From Fedora Project Wiki
No edit summary
(3 intermediate revisions by one other user not shown)
Line 7: Line 7:
These question have been sent to the candidates:
These question have been sent to the candidates:


=== Main ===
=== Main Questions ===


# [[#question_1|If you could single-handedly change one thing about Fedora, what would it be and why?]]
# [[#question_1|If you could single-handedly change one thing about Fedora, what would it be and why?]]
Line 19: Line 19:
# [[#question_9| How will you make the work for in Fedora easier and more fun?]]
# [[#question_9| How will you make the work for in Fedora easier and more fun?]]


=== More ===
=== More Questions ===


#[[#more_1| Who is Fedora for?]]
#[[#more_1| Who is Fedora for?]]
Line 554: Line 554:
|}
|}


===More===
== More Answers==
{|
{|
|colspan="2" style="font-style:italic; background: #ddd; font-weight: bold;"| <span id="more_1">1. Who is Fedora for?</span>
|colspan="2" style="font-style:italic; background: #ddd; font-weight: bold;"| <span id="more_1">1. Who is Fedora for?</span>
Line 1,092: Line 1,092:
|-
|-
|}
|}
== Notes ==
Some experiences and notes that might be helpful for future Candidate Questionnaires:
Workflow:
* aim to collect answers in private and publishing them in one go -- then candidates have no chance to look at the answers from other candidates; that sounds like a minor detail, but seems to have helped a lot to encouraged the candidates
* to help with that only add candidates answers to the official answer documents/pages that were handed in before the deadline
* prepare the questions early and have them ready and public in the wiki early -- ideally at the beginning of the nomination period or something like four days or one week before the end of the nomination period; set a tight deadline for handling in the answers like "end of nomination plus two days". That should make sure the whole election process doesn't take to long and give everybody that nominated enough time to sent the answers -- sure it's tight if people nominate late, but that's their fault ;-)
* to help with that make sure the deadlines are known before the nomination period starts
* seems a lot of people liked the OpenOffice table for comparing the results we had in the past, as it's possible to easily hide candidates and answers/questions you are not interested in; to reduce copy'n'paste (and thus reduce error potential) consider to put such a basic table with the answers (one per row) into the wiki that candidates (one per column) need to use to hand in the answers; then all that is needed is to merge the rows with the answers into one table, which is quite easy; from that table it's easy to export them into many other formats without too much work
Questions:
* "open ended" questions obviously are highly preferred
* someone needs to review the questions and merge similar questions into one and remove others to make sure it are not to much questions (something like 16 to 20 likely should do)
* some of the old question were quite good and universally;
* some people dislike questions like "do you prefer Gnome or KDE" that are/should be mostly irrelevant for Fedora as whole and the position the candidate is nominated for, but others people really like the answers as they give a impression about the person itself. So a few of those question are good and definitely acceptable, but "few" is likely important

Revision as of 18:50, 14 June 2009

Election Questionnaire

Please place questions you would like to see candidates for the various offices answer. If your question is specific to one office (such as the Board or FESCo), please indicate that.

Questions

These question have been sent to the candidates:

Main Questions

  1. If you could single-handedly change one thing about Fedora, what would it be and why?
  2. Please name three things you plan to work on and realize while being on the Board or FESCo!
  3. If elected, how will Fedora as a project be better as a result of your leadership? Or IOW: What strengths will you bring to the Fedora board/FESCo that are currently missing?
  4. What do you see as Fedora's greatest strength and weakness, and what will you do to improve upon that? Or IOW: Which processes have worked best in Fedora, and which processes need to be improved?
  5. What is Fedora's place in the larger community with respect to other distributions?
  6. What are you going to be doing in the Fedora Board that you cannot do outside of it or how would being in board/FESCo help in what you want to accomplish?
  7. What do you consider to be Fedora's raison d'etre? In the past the focus of Fedora has shifted from release to release. Do you see a long-term goal or a "target audience" Fedora should strive for? How do you define your role in helping the project reaching that goal?
  8. What are your unique strengths and what are your weaknesses?
  9. How will you make the work for in Fedora easier and more fun?

More Questions

  1. Who is Fedora for?
  2. Should any steps be taken to make sure releases don't get as much last minute delays as in the past? If yes: which?
  3. How or to what extent would you say are Fedora's governance bodies responsible for protecting volunteers and volunteer based efforts against interference from within Red Hat by either person or policy?
  4. The Fedora Project suffers from a lack of communication. In 2009 for 8 out of 21 FESCo meetings no meeting minutes were sent to the lists. What would you do to improve communication between the different groups in Fedora and especially between FESCo, Board and the community?
  5. What would you do to cleanup and organize the Fedora project packaging guidelines, rules and other wiki pages to make it more consistent and easier for new contributors?
  6. There is a proposal out there to moderate fedora-devel. What should be disallowed? Racist or sexist speech? Profanity? At what point does a complaint against a small project become a personal attack? ie "Your idea is utterly stupid." Should we disallow trolls or posts likely to start flamewars? Who will decide the difference between intelligent debate and flaming?
  7. Please give three examples of other boards or communities you have participated in and the positive differences you made there.
  8. Simple question: Among the two, do you prefer Gnome or KDE?
  9. Do you think a Fedora Foundation is still worth pursuing? Why or Why not?
  10. Should kernel module packages be allowed in Fedora?

Answers

For alternate review methods, we provide two other versions of the answers:

* text based (copy'n'pasted from the emails with the answers)
* in a openoffice table

