's stop here, and take those two subjects to FAB threads.
17:21:39 <mmcgrath> notting: another issue for another day :(
17:21:51 <stickster> And I would like to see replies from as many of the Board members as possible.
17:21:56 <caillon> notting, you should get some, i recommend it.  makes my life easier :P
17:21:58 <stickster> Any objections?
17:22:13 <stickster> OK
17:22:16 <stickster> meeting ends in 5
17:22:17 <stickster> 4
17:22:18 <stickster> 3
17:22:20 <stickster> 2
17:22:21 <stickster> 1
17:22:24 <stickster> #endmeeting