00:03:28 <yn1v> #startmeeting fedora board town hall May17th 00:00UTC
00:03:28 <zodbot> Meeting started Mon May 17 00:03:28 2010 UTC.  The chair is yn1v. Information about MeetBot at http://wiki.debian.org/MeetBot.
00:03:28 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
00:03:57 <yn1v> let's have nominates self intro in one line.
00:04:19 * lcafiero is Larry Cafiero, the Regional Ambassador for the US West Coast States and an Ambassador mentor for NA.
00:04:27 * jjmcd has been working with Docs.  Retired, long experience, all about process
00:04:50 <smooge> My name is Stephen Smoogen, a member of Fedora Infrastructure and an employee of Red Hat both now and in the past.
00:05:01 <rdieter> Rex Dieter, former board and fesco member, a leader in kde-sig
00:05:10 <spot> Hi everyone, my name is Tom "spot" Callaway, Fedora Packaging Committee, current Board member, SPARC lover, lousy dancer, and Red Hat employee since 01.
00:05:23 <mizmo> Hi, I'm Máirín Duffy, I'm the Fedora Design team lead and a user experience designer for Red Hat, and I don't have turtles.
00:07:27 <yn1v> What does each candidate see as the most pressing issue that the board needs to work on in the coming year?
00:08:06 <spot> yn1v: do you want us to free-for-all reply?
00:08:24 <yn1v> I will call in the order nominated.
00:08:32 <yn1v> can you start spot?
00:08:37 <spot> I'd be happy to.
00:09:24 <yn1v> then smooge, mizmo, lcafiero, rdieter, and jjmcd
00:10:04 <spot> I think it is important for the Board to set some big, long term goals for the project. For a long time, we've been a group of people all doing their own things without any sort of real guidance, and now that we have a vision statement and an idea of who Fedora is targeting, we should start setting goals that span multiple releases and encourage the community to help us define Features and improvements that help us meet those big goals.
00:10:33 <spot> Its great to say "First", but I feel strongly that we need to continue to innovate and improve.
00:10:42 <spot> EOF
00:11:18 <yn1v> smooge is next
00:11:25 <smooge> typing.. one sec
00:13:55 <smooge> I believe that the major goals of the board in the next year are: working on getting the community into a less crappy atmosphere, finding a continual set of targets for us to focus on and improve on, making sure that Fedora Engineering has the resources and input to make a release on time. In many ways they all tie into each other.
00:14:39 <smooge> When we spend our time fighting over who can have the most drama.. we aren't working on the other two. And vice versa.
00:14:42 <smooge> EOF
00:14:54 <yn1v> great ... mizmo ?
00:15:09 * mizmo typing
00:16:04 <mizmo> I think the most pressing issue for the Board to focus on in the coming year is refining the work the current Board has done in defining Fedora's vision
00:16:19 <mizmo> building it into an actionable roadmap that we can align all our efforts under
00:16:50 <mizmo> I think right now we have a lot of islands in the project, and I think some of the discontent and drama that folks have mentioned (there was even a question on the candidate questionnaire regarding it)
00:17:10 <mizmo> has come from folks not understanding how their work fits in with others and the friction that comes from that.
00:17:26 <mizmo> I think this is the most pressing issue because solving it will help alleviate multiple problems at once.
00:17:27 <mizmo> EOF
00:17:57 <yn1v> lcafiero, your turn
00:18:00 <lcafiero> Keeping the cats herded and going in the same direction is no easy task, and that's what the upcoming board faces going forward. It bears mentioning though that of the top distros, Fedora does a lot of things right despite the perception of increasing "drama," and the key is to keep things going forward, meeting goals and keeping things positive.
00:18:03 <lcafiero> eof
00:18:07 <lcafiero> thank you, yn1v
00:18:19 * lcafiero has 24 fingers
00:18:20 <yn1v> pre - writing good
00:18:36 <yn1v> rdieter ?
00:19:49 <rdieter> A top priority for me would be increasing community outreach at the board level : to have a continual measure of the project pulse, acting to facilitate excellence and productivity, (and as a practical matter) assimilating and redirecting dissent.  Contributors need to know and feel they have a voice and are being heard, that the board supports all of their efforts.
00:19:53 <rdieter> eof
00:20:20 <yn1v> finally jjmcd
00:20:26 <jjmcd> As I said in responding to the questionnaire, I think we need to work on ways to make information easier to find and more transparent.  We need to instill a culture of making the background for each decision obvious, and make it easy to find the information.