Main Answers

1. If you could single-handedly change one thing about Fedora, what would it be and why?
Tom Callaway (spot) If I could snap my fingers and have the world redraw itself around me, I would lift all the legal barriers preventing patent and DMCA encumbered FOSS code from being included in Fedora. RPMFusion does a very good job of keeping these packages maintained, but it would be much nicer to have these things properly integrated into Fedora. Of course, I realize this isn't likely to actually happen anytime soon. :)
Josh Boyer (jwb) This is an odd question to me, given that if I could single-handedly do it myself, then I'd just DO IT. Most of the issues we face take collaboration and in that spirit I'll answer with: The way we do communications at times. We have a myriad of lists of varying topics, the wiki, IRC, etc. And yet we still seem to have trouble getting the proper information to the people interested in it. I have no great solution for this, but it is the one thing I would like to see improve.
Mike McGrath (mmcgrath) Better market recognition of what fedora is and who should use it.
David Nalley (ke4qqq) I'd introduce time-shifting to allow us to accomplish everything without slips and yet for all appearance look like we hit schedule with no struggle. I'd also use this to schedule vacations for those contributors who constantly seem busy and overworked. Seriously though, I see continued growing pains in different areas of Fedora - I wish I had a way to single-handedly mitigate some of those growing pains.
Dennis Gilmore (dgilmore) Enable people to find a place that they fit in fedora, more easily, sometimes it seems to be hard to find out how to scratch your itch.
Kevin Fenzi (nirik) I would change the perception that Fedora is just a testbed for RHEL and not a very nice distribution in it's own right. I think we would gain more contributors if this was more widely known.
Kevin Kofler (Kevin_Kofler) I'd rename the "Desktop Live" spin to "GNOME Live". I think we really ought to call a spade a spade and not falsely imply there's only one desktop.
Adam Miller (maxamillion) Marketing/Ambassador focus - Fedora already leads the GNU/Linux world (and in my opinion the entire Open Source world) in innovation and upstream contributions. The Fedora 11 feature list is second to none but where I feel sometimes we fall behind is the efforts put forth to get Fedora "out there" and I don't want anyone to think that I am saying our Ambassadors or Marketing Teams are not doing a great job, because I think both are doing a bang up job but I feel that other GNU/Linux projects that are ranking as "more popular" on user polls on sites like linuxmagazine.com due to their hightened marketing focus from their respective project.
Bill Nottingham (notting) The culture of the development list (and by extension, possibly the development community itself) needs to change. Every week there's a new hot issue (or three) that has raised the ire of someone and leads to a long and protracted discussion that rarely has any benefits.In a couple of weeks, it ends up being forgotten as the next big issue pops up. It gets to the point that when I see any reasonably high number of unread messages, my initial thought isn't "ooh, let's see what interesting work is being done"; it's "oh no, what's the problem *now*?". I've even received e-mail and messages from others who find it so bad that they don't even want to post development discussion there, as they feel it will just get drowned in more pointlessness and flaming. If the attitude on the mailing list is that bad that we're actually driving contributors away, I think it goes to the top of the list as a place we need to change and a problem we need to solve.
Jens Petersen (juhp) Keep rawhide installable most of the time. Maybe by having a devel testing repo of critical/unstable core packages and/or using bodhi voting. (My second one would probably improving our web UI.)
Andreas Thienemann (ixs) I'd singlehandedly magic up an unlimited supply of developers, documentation writers, UI designers and translators etc. That way we'd have all our bugs fixed, all missing features added and would never miss a single release again. Unfortunately, our real problems are not that easily fixed but need collaboration between our contributors.
Seth Vidal (skvidal) Well, I'd want to have the military arm of fedora to be much stronger. Our strikeforces should legitimately be able to occupy the Isle of Man and secure all resources there indefinitely. :) Seriously, though, I'd love to have fedora-devel-list and fedora-test-list be entirely troll-free in the blink of an eye. Unfortunately we can only do it a little bit at a time.
Ian Weller (ianweller)
Christoph Wickert (cwickert) Introduce corporal punishment for maintainers who don't react to their bugs. Just kidding. I think a single change is not enough and I'm not running for FESCo just to change a single thing, so I cannot really answer this question.
David Woodhouse (dwmw2) This doesn't have to be realistic, right? Or practical? I'd want to shift it to the Free World so that we can incorporate the RPMFusion packages, then do a crippled spin for those who really need it. While we're set up to do these 'crippled' spins, we could make a flagless spin too, one without obscenity, one without naked women... pick your own nation's fetish :)
Return to top


2. Please name three things you plan to work on and realize while being on the Board or FESCo!
Tom Callaway (spot)
  1. Simplifying the Fedora CLA: The existing CLA is a mess, and it is a bit too broad for the needs of Fedora. I plan to work with Red Hat Legal and the Fedora Board to reword the CLA so that it is a minimal, easy to understand document with a smaller audience. Ambassadors and people who want to make wiki changes should not need to sign the CLA, for example.
  2. Defining the primary Fedora target: This is a tough one, because I do believe that it is a good thing that Fedora has a broad community of users and developers, but it is also important for the Board to define who the _primary_ target of Fedora is, so that we can make smart design and development decisions.
  3. Working on making it easier for people to participate in Fedora. I want to eliminate red tape and bureaucracy wherever possible and ensure that people are able to be involved and happy in the Fedora community.
Josh Boyer (jwb) These are in no specific order:
  1. Continuing the Secondary Architecture effort.
  2. Continue to try and encourage users to become contributors in a number of ways.
  3. Aid in defining "what is fedora". That seems like an entirely challenging topic to tackle, and I like a challenge.
Mike McGrath (mmcgrath) Stronger focus for the project. More freedoms for contributors. Growth of Fedora the project beyond Fedora the OS.
David Nalley (ke4qqq)
Dennis Gilmore (dgilmore) Secondary arches more widely accepted and worked on. EPEL moved to koji/bodhi requirements for new VCS
Kevin Fenzi (nirik)
  1. I would like to try and work on the backlog of merge reviews. I think if we can get that monkey off our back we could open up a lot of great possibilities, like re-reviews of existing packages, less frustration for contributors and increase quality.
  2. I would like to try and see about increasing communication between the various Fedora support channels: IRC, forums and mailing lists.
  3. I would like to try and increase communication also between FESCo and the community.
Kevin Kofler (Kevin_Kofler)
  • Fight for KDE as a first-class citizen (on the same level as GNOME): KDE should get equal space on the download page, on the installer DVD and in naming of spins, changes in shared desktop technologies such as PolicyKit need to keep KDE in mind, menu entries need to mention the actual name of the application because there is generally more than one for each task etc.
  • Fight against policies such as restrictions on country flags which are both at odds with my idea of freedom and sending us on a collision course with upstream projects (including, but not limited to, KDE).
  • Act as a liaison between FESCo and the KDE SIG.
Adam Miller (maxamillion)
  1. The main thing I want to bring to the group is a non-biased approach to forward innovation, the Steering Committee is only there to guide the project along, not control. I would most like to cater to the needs of the developers who are coming to the Committee with the ground breaking features and ideas of tomorrow.
  2. I would like to address issues such as KDE and Xfce's status. I agree with the fact that we need a "default" such that there is a "Fedora look" that is uniform for users but I also think that some of the communities of none Gnome groups would like to see some more effort to equate the playing field, which somewhat goes along with my non-biased sentiment of keeping things equal.
  3. Passion - sound corny? Sorry, but its true. I live for this stuff, I am an Ambassador and I look forward to every chance I get to talk to someone about Fedora and to promote the wonder that is our distribution, our project and our community.
Bill Nottingham (notting)
  1. Changing the development culture to be more productive, if at all possible
  2. Try to work towards consistent guidelines across packages where possible, instead of 'every developer has their own guidelines'
  3. Continue to drive the feature process to highlight the great work being done in Fedora, and to hopefully catch more things before they become issues that delay the release
Jens Petersen (juhp) Making Fedora better and more appealing for international users (eg YumLangpackPlugin feature). Help with improving our processes for better quality and increase stability during the devel cycles. Try to get the layout of Bugzilla improved for better usability.
Andreas Thienemann (ixs)
  • Helping FESCO and Fedora focus on technical excellence. We are supposed to be the leading distribution for interesting new features. It's even part of our slogan. Right now however, there's too much time spent on discussing politics and not enough time left for working on features and fixing ensuing bugs. This should get better with less politics in the way.
  • Strengthening the non-desktop aspects of Fedora. Fedora does a lot of development for the desktop use case, but we've got to take other users into account as well. The Server SIG will obviously benefit from any such involvement but also projects such as ovirt, the Fedora Appliance Feature or the Robotic SIG. All these are interested in a more lightweight Fedora which we should look into how to best achieve this.
  • More QA and testing. I've been very glad to see a stronger focus put onto QA this release. Test Days have proven very useful with users having prepared test plans available and developers and other knowledgeable contributors at the ready to track down any occurring problems. I think this and other initiatives are the way to go to improve our release quality and I'd like to support this wherever I can.
Seth Vidal (skvidal)
Ian Weller (ianweller)
Christoph Wickert (cwickert)
  1. Cut the dependency bloat. Pidgin for example must not depend on evolution-data-server just because there is an optional feature of syncing contacts (which does not work reliable and is useless anyway if evo is not installed). Programs must be packaged more fine grained and we probably need to make more use of alternatives and virtual provides. But these are details that need to be discussed in FESCo and FPC first.
  2. Improve communication (read below).
  3. Improve work flows. Some processes are just too bureaucratic, which delays development and scares off contributors.
David Woodhouse (dwmw2) In general, I'd like to strive to make Fedora the best Linux distribution -- and the best Operating System overall -- for as many people as possible. That's not a Fedora fanboi mindset -- if _we_ take that approach, and if our 'competitors' take the same approach, then everybody wins. More specifically, that means...
  1. I'll do what I can to make Fedora work, and _keep_ Fedora working, on as many platforms as possible. So I'll to what I can do assist with secondary architecture development so that hopefully we can actually get a secondary release out the door some time soon.
  2. I try to resist the 'GNOME disease' which manifests itself by telling users "your use case is uninteresting; go away". I think we do too much of that (cf. recent flamefests).
  3. Promoting packager responsibility. The value-add of a Linux distribution is that have put in the effort to make everything we ship work together, as a coherent whole that we call 'Fedora'. I think it's very poor if some packages are second-class citizens and don't get proper maintenance, because their packagers refuse to, or are incapable of actually fixing problems in the package.
Return to top


3. If elected, how will Fedora as a project be better as a result of your leadership? Or IOW: What strengths will you bring to the Fedora board/FESCo that are currently missing?
Tom Callaway (spot) Keeping in mind that I am running for re-election to the Fedora Board, I think that I bring a significant amount of experience with Fedora to the Board, having been involved with the project since the Red Hat Linux days. I also feel that as an active packager, I have a high level of appreciation for the pain that Fedora packagers feel, and I am a good representative of that aspect of the community.
Josh Boyer (jwb) I think the Board is fairly well balanced today. We have a diverse set of members from a variety of backgrounds. So I don't think there is anything major that is currently missing. What I will try and contribute is a view on overall release quality and package robustness, as well as representing the community as best I can.
Mike McGrath (mmcgrath) I think a better focus and more defined project goal will go a long way for Fedora. We really aren't SuSE, Debian or Ubuntu and shouldn't get compared to them.
David Nalley (ke4qqq) Compared to most of my running mates I am a n00b, having only joined the Fedora Project in early 2006. I hope to bring some of that vantage point to the Fedora Board.
Dennis Gilmore (dgilmore) since i'm on FESCOo i don't offer anything new there. For the board I offer someone to represent the the smaller portions of the fedora community.
Kevin Fenzi (nirik) This seems like more a question for someone not already in fesco, but I think I have a good deal of understanding of how things work in fedora and work in many different parts of it. I think this perspective is very usefull.
Kevin Kofler (Kevin_Kofler) Fedora will benefit from a KDE point of view, and more generally an alternative point of view, being represented in FESCo. I believe we need to get more diverse points of view into FESCo to avoid the current paradox where several discussions found a large consensus for one opinion on the mailing list, but FESCo subsequently voted the exact opposite. Statistically speaking, a less biased sample is the best way to reduce that effect.
Adam Miller (maxamillion) I don't necessarily think the the Board or FESCo is currently lacking so much as I think it might be a time for some new minds to be brought into the equation so that new ideas can be seen within FESCo. I think I would be able to bring a new outlook to FESCo because I would be a new and ambissious thought process added to the discussions.
Bill Nottingham (notting) My biggest strength is that I bring a broad variety of experience working on Linux, on Linux distributions, and in Fedora. I've been doing this for long enough that I feel I've developed a pretty good eye for the right way and the wrong way to do things and implement things. I'm running for re-election, so it would be disingenous for me to claim that that's currently missing, of course.
Jens Petersen (juhp) International and Asian experience and perspectives: I am originally from Denmark and Britain and I lived 12 years in Japan and now in Australia. I have been working on i18n since I joined Red Hat.
Andreas Thienemann (ixs) Having followed some of the FESCO meetings on IRC I would have liked FESCO to act a bit less hesitant in controversial cases. Instead of trying to reach a compromise or making a final decision these controversial topics are often deferred. I'd like to see more assertiveness in FESCO: If a compromise cannot be reached because one party is not willing to compromise, it is necessary to make a final decision even if it is difficult. If a feature is not ready for a release, it is perfectly okay to push it back for one release or even longer. While these decisions are not easy and guaranteed to make people unhappy, it's better for the project and the distribution to have this cleared up early. The necessary feature readiness review was rather late this time. I'd prefer this earlier in order to give necessary feedback and ultimately helping the developers reaching their goals.
Seth Vidal (skvidal) My main contribution to Fesco is a certain hard-minded-ness that has been useful on the board for the last many years. Occasionally, folks get too caught up in the process and forget that sometimes the process is: 1. not required and 2. not helpful. I tend to bring a "is this really necessary?" mindset to organizations I work on. I think that questioning those assumptions is a big strength I bring to fesco.
Ian Weller (ianweller)
Christoph Wickert (cwickert) I cannot pledge that Fedora will be better because I'm in FESCo. Remember: If I get elected, I'm only a single member and I don't make decisions by myself. My strength is a high pain threshold. I'm not afraid of controversial discussions.
David Woodhouse (dwmw2) Proper spelling. But seriously, see above.
Return to top