00:20:36 <jjmcd> This helps not only ease the transition of newer contributors, but also makes it harder for the more "passionate" contributors to disagree endlessly.
00:20:42 <jjmcd> eof
00:21:07 <yn1v> that was a question from bcotton.
00:21:21 <yn1v> now we have a question from jds2001
00:21:32 <yn1v> what can the board do to minimize the effect of "posionus people"? As stated in a thread on advisory-board, one can be posinous yet still be "excellent to each other". I'm curious about the candidate's stance on what to do with and how to handle these community members.
00:21:52 <yn1v> anyone wants to take that one first?
00:22:53 * spot will take a crack at it
00:23:06 <yn1v> great!
00:23:14 * lcafiero next
00:23:28 <smooge> smooge next
00:23:31 <jjmcd> then jjmcd
00:23:56 <rdieter> me
00:24:22 * mizmo last!
00:24:35 <spot> I'm not convinced that one can really be "excellent" to their community members when they're poisonous, but there are some things we can do. At some point, we have to step up and tell these poisonous people that the damage that they do to the community outweighs the good of their contributions and bounce them, but I think we should reserve that action as a last resort option.
00:25:25 <spot> Its a hard problem to solve, because you're having to tell people that they're wrong, and no one likes to hear that.
00:25:48 <spot> But not everyone can be right all the time, and that's fine. The trick is how you handle disagreement.
00:26:21 <spot> As long as people are being respectful and constructive, I think it's okay, but when it stops being those two things, they are not excellent.
00:26:31 <spot> EOF
00:26:53 <lcafiero> It depends on what you define as poisonous people -- people can disagree without being disagreeable (though in some cases, that has not been the case and (to be honest) I may be guilty of being less than excellent in the past. Nevertheless, I think as long as the group as a whole stays focused, ideally the "poisonous people" will either reform or go away. Ideally. Bouncing them, as spot says, should be a last resort
00:26:53 <lcafiero> , and I believe we have a mechanism in place (or at least one has been discussed).
00:27:59 <lcafiero> In any group you're going to have friction, no matter how focused all are on the same goals. The key is defusing any conflict early and move in the same direction.
00:28:02 <lcafiero> eof
00:28:59 <smooge> I do not believe a person can be poisonous and 'be excellent to each other'. A person can disagree, can point out problems, and be in dissent and not be poisonous. Poisonous to me means that a person is inciting to kill/maim others.. to break their spirit to work with others. These sort of people might think they are just plain old dissenters, but in the end their actions make it harder for others to work. I believe in a 3 strikes, where the
00:28:59 <smooge> board comes to an agreement to warn the person, and if they do not correct then bounce.
00:29:33 <smooge> sorry if that is a bit disjointed the finger cast keeps getting in the way.
00:30:54 <yn1v> can I pass to jjmcd ?
00:31:01 <smooge> anyway dissenters are not always poisonous, but we have to remember that just being a dissenter does not make a person a 'hero'. In any case, I figure the bounce action would need to be documented and in the open so that it doesn't look like its the popularity party
00:31:04 <smooge> eof
00:31:17 <jjmcd> I think I responded to that.  More transparency makes it hard to keep on with disagreements.  When you you can clearly articulate why some decision is taken, preferably in a quantitative way, it takes the steam out of endless conflict.  Of course, again, if everyone is on the same page ahead of time, conflict is less likely in the first place.  But we shouldn't forget Larry's earlier comment; I recently got a peek as some other projects and we have it
00:31:18 <jjmcd> really good here.
00:31:34 <jjmcd> Sometimes you may need to bounce someone, but better to avoid the need in the first place.
00:31:40 <jjmcd> eof
00:32:24 <rdieter> one persons' poison is often anothers... umm... candy, nectar?  hopefully you get the idea.  anyway, I feel strongly that the 'excellent to each other' guideline fits and works well to guide conversations in constructive directions.  However, there does exist a point where sometimes parties will simply have to agree to disagree.
00:32:36 <rdieter> And that's ok!
00:32:55 <rdieter> That's the point where I believe reaching out is very important, to reassure folks they've been heard and not just ignored or dismissed.  After that, if all else fails, then the board can deal with the remaining 'poison' as neccessary (but I honestly don't think it'll ever have to come to that).