4. What do you see as Fedora's greatest strength and weakness, and what will you do to improve upon that? Or IOW: Which processes have worked best in Fedora, and which processes need to be improved?
Tom Callaway (spot) I think our greatest strength is our community as a whole, we are passionate and hard-working. There are a few key weaknesses that we still need to improve upon:
  • QA: We're making good steps here, but we still need to build automated QA infrastructure.
  • Package Review: We need to streamline this process, and automate as much of it as we can.
Josh Boyer (jwb) Fedora's greatest strength is it's contributor base. We have a great set of contributors that continute to drive Fedora forward. I think one of our weaknesses (and we do have more than one), is that we as a community can get too hung up on tiny details and can easily lose some of the 'bigger picture'. For example, creating policies and processes for everything under the sun. It is my belief that a process should be a benefit and _help_ to the overall project, not a hurdle. So trying to processize every possible issue that comes up just adds more hoops to jump through and generally increases the amount of effort needed to participate.
Mike McGrath (mmcgrath) Fedora's greatest strength is it's large contributor base, it's weakness is letting that contributor base flounder a bit.
David Nalley (ke4qqq) One of the strengths Fedora possesses is the relationship with Red Hat. There is made available to the project, because of this relationship, incredible resources in people, time, money, and equipment that would likely not be available otherwise. This relationship has also been a hindrance at times, though largely it's gotten far better than most would have hoped.
Dennis Gilmore (dgilmore) Fedoras greatest strength and weakness is the rapid pace of development. constantly pushing the boundaries and being first in market in many cases.
Kevin Fenzi (nirik) I think our contributors are our greatest asset. We need more of them to share their passion for Fedora. Our weakness is growing that pool of contributors. I would love to see us engage our end user community and show them how great and fun it is to become an active contributor.
Kevin Kofler (Kevin_Kofler)
  • Fedora's greatest strength is its "always up to date" character. In fact, it works so well there isn't much to improve, though some packages could benefit from more aggressive version upgrades during stable releases. (The kernel and KDE are good examples to follow, those packages not even getting fully-compatible bugfix point releases in some Fedora releases are examples where improvement is possible.) Another huge strength of Fedora is its strict commitment to Free Software.
  • Fedora's greatest weakness (leaving aside things which cannot be fixed such as software patent issues) is the subtle GNOME bias which keeps haunting us from time to time (despite all the improvements having happened over time and the work done by the KDE SIG) and decredibilizing us in KDE circles. In particular, feature pages advertising GNOME-only features need to say so more clearly, the download page needs to present KDE right next to GNOME rather than hiding it behind an extra link and using "Desktop" as a synonym for GNOME needs to stop. Justifying restrictive policies such as a ban on flags by saying "it's the GNOME project's policy" as at least one FESCo member did also makes no sense whatsoever.
Adam Miller (maxamillion) Greatest strengths in my opinion are the leadership in Open Source, upstream contributions, development of bleeding edge technologies, and an incredible community at large. The question on Fedora's weakness was already posed so I won't reiterrate but there is another concern I have which is the inability to find information due to unmaintained wiki pages which is something I would really like to see get fixed, while this isn't specifically a FESCo issue I do feel that FESCo can make a push for features owners and SIGs to keep their pages up to date.
Bill Nottingham (notting) I don't think those two are actually the same question, as the strengths and weaknesses aren't necessarily processes. (In fact, if the strength of Fedora is only in our processes, we're in trouble.) Our biggest strength is our vibrant community; we've attracted a large variety of contributors working in lots of areas, whether they be art, development, or our fine infrastructure team. We've successfully produced not only a world-class Linux distribution, but also a hosting environment for fostering open-software development, a reference case for quick global infrastrucutre deployment, and a translation infrastructure that's now being adopted by other projects. Furthermore, for most usage cases out there, we've got packagers in Fedora working to bring software for that case to the users; we do a good job of satisfying the long tail. Our biggest weakness is that we have this great diversity of people, but we don't particularly focus them. So, we end up with large groups of people each saying that their own usage case or methodology is the most important, and can't possibly be made worse at the expense of some other contributor, or even the good of the project overall, even if the harm is completely imaginary. So we end up in endless arguments (most of them silly), and miss the opportunity to make the overall project and release better.
Jens Petersen (juhp) I think innovation is Fedora's greatest strength - shipping close to latest and greatest with reasonable stability. The weakness is probably the processes and overcomplicated workflow for package reviews, pkg cvs admin, translations, etc. I think people are aware of the need of a unified web dashboard for contributors and so I believe things will continue to improve.
Andreas Thienemann (ixs) Fedora's greatest strength is no doubt our community. Having met a variety of people during several FUDCons I'm amazed what a huge range of contributors from different countries we have and how easy it is to get along with most of them. The friends part of our slogan should not be underestimated as it makes working together so much easier and fulfilling. I do not know what our greatest weakness is, but I see several key problems preventing us from being the one distribution to rule them all:
  • Quality. We're getting there, but I'm still finding too many problems in core components.
  • Communication. We have to make our main communication channels useful again.
  • Increasing amount of bureaucrazy. Needs less.
  • Community Input. We need moar input from affected parties in our decision finding process.
Seth Vidal (skvidal) The mailing lists need some work. They get out of hand too quickly and it deters useful/productive work. Amazingly bugzilla and the triaging process is one of the more functional processes. Everyone beats up on bz b/c it takes a while to do certain things. However, whenever it is compared to anything else, it wins pretty much hands down.
Ian Weller (ianweller)
Christoph Wickert (cwickert) Strength: The great community and it's contact to upstream. We have many contributors that are also leading personalities in popular projects like X.org or Firefox. Weakness: Quality assurance. We have improved recently, but as we all know it's still a long long way to go.
David Woodhouse (dwmw2) One of the things which concerns me is the way that some packagers are protective of 'their' packages and don't want anyone else to touch them. I've been vocal about the dangers of that attitude, and I'm very pleased with FESCo's decision to require FESCo approval for any package which wants to disallow provenpackager commits. At the moment, there's nothing outstanding that I'm _desperate_ to fix, in the way that I was to fix that problem. There are always tweaks required, but in general I think we're doing a good job. We do need to be careful of having _too_ much bureaucracy and too many policies, rather than relying on common sense.
Return to top


5. What is Fedora's place in the larger community with respect to other distributions?
Tom Callaway (spot) Fedora is a leader. We are pushing the boundaries of Free Software every day, and we have a commitment to working with upstream to make significant improvements. We are also working with other distributions to collaborate on complicated efforts or to standardize our processes. For example, I'm meeting with members of the OpenSUSE community at LinuxTag to see where we can have a common set of packaging guidelines.
Josh Boyer (jwb) I view Fedora as the leading (sometimes bleeding) edge distribution overall. It is often the first to ship the latest versions of packages, and it strives to do so in a quality manner. More importantly, it is often the first to ship the combination of the latest packages, which helps shake out bugs in the package interactions and has a trickle down effect for other distros.
Mike McGrath (mmcgrath) Fedora works with upstream and gets new technologies first.
David Nalley (ke4qqq) Fedora is the place of innovation. This means that many people come here to deploy new technology and features first, and come with the idea that this is the proving ground, that deploying within Fedora will gain them a huge userbase and perfect their product. We've are also rapidly becoming the standard bearer for freedom and openness.
Dennis Gilmore (dgilmore) Fedora is a leader and prover of new technologies. it is a place where everyone can scratch there personal itch and come together to make a great product for everyone to use.
Kevin Fenzi (nirik) Well, Fedora is not an enterprise distro with support channels and such, but otherwise I think it fits a large number of use cases. I think it's ideal for todays desktop/laptop use. I think it's great for development servers for exploring new tech.
Kevin Kofler (Kevin_Kofler) Fedora is the distribution doing most of the innovation. All distributions benefit from us pioneering new technologies.
Adam Miller (maxamillion) I think we are the bar in which other distributions are measured against. Though I might be biased in that thought. I think Fedora leads in every adpect that a free and open source project should with loyalties to the ideals that entails.
Bill Nottingham (notting) Fedora is committed to making a distribution that anyone can use, modify, and redistribute freely for any purpose (commerical or otherwise). We don't include poison-pill software that only has guarantees if received from Fedora; we don't include binary-only software that the user can't modify, etc. We're committed to solving problems the right way whenever possible, even if it's more work, not papering over them with hacks.
Jens Petersen (juhp)
Andreas Thienemann (ixs) I believe the claim of Fedora being the most advanced Linux distribution, combining and showcasing the newest developments is spot on. Fedora is the distribution closely following and often even driving upstream development of a wide range of free and open source software. While other distributions might ship similar versions of packages, it's Fedora influencing the direction of future Linux offerings. I fear however, that during the process of defining the future we're sometimes losing track of our current users and their real world problems with out distribution. If that's true, it should be rectified ASAP.
Seth Vidal (skvidal) Freedom-focused, rpm-based leading edge distro. To use terms folks will get more: from a freedom standpoint were Debian-or-better of the rpm-based distro world from a newest-bits standpoint we are alone in trying to get all the newest items Our policy of upstreaming patches has set us apart from many other distros and we're starting to see upstream maintainers realize this advantage to their projects.
Ian Weller (ianweller)
Christoph Wickert (cwickert) First of all Fedora is one of the big players. We have a large user base and a lot of installations, although (fortunately) we don't have a lot of fanboys advocating Fedora at distrowatch or slashdot. Second, Fedora is the driving force of innovation. Many things we do get picked up by other distributions sooner or later. We need to enhance cooperation to make it happen rather sooner than later.
David Woodhouse (dwmw2) To strive to be better than the others. And to learn from the others when we don't manage it.
Return to top