00:33:13 <rdieter> (I like the other answers I've heard so far too... :)
00:33:14 <rdieter> eof
00:33:23 <mizmo> I don't believe anyone comes to a project like Fedora with the goal of being poisonous, although it can be hard to remember that from within the heat of a flame war. :)
00:33:32 <mizmo> I do think it's challenging to gently lead folks whose actions are poisonous onto the right path if the right path isn't clearly documented and communicated. I think today if we try to say to someone, 'Well, that doesn't actually fit in with our vision for Fedora so it's probably not the best idea', it places the current vision in a position of being questioned - and it will be questioned. That will lead things even more off-track.
00:33:41 <mizmo> An effective tactic I've seen in diffusing such situations is pointing out the common goals of the sides involved. Fedora's vision should serve that role, and can be a useful tool in getting folks on the right track. I've noticed in a lot of conflicts, having a reasonable answer as to why something is the way it is, even if the person doesn't agree, goes a long way in repairing the situation.
00:33:47 <mizmo> That being said, I have had to put at least one person on moderation on the Design Team (at the tie art team) list in the past. We can't be afraid to moderate folks who are truly being a problem, but we do need to give them the benefit of the doubt and try to work with them before resorting to moderation.
00:33:48 <mizmo> EOF
00:34:01 <yn1v> great!
00:34:26 <yn1v> inode0, asks: People are less cranky when having more fun. Can the board reduce Fedora crankiness by increasing contributor fun?
00:35:30 <yn1v> can we do the same? each one call to set the order of answers ?
00:35:36 <spot> Man, I wish it was that easy. Not everything in Fedora is always going to be fun, and what is fun to one person may not be fun to another.
00:36:32 <spot> I strongly encourage fun things, but I'm not so naive to think that everything will always be fun. I'm always open to suggestions on how we can improve things, to make them better, more efficient, more fair, and heck, more fun.
00:36:34 <spot> EOF
00:37:01 <yn1v> how's next?
00:37:07 <rdieter> I can go
00:37:12 * lcafiero next
00:37:18 <yn1v> go rdieter
00:37:21 * jjmcd next
00:37:41 * rdieter first hides his anti-fun platform
00:37:51 * mizmo next
00:38:25 <rdieter> speaking for myself, I get the most rewards from having a sense of accomplishment.
00:39:17 <rdieter> so best to match contributor's talents and interests to get the most of that...
00:39:57 <rdieter> and the board needs to do everything it can to facilitate that
00:40:18 <rdieter> eof
00:40:20 <lcafiero> I'd echo spot -- I'm open to ideas, short of things that are immoral or illegal. eof
00:40:22 <smooge> smooge last
00:40:47 <jjmcd> Is this a plant or what?  I keep coming back to the same message.  When technical people know what to do, they do it and enjoy it.  The answer is making it clear what needs to be done.
00:40:58 <jjmcd> Although as Tom says, not everyone has fun at the same things, the people who do enjoy X are willing to put up their hand to do it, if they know it needs to be done.
00:41:08 <jjmcd> I hate to look one dimensional, but you keep playing to my message!
00:41:10 <jjmcd> eof
00:41:50 <mizmo> I agree that folks are less cranky when there's a sense of fun and play in the environment you're working in.
00:41:57 <mizmo> I think if we have a clear vision that folks are united towards achieving, then I think a lot of the things that frustrate folks will be easier to deal with, and more room to have fun will open up.
00:42:06 <mizmo> I think the Board can help facilitate that in our approach to communicating decisions and plans.
00:42:07 <mizmo> eof
00:44:53 <smooge> slow typing.
00:46:07 <smooge> Ok I would like to say that to make things fun for contributers we need to know more what contributers want to do. It is part of knowing where we are going, because if we have a strong direction and contributers know it and are attracted to it, we can find fun things to make it happen.
00:47:49 <smooge> Beyond that, I would like us to get more FAD's going around the world and seeing what we can do about making ambassadors empowered and energized (eg its what they are good at so what do they need).
00:48:03 <smooge> sorry my agenda is mostly anti-fun I think :)
00:48:08 <smooge> eof
00:48:19 <yn1v> I will squeeze my question Fedora is growing and expanding, how the board will built bridges to keep us together, coordinated and in sync ? bcotton also has the idea of making ramps for newcommers
00:48:52 * spot has been going first all night, i call last on this one. ;)
00:49:04 <yn1v> noted!
00:49:38 <mizmo> i can go first on this
00:49:44 <yn1v> go ahead!