6. What are you going to be doing in the Fedora Board that you cannot do outside of it or how would being in board/FESCo help in what you want to accomplish?
Tom Callaway (spot) I think that participating in the Fedora Board helps me ensure that Fedora makes intelligent and strategic long term decisions, and that the best interests of the Fedora community are well represented.
Josh Boyer (jwb) To be honest, not much. One of the things I like the most about Fedora is that you don't have to be in a committee or on the Board to really make an impact on the project. We strive to keep that true and I continue to be impressed at the amount of work that gets done without any sort of Board/FESCo interaction. That being said, I do want to participate in some of the higher level project discussions that the Board tackles. Things like trademark guidelines, 'What is Fedora', etc.
Mike McGrath (mmcgrath) I'm hoping to focus more on visionary goals for Fedora on the board. In my current position I work a great deal on implementation of those goals but not so much in forming them as that's the board's job.
David Nalley (ke4qqq) I personally would love to help drive consideration of a Fedora foundation-like entity.
Dennis Gilmore (dgilmore) Make sure that those smaller groups inside fedora get heard. Provide a strong voice for secondary arches.
Kevin Fenzi (nirik) I'm not sure that it does really. Anyone can contribute and step up without being in fesco. Being in fesco allows one to help guide things as a whole and help others come up with the best way forward. I look forward to helping others with their ideas in fesco and bringing my ideas and plans there.
Kevin Kofler (Kevin_Kofler) As I am someone who wants to bring change into FESCo, getting actually elected into FESCo (hopefully together with others on a similar mission) is the best way to get there.
Adam Miller (maxamillion) I can't answer this question because I'm not walking into FESCo with an agenda to push like some politician being shoved in by lobbyists, but instead as someone who is passionate about the success of the project. This question in my opinion is loaded because nobody really knows what they are going to do until a situation occurs and the discussion is had, information and ideas are exchanged and educated decisions are made and this is exactly what I would do my best to accomplish if I were elected.
Bill Nottingham (notting) I want to ensure that Fedora remains the best, highest quality, forward-looking Linux distribution out there; one that I find compelling to use and interesting to focus work on. By serving on FESCo, I'm able to try and ensure those goals, in the approving of Features, packaging policies, and other development giudelines.
Jens Petersen (juhp) I think it would be good to have someone on FESCo outside the US and Europe.
Andreas Thienemann (ixs) The planned QA support and my interest in Fedora as a non-desktop system do not need a position in FESCO. The aim of easing the maintainers work however does, as we've seen in the past. But to directly influence the policies governing the work of our maintainers and other contributors as well as and the direction Fedora takes in the future a place on Fesco is needed.
Seth Vidal (skvidal) Vote on policy changes.
Ian Weller (ianweller)
Christoph Wickert (cwickert) Voting. I have the feeling that a growing number of decisions in FESCo do no reflect the view of the community and this is something that can only be changed inside of FESCo. Before FESCo votes on something, it needs to ask people for feedback. Of course this can slow things down, but nevertheless I think it's absolutely necessary. FESCo needs to keep in touch with the community and must represent it's views.
David Woodhouse (dwmw2) As a member of FESCo, I can contribute to decisions in a more concrete way and can propose acceptable compromises where necessary. And vote for what I think is right, of course :)
Return to top


7. What do you consider to be Fedora's raison d'etre? In the past the focus of Fedora has shifted from release to release. Do you see a long-term goal or a "target audience" Fedora should strive for? How do you define your role in helping the project reaching that goal?
Tom Callaway (spot) This is a difficult question to answer. I think that the long term goal of encouraging Free Software is key to Fedora, and I do believe that we should define a primary target for Fedora to ensure that our efforts are not spread too thin or confused. I also don't think that we should take any steps that would limit the effort of motivated community members to shape Fedora for their own needs or wants. To put it bluntly, if our primary target is the Desktop PowerUser, we should still try to make every effort to not make life difficult for a SIG working on making a Fedora Server spin. Its a difficult balance to walk, but I think we have to try.
Josh Boyer (jwb) I think this is something we as a community are struggling with right now. For a long time, just getting a quality release out the door on time was enough to satisfy people. And that may still be valid for many. I know that is usually enough for me, as I tend to enjoy the whole rel-eng aspect of things. However, others are looking for this long-term goal as a way to guide their contributions. The Board has recently started tackling this issue with the 'What is Fedora' discussions, and I do hope to participate in those.
Mike McGrath (mmcgrath) Fedora is the Linux technology leader. I also want to help Fedora become more of a tool for the larger OSS and technology players. When Dell, IBM, or whoever says "I want to put $NEW_PRODUCT together and give it a try." I want Fedora to be the first in that list for deployment and partnership. I'd also like Fedora to be a way for the big players to get in touch with the tons of little players for feedback, testing, input, etc. Fedora is a perfect platform for this.
David Nalley (ke4qqq) I think focus that shifts from release to release is a good thing. One it's driven by the people doing the work, and as this is a meritocracy, the people doing the work should continue driving the direction of Fedora. I can't help but think that by pigeonholing ourselves as a 'desktop focused' distro or 'server focused' distro that we end up losing valuable contributors, and that it would unnecessarily constrain innovation. I honestly think that the four foundations really encapsulates a lot of Fedora's mission and fear too much being set in stone. I don't envision the Board as being the oracle on high who tells us which direction to go, but rather the guard rail on the edge of the precipice that keeps the project within bounds and out of danger.
Dennis Gilmore (dgilmore) I think fedora is more of a techie, developer distro. it is a place where you can prove that your new shiny thing works and is valuable to all. it is useable by everyone but not neccesarily always the best choice. Fedora is not for those wanting long term support, that is where CentOS and RHEL fit in. I personally run servers with fedora on them.
Kevin Fenzi (nirik) I think it's impossible to come up with one 'target audience'. Fedora is what we contributors make it. If you want to have a distro that works well on desktops for end users, then work on that. Of course some use cases are out of scope or incompatible. In those cases I think we need to choose where best to spend out resources. We should strive to allow various use cases where there are contributors working on those areas, and if its possible to.
Kevin Kofler (Kevin_Kofler) Fedora's raison d'être is to be 1. 100% Free Software and 2. always current, but not unstable ("leading, but not bleeding"). We should not compromise on these goals. So our target audience needs to be prepared to get updates, including version upgrades, regularly, and in particular to have sufficient bandwidth to download them (at least as DRPMs). They also need to be prepared to upgrade to a new release at least once a year. And finally, they need to accept that they will not find proprietary or patent-encumbered software within Fedora, but only in third-party repositories. But within those constraints, I don't think we should exclude anyone. However, not excluding anyone also means not designing exclusively for new users (as GNOME is arguably doing), and this one of the reasons why supporting KDE well is important, and this is where the KDE SIG, in which I participate, enters the picture. You can also count on me to stand up for what I see as our raison d'être (i.e. "Free and current") and against any attempts to encroach these principles.
Adam Miller (maxamillion) I don't think we have a "target audience", I think fedora should be for everyone. That's one of the things I love about Fedora is that in the event a group of Fedora community members want to cater specially to a specific group of users they are free to launch a SIG and/or a Spin in order to accomplish this, which is a wonderful thing because those community members continue to contribute to the project as a whole instead of branching and creating child distributions. I also don't think a long-term goal is very feasible in such an evolving environment where we work with next generation technologies every day, I can't know what's to come and therefore can't make a long-term destination.
Bill Nottingham (notting) Fedora should probably focus more than we do now on a particular audience; by not targeting any audience (or designing for 'all' audiences, you ensure that you're not optimal for any audience. It's why the Fedora DVD spin is traditionally the weakest offering; it's a jack of all trades and master of none. With a target in mind, you can actually make meaningful decisions about what you're doing. That being said, the target audience is more a board question than a FESCo question; FESCo would just be the implementer.
Jens Petersen (juhp) I think it has to come from the grass roots from contributors. Fedora is both a testing ground for new technology, good devel environment and desktop(s) and a great base for spins and rebranding.
Andreas Thienemann (ixs) I would like Fedora to be the universal operating system. Fit for the server, versatile enough for the desktop, small enough for the netbook, modular enough for building an appliance, modern enough for the developer. This needs more work on the packaging side however, which I'm willing to support FPC for. About shifting focuses: I think we should sit down and clearly define our focus. Keep it wide enough to encompass our important use cases and users but narrow enough not to be faced with illusive demands. Based on this focus, we then define our goals and are free to keep them for a number of releases. How I'm helping the project? During the Berlin FUDCon Barcamp in June I'll be leading a discussion session about exactly that topic. The aim is to gather input from contributors what we should be aiming for. As a member of FESCO, I'd be in the unique position to actually execute the necessary work.
Seth Vidal (skvidal) The need for a 'target audience' for fedora is just so we can set goals and say definitively "we will not be implementing feature-Y b/c we aren't targetting them". Target audience goals are about saying "no", not about saying yes. Part of the reason we need a target audience is that a lot of software development is about intelligently saying no. Fedora needs to say no more often. Being everything to everyone is chaos.
Ian Weller (ianweller)
Christoph Wickert (cwickert) Actually I don't think that focus has shifted that much. It only did in subareas, but there always was a central theme: Innovation and technical excellence. I think ware are doing an good job in this area, but of course we still can do better. Our target audience are mainly early adopters and developers, but that doesn't mean that we should limit our focus to those users, because I have seen many people with different knowledge and skills using Fedora without problems.
David Woodhouse (dwmw2) I don't really want to restrict the target audience. I think we can provide a general-purpose distribution which is applicable to all kind of targets through servers, desktops, netbooks and handheld devices.
Return to top