00:50:01 <mizmo> How do we know what types of folks to recruit if we don't know where we want to go? How do potential contributors know this is the right project for them, whether or not it is worth the time/effort investment if they don't know where we are going (or that we agree on where we're going at all?)
00:50:02 <mizmo> ...
00:50:49 <mizmo> I don't mean to sound like a one-trick pony, but I think the Board facilitating an actionable and inspirational roadmap for Fedora's future will help us get better at recruiting and placing new contributors in positions where they can feel fulfilled / productive and our project can succeed faster.
00:51:01 <mizmo> EOF
00:51:20 <yn1v> who' s next?
00:51:34 <smooge> I can take 3rd.
00:51:38 <smooge> typing now
00:52:03 <lcafiero> rex? You want to go or shall I?
00:52:36 * jjmcd can go whenever suits
00:52:51 <lcafiero> I'll yield to jjmcd and go after smooge
00:52:58 <rdieter> I can go last
00:53:09 <jjmcd> spot called last
00:53:15 <jjmcd> Part of making the necessary information accessible is making the path to it map the work process.  Onboarding new contributors, as I mentioned, is a key work process to address, and it is different on each team.
00:53:24 <jjmcd> Clearly, we need to do a better job of articulating our vision, but again, it's all about communication.  I get the impression that the current board has made some progress developing a vision, but it appears to still be secret.
00:53:32 <rdieter> oh, last-1
00:53:35 <jjmcd> Perhaps it needs to be fine-tuned, but in any case, it needs to be well understood, inside and outside of the project.
00:53:47 <jjmcd> eof
00:54:53 <smooge> My view is that we can't keep growing in either packages or people without major changes in governance and structure. Some of the problems we are having is due to the social entropy that every group of people has as it grows in size. At certain times, you reach a point where you can't keep people 'cool' enough to get work done and things break down. At such points an organization has to decide what it wants to do. Change governance and contin
00:54:53 <smooge> ue growing, or keep current governance and grow down in size to a more managebel point
00:55:42 <smooge> The next year will be 'us' (as in all people with voting membership) working out which of the two we want and how we are going to accomplish it.
00:55:53 <smooge> eof
00:56:07 <lcafiero> jjmcd is right about different paths for different parts of Fedora -- I know we may do things differently in Ambassadors. However, I think there is too much focus on what the board can do for Fedora, as if it has to come from the top down, when it should be "what can the community do for Fedora," coming from the bottom up. The board should be facilitating process that move the project forward, and to do that it need
00:56:07 <lcafiero> s input from the community at large.
00:57:37 <lcafiero> eof for now (though I think I'll address this further at another time)
00:57:52 * lcafiero is somewhat distracted by $DAYJOB temporarily
00:57:54 <yn1v> rdieter, your turn
00:59:35 <rdieter> ok, what lcafiero said ^2.  I think vision has less significance here honestly, there's plenty of onroads into the project already, imo.  facilitating those, making things clearer, easier, lending a helping hand when necessary will go a long way.
01:00:51 <rdieter> and, encouraging participation in higher level entities, things like fesco, infrastructure, design, rel-eng , makes teams feel connected as well
01:00:52 <rdieter> eof
01:01:09 <yn1v> spot, your turn
01:01:13 <spot> I think there are a few obvious things we can do to build bridges:
01:01:20 <spot> 1. Have common project-wide, well defined, clear goals, with measurable success and end dates.
01:01:29 <spot> 2. Act transparently.
01:01:37 <spot> 3. Encourage open and cross group communication.
01:01:44 <spot> 4. Make it easier for people to get involved, through positive mentoring, good docs, simplified processes, and minimal red tape.
01:01:53 <spot> I'll note that this is pretty much the same thing that my fellow candidates have said, but I would add that it is vital that the Board continues to act as an example and set the bar high in these areas.
01:01:57 <spot> EOF
01:02:14 <yn1v> we have one last question
01:02:24 <yn1v> inode0 There is a perception (reality) that the project is drifting (being nudged) back in the Core direction. Are there particular things from the Core days that in retrospect you view as good for the Project that need to be restored?
01:02:50 <yn1v> who want to be last this time?
01:03:01 <smooge> me
01:03:20 <spot> i'll go first
01:03:25 * jjmcd whenever
01:03:35 <rdieter> 2
01:03:40 * lcafiero will go third
01:03:49 <yn1v> go spot!
01:04:20 * mizmo 4th
01:05:19 <spot> I think that the idea of having a set of packages which receive a higher amount of attention, coordination, testing, and consideration is a useful goal.