8. What are your unique strengths and what are your weaknesses?
Tom Callaway (spot) Strengths: I haven't come across a piece of code that I couldn't get into an RPM package, and I understand FOSS licensing better than most hackers. Weaknesses: I'm not as good of a software coder as I would like to be. I (occasionally) have great ideas, but lack the skillset to implement them.
Josh Boyer (jwb) I have no amazing super-powers. I generally consider myself fairly competent on the engineering side of things, with a focus more on code than packaging. I also try to be open minded, but I'm not afraid to take a position on something if I feel strongly about it. As for weaknesses, I have those too. I'd like to be more educated in how the non-engineering communities of Fedora function. I can be overly negative at times. I also have a decreasing tolerance for trolls and flame-fest email threads, though I do still try to read them. I dislike kittens and ponies.
Mike McGrath (mmcgrath) I'm a fast learner who knows when it's time to work, and when it's time to have fun. I'm a results oriented person though the path from point A to point B is also important. I also value simplicity which I see as a strength. My weaknesses include the occasional short temper and horrible spelling.
David Nalley (ke4qqq) As a relative n00b I think I understand the challenges faced by some of the more recent contributors, and the people who have yet to begin contributing. At the same time, I don't have the experience leading the Fedora Project that most of the other candidates do.
Dennis Gilmore (dgilmore) I have a good understanding of how the different parts of fedora work how they all fit in together from buildsys to releng and everything in between. I may not always know how a regular user may use something. helping my mum use fedora has shown me that.
Kevin Fenzi (nirik) well, I have been involved with fedora for a long time, and know a lot about how things are put together. I think thats both. New perspectives may come up with a better ideas, but knowing the history and reasons for things is important as well.
Kevin Kofler (Kevin_Kofler)
  • Strengths: I have experience with both Free Software development and packaging, I also have a formal university education in Mathematics and Computer Science, I speak 4 languages fluently (English, French, Italian, German), I'm used to IRC meetings from my involvement in KDE SIG. People at my university consider me a brilliant student, in KDE SIG I'm known as a "Qt/KDE ninja" for my successes at fixing critical bugs or tracking down their causes.
  • Weaknesses: I tend to overreact when something angers me (and need to try really hard to stay polite) and can't always avoid to "feed the trolls".
Adam Miller (maxamillion) I would prefer not answer this one.
Bill Nottingham (notting) For strengths, see question #2 above. My biggest weaknesses are that I'm really bad at delegating, and I don't find nearly enough time to do actual development and bugfixing work. In combination, that can be a bit of a problem for software maintenance.
Jens Petersen (juhp) Attention to detail, and saying yes more than no.
Andreas Thienemann (ixs) Strengths: I have vast experience getting diverse groups of people to agree on a compromise as long as people are interested in finding a solution. If this is not the case, the end result is usually still acceptable but not universally liked. Tough luck but it's still necessary to come to a decision. Weakness: I don't have an army of ninjas to promote an attitude of violence. This could have been useful in speeding up some processes.
Seth Vidal (skvidal) I have an Adamantium-laced skeleton and there's a mutant healing factor in there somewhere, too. My biggest weakness is that I'm slow to accept changes I don't drive. I will push back against change that I don't see as necessary. I do not believe all change is progress. Nor do I believe all progress is change.
Ian Weller (ianweller)
Christoph Wickert (cwickert) I'm used to cope with criticism, in fact I need it as a vital feedback. I think this is something that every FESCo member should have, because in the steering committee you are in an exposed position. On the other hand my criticism sometimes is a little harsh. I can be demanding and dogmatic. Other weaknesses include poor English and bad typing (still handicapped from an accident two years ago).
David Woodhouse (dwmw2) I am relatively good at seeing technical solutions and implementing them. I'm not always the most diplomatic person you'll meet.
Return to top


9. How will you make the work for in Fedora easier and more fun?
Tom Callaway (spot) Less paperwork, less manual steps, more areas of opportunity. :)
Josh Boyer (jwb) Fun is a relative term. For example, if you don't find packaging fun to begin with, I doubt I can do anything to really make it more fun. And 'easier' isn't always a goal either. To be honest, some of the items and issues we face _are_ hard and trying to make them easier usually doesn't get any progress. There are times we need people to simply dig in and work such items, even if they aren't fun or easy. That being said, there are things we can do to make sure we aren't making things harder than they really are. I answered a previous question with the example of too many policies and procedures. Things like that can and should be avoided and should hopefully make Fedora a more enjoyable project to contribute to.
Mike McGrath (mmcgrath) I think Fedora is in an interesting transition period right now in an industry that's currently in transition. Getting a better focus in place so that all of our thousands of contributors are working towards the same goal will keep expectations in line more. Additionally, in the more practical sense, I'll continue my work in Fedora Infrastructure to keep things fast as well as continue to help make more tools feature rich and useful.
David Nalley (ke4qqq) I would hope that staying out of the way would help work progress in a fun and easy manner. Perhaps I could tell a few jokes, though my wife claims I have no talent when it comes to humor.
Dennis Gilmore (dgilmore) I will try to make it simpler and more consistent.
Kevin Fenzi (nirik) I would like to see more engagement and community amoung contributors. Perhaps some non tech gatherings on irc, or more use of #fedora-social or the like.
Kevin Kofler (Kevin_Kofler) By opposing policies which require changing upstream projects' software for no good reason and thus place a pointless workload on the maintainer(s). Again, the restrictions on flags are a good example of such an annoying and useless policy. I also hope to be part of a more diverse FESCo which will lead to fewer decisions diametrally opposed to the majority's views and thus reduce frustration.
Adam Miller (maxamillion) I don't understand the question.
Bill Nottingham (notting) Ice cream and ponies for everyone! More seriously, I'm not sure that that's the proper goal. We shouldn't make things unnecessarily hard or irritating, of course. When it comes down to it, fixing a bunch of blocker bugs isn't necessarily going to be 'easy'. Filling out an update request isn't necessarily going to be 'fun'. But if we can automate QA of releases and updates, we can catch issues earlier, making less pain for our users and developers. If we can have clear policies on when to, and when not to, update library versions, we can prevent developer aggravation if they feel they have to do unncessary rebuilds. If we get changes into RPM to make packaging guidelines obsolete, we make life easier for packagers when they can simplifiy their spec files. But most of these things don't actually require FESCo or Board privleges to do.
Jens Petersen (juhp) I hope we can get a better web interface in place that will make contributors' lives easier.
Andreas Thienemann (ixs) Sane policies, no micromanagement, letting people do what they do best without unnecessary meddling from third parties. That should make the work easier. More fun? Organising events such as FUDcon in order to allow our contributors to gather, to catch up with each other in real life is a very important part. It helps people get to know each other, form a better understanding or clear up any issues remaining from discussions online and get back home revitalized having gained some new friends. All this is what I consider fun. Logically, this means, we need more events, as this equals more fun. \o/
Seth Vidal (skvidal) How do you make work easier and more fun? Umm. If it was easier and more fun it wouldn't be work, would it?
Ian Weller (ianweller)
Christoph Wickert (cwickert) Easier: Improve communication. In an project with many many people involved communication is the key to success. I'm sure better communication will More Fun: Try to establish local events where people meet face to face, like the "Fedora Round Table" which was a lot of fun.
David Woodhouse (dwmw2) Mostly by striving to avoid gratuitous bureaucracy.
Return to top

More Answers

1. Who is Fedora for?
Tom Callaway (spot) That's the big question, isn't it! I truly believe that Fedora is flexible, in that you can make it work for your needs, but in the same breath, that there needs to be a well defined primary use target. This may be controversial, but I don't think that primary target is "Aunt Tillie, Desktop User, afraid of the Command Line". I think Fedora is more for a power-user, someone who is comfortable with using a computer, but also wants good tools (GUI and CLI) to help them be productive with their day to day tasks (Email, Web, Games, Development, Communication, Multimedia). I think Fedora is for folks who like to know how things work, even if they will never make a single code change.
Josh Boyer (jwb) Give us your tired, your poor, your huddled masses... More seriously, this is a really hard question to answer.
Mike McGrath (mmcgrath) Fedora should be for enthusiasts, hobbyists and people that are actually doing the work that ultimately becomes a free software product. Every Fedora user could be a contributor.
David Nalley (ke4qqq) What I think you are really asking is 'What is Fedora's target demographic?' The people that Fedora should be targeting is contributors. This isn't to say the Fedora is a 'developer's distribution' necessarily, nor is it to say the contrary. What it does say is that it takes a ton of work, a huge portion of that volunteer, to make Fedora successful. That means, at least in my mind, we should stop trying to nail down where we are in the sea of free software and let the waves of contributors direct us by their actions. If we seek out and grow Fedora contributors we make Fedora more successful. Dennis Gilmore wrote 'I work really hard on those things I believe in.' in his nomination for the board, and I think that sentiment is true in large part all across Fedora. When people care about something, they work passionately at it. Fedora is for those passionate contributors. We're happy that others, like OLPC, Moblin, RHEL, and the millions of Fedora users across the globe like recognize the excellence that those passionate contributors make, but those are largely serendipitous.
Dennis Gilmore (dgilmore) everyone
Kevin Fenzi (nirik) Many folks. Early adopters. Folks who like new tech. Desktop users. Development server users. Many others.
Kevin Kofler (Kevin_Kofler) Everyone (independently of experience level) who wants/needs a Free Software operating system which is as up to date as possible without being unreliable. People who are happy with running old software (be it the non-technical user scared of a menu item moving in their application due to an upgrade or the conservative, "each change needs approval" corporate deployment) are better served by more conservative distributions. It is obvious to me that no single distribution can please both those target audiences at the same time.
Adam Miller (maxamillion) Everyone.
Bill Nottingham (notting) Hm, sort of a repeat of question #7 above. Fedora should be for everyone to use as they see fit, or anyone willing to contribute in its framework. However, I do feel that it should at least have a primary focus, even as other uses, spins, and ideas are allowed. But again, that's probably more a Board issue than a FESCo one.
Jens Petersen (juhp) Every hacker and their grandmother. On the one hand we can't expect everyone will be using fedora tomorrow on the other hand we seen initiatives like Moblin and OLPC that bring the potential of fedora to large new audiences.
Andreas Thienemann (ixs) Contrary to the question above, for whom in my opinion Fedora should be, I honestly do not know the right answer for whom Fedora _currently_ is. We're constantly shifting focus. Once it was claimed that Fedora is giving a preview of what will be available in Red Hat Enterprise Linux, then it was said that we'd be _the_ distribution for the Desktop. At other times Fedora had been promoted as the building block for remixes, spins and other distributions or that it would be the software developers distribution of choice. While all these and other claims were true at the time and are still valid, I am wondering if it is a good idea to try to support such a wide range of users. I have the feeling that the base of dissatisfied users complaining about regressions and other problems is growing. Alternatively, they are at least getting more vocal. The question who Fedora is for, for whom we want Fedora to be and what we have to do to achieve that should be discussed by more contributors though. This is why I'm planning to lead a discussion during the upcoming FUDCon Barcamp in Berlin about exactly that question. You're all cordially invited to attend. The outcome should influence future Board and FESCO decisions.
Seth Vidal (skvidal) Me.
Ian Weller (ianweller)
Christoph Wickert (cwickert) Human beings. I know there is another distribution that claims to be for human beings, but nevertheless: Fedora is for everybody. Seriously. A lot of my customers are using Fedora even if they have not used Linux before. And they are doing pretty well, I have less problems supporting them than many other customers.
David Woodhouse (dwmw2) Isn't this a duplicate of the question about Fedora's raison d'etre?
Return to top