01:05:20 <yn1v> what makes jjmcd 5th and smooge last
01:05:43 <spot> However, the way that Red Hat achieved it back in the "Core" days isn't the only way (or even, a very good way) to get there.
01:06:26 <spot> I think that our community can work together and accomplish a solid installation footprint for Fedora, I think that we do a pretty good job of that today.
01:07:02 <spot> I also think that the fact that people are proud of their packages helps make them better, and the fact that people want Fedora to be awesome makes it that way.
01:07:35 <spot> But no, I don't really want to hand the main set of packages back to Red Hat and say "we don't want input, you do it for us"/
01:07:45 <spot> That part of Core, I don't miss at all. :)
01:07:46 <spot> EOF
01:08:13 <yn1v> rdieter, your turn
01:08:31 <rdieter> critical path considerations ... is evolving into a Core-like thing.  A good thing too, but it'll take some iterative steps to get the right fit, a balance of safety vs productivity.  Otherwise, I feel very uncomfortable dictating any sort of segregation, esp at the board level.  I think we can and should do better.
01:08:40 <rdieter> eof
01:09:36 <lcafiero> Clearly I will yield to those who have been around longer than me. I started using Fedora regularly at Fedora 7, which I think is the first "post-core" Fedora, so I am not cognizant of what things were like "in the old days."
01:09:39 <lcafiero> eof
01:09:51 <mizmo> One of the big advantages to switching away from the Core-Extras split was that it allowed contributors outside of Red Hat to contribute to the Core. Should we switch back to a Core-like model, this clear advantage won't be lost - anyone in the community can contribute to the Core.
01:09:58 <mizmo> I think the *focus* the Core model had was a major advantage. We've recently started a critical path policy which I think the question is referring to. Focus is a good thing. It helps guide us for which packages updates are more destabilizing for and for which packages issues should be considered more seriously as blockers than others.
01:10:02 <mizmo> We can't deliver 100% on every single package and release every 6 months, and we can't block the release for issues in any given package. Having a core of packages to focus on I think increases our chances of delivering excellent software in a consistent and dependable manner.
01:10:03 <mizmo> EOF
01:10:22 <jjmcd> We have a LOT of packages.  If the user experience is to be good, then some of those packages are going to have to be vetted more thoroughly than others.
01:10:33 <jjmcd> But does that mean that some packages are "more equal"?  Maybe, but that doesn't mean we need to revert 100%.
01:10:44 <jjmcd> I agree with Rex that the board shouldn't bless a particular set of packages as being special, I think that is a job for metrics.  But at some point, we will need something like that.
01:10:53 <jjmcd> eof
01:11:36 <yn1v> smooge, your turn
01:11:57 <smooge> Well the issue of going back to Core days is mostly scalability. We have only so many engineers (both volunteers and paid) and only so much time. We would like to make it so we have a set that will work and be built at a certain time. We can try to push the envelope but we seem to end up with more engineers taking more packages and then spectacularly burning out. The difference I see between Core and Critical Path is a) people both inside and
01:11:57 <smooge> RH are working on things and agreeing this is the way to go, b) package selection is more open, c) we are approaching it as an iterative process. If we have a better method can we grow it? can we help teams build their own 'crits' to make say gaming or mnopDE or pony-rides spins have their own set of core things they need and can keep track of
01:12:24 <smooge> I think that it is also a common cycle in a growing organization.
01:12:35 <smooge> Things grow, things recede, things grow again.
01:12:44 <smooge> Without that you end up with core rot.
01:12:45 <smooge> EOF
01:13:03 <yn1v> I feel that we got good questions and good answers. I think that we have a great time and we were able to see strengths on our nominates
01:13:22 <yn1v> I want to thank for your time (we run a bit late!) and also for your patience as this is my first meeting as moderator.
01:13:26 <smooge> thank you
01:13:34 <spot> thanks yn1v for moderating
01:13:35 <yn1v> anything else before we close this meeting?
01:13:35 <smooge> and thanks to people asking things.
01:13:37 <rdieter> thanks, good questions.
01:13:38 <jjmcd> Thank you, good session
01:13:40 * inode0 would like to thank yn1v for moderating tonight too
01:13:46 <lcafiero> Thank you yn1v and thanks townhall-public denizens
01:13:49 <spot> and thanks to all who participated, candidates and questioners
01:14:01 <smooge> I would #endmeeting
01:14:22 <yn1v> #endmeeting