2. Should any steps be taken to make sure releases don't get as much last minute delays as in the past? If yes: which?
Tom Callaway (spot) I think that Release Engineering makes every effort to try to minimize last minute delays, but some are simply unavoidable. With the exception of Fedora 10, which was delayed due to the security intrusion during the release cycle, I think we've had a better handle on what needs to be done to minimize "slips" in our release cycle. Improving the QA and bug triage earlier in the release process is key to minimizing last minute delays, and we're making positive progress in those areas.
Josh Boyer (jwb) Possibly, yes. There is a Fedora Activity Day June 8-10 that will try and address some of this. I am participating in that and look forward to what we can come up with to help here.
Mike McGrath (mmcgrath) More QA in earlier steps, earlier feature freeze. The former of which requires more people.
David Nalley (ke4qqq) This is FESCo's domain, and not a Board issue. From a personal perspective, I'd love to see no delays, and I think some of the current discussions on fedora-devel about length of composes, time spent running updates, etc are beginning to show some of the growing pains that we are experiencing as a project. I look forward to seeing some of the novel approaches to dealing with this.
Dennis Gilmore (dgilmore) As much as we can we should try to always improve how we do releases/ handle blocker bugs. but i fear we will always find last minute blockers.
Kevin Fenzi (nirik) I'm not sure what could be done, aside from trying to slow fedora down. Delays happen. Perhaps we could place more resources on those areas that have caused delays in the past.
Kevin Kofler (Kevin_Kofler) I do not see a concrete need there, the slips have happened for valid reasons, it's better to release a fixed version a bit late than to rush out a brown paper bag release. The delays have all been of a tolerable magnitude.
Adam Miller (maxamillion) Yes and no, it would be nice to have efforts to attempt to have less last minute delays, but the reality of the development world is that many things are unpredictable and we'd rather push back a release date than ship a broken product.
Bill Nottingham (notting) Automated QA. This is probably the most important project that can be done in this regard. Jesse Keating, James Laska, Will Woods and others are working on a framework for this, they'd love the help. Outside of that, we need to be better about more strictly applying the feature freeze dates and not letting features in that land later than those dates. Most of the recent last minute delays are due to features that landed late, but by the time we're looking at these last mintue issues, we're at the point of last return.
Jens Petersen (juhp) I think if we want to keep our deadlines then we need to really following the devel freezes consistently for /all/ packages, particularly base core packages and the installer. Maybe even some core parts should be frozen earlier than the rest of the distro - of course that involves some price in development and innovation but it would lead to more stability. At least IMHO it would be good to have more consistency on this across the distro and less exceptions.
Andreas Thienemann (ixs) While delaying a release is not pleasant it is not neccessarily as bad as people make it out to be. We have a very tight release schedule with a lot of churn between releases and it speaks of our QA that we do catch and are able to fix bugs surfacing that late. There is currently a discussion about this topic happening on fedora-devel with some viable suggestions. While I'd love to see our releases being always on time, I think it is up to the specific teams e.g. release management to decide what can be improved. My understanding is, that the upcoming Fedora Activity Day will be used to better analyze this issue then it can be done during a FESCO election by the candidates. Personally however, I think the Fedora Test Days during this release cycle have been a huge help increasing the stability of our newly developed features and I do hope that we'll increase further testing of rawhide to discover problems early on in a release cycle.
Seth Vidal (skvidal) Not really. We will slip releases. It happens all the time. We should be aware of this and not try to push the deadlines too quickly. The f12-devel cycle is going to be SHOOOOOOOOOOORT.
Ian Weller (ianweller)
Christoph Wickert (cwickert) To be honest I think all things considered we are doing pretty well. If there are release critical bugs, they need to be fixed even if the release gets delayed. Don't get me wrong: I'm not saying we need to do like Debian, but we had releases with severe bugs (e. g. the extrascdinstall bug in FC6) where we should have delayed the release to fix this. We need to make sure that changes in the core parts like anaconda are not done last minute, so they can get wide testing. The test days we had are a good approach, but they need to be earlier in the development cycle.
David Woodhouse (dwmw2) Ideally, perhaps -- but on the whole I'd prefer to have a _good_ release than a _timely_ release. Perhaps not taking it to the extent that some other distributions are notorious for, but the slips we've had aren't something that deeply concern me, per se. What it _does_ show is that we perhaps need a few more people actively testing rawhide -- but we know that, we've always known that, and we're always looking for ideas on how to make it more 'attractive' to potential testers.
Return to top


3. How or to what extent would you say are Fedora's governance bodies responsible for protecting volunteers and volunteer based efforts against interference from within Red Hat by either person or policy?
Tom Callaway (spot) [Disclaimer: I work for Red Hat, but they do not tell me what to say.] In general, I think that in almost every possible situation, Red Hat is content to let Fedora do whatever it needs to do to be successful. The only places where Red Hat steps in are areas where Fedora's actions would construe a legal, financial or security risk to Red Hat as a company, and these are few and far between. I think that outside of those conditions, the Fedora Board (and FESCo) should feel entirely unencumbered by Red Hat's interference, keeping in mind that Red Hat (and its employees) are major contributors.
Josh Boyer (jwb) I don't understand this question. It seems to be predicated on the theory that Red Hat has some nefarious notions in store for Fedora, and I simply don't think that is the case. Fedora has existed for a number of years now, and during that time it has continually become more open and less Red Hat governed. I'd be happy to discuss it further if there were more concrete examples or concerns.
Mike McGrath (mmcgrath) I think it's the FPL's position to fight for Fedora freedom wrt our primary provider. In particular I think our agreement with our contributors should protect each party equally.
David Nalley (ke4qqq) As I said earlier I see the Board (and FESCo and every other governance body) as a guard rail. Ideally they never get in the way and are largely unnoticed by the vast majority. However, they serve an excellent purpose as a safety measure to keep things from going awry. As I mentioned earlier the bond with Red Hat is great for Fedora and at times bad for Fedora, even if it's only in perception. I see the various governance bodies as there to make it easy to contribute and get things done. To the extent that RH (or even groups within Fedora) makes that difficult, I definitely think that it's within their purview to try and fix the situation. I'll note that previous Boards dropped the onerous CLA process from something that often took weeks to complete to something that is click-through now. Another example is FESCo's relatively recent move to make getting sponsored as a packager easier, by limiting the rights that packagers have and creating a new class called provenpackager.
Dennis Gilmore (dgilmore) I think we need to ensure that all people are free to do the work that interests them regardless of who there employer is.
Kevin Fenzi (nirik) I'm not sure I understand the question. What sort of 'interference' ?
Kevin Kofler (Kevin_Kofler) FESCo needs to intervene in case of blatant abuse of power by one person or company, whether the offender happens to be Red Hat or anybody else. And no, the offender is not always Red Hat. But we need to be careful not to try to micromanage: FESCo dictating every single contentious decision isn't going to scale either and it might turn out to cause more problems than it solves. Sometimes a decision needs to be left to the maintainer(s), a balance needs to be struck. But singling out Red Hat "interference" as the source of all conflicts is not going to help, either. Red Hat even gets blamed for decisions which weren't even theirs, e.g. the decision to ship Fedora 9 with KDE 4.0.3 (which I still think was the right thing to do, but that's off topic here).
Adam Miller (maxamillion) I don't like the question, every contributor to Fedora is a member of the Fedora community and singling out those who are fortunate enough to do it for a living, either funded by RedHat, Dell, Intel, or other company should not be singled out. If a community member causes interference then this should be taken up with the Board or FESCo depending on the situation.
Bill Nottingham (notting) If other contributors (inside or outside Red Hat) aren't acting in accordance with the policies or procedures, then the Board (or similar) entities should probably be contacted for mediation when other methods fail. That being said, this question seems like it's leading to a followup, or an example of a specific greivance, which isn't listed here. Without that, it's hard to go into more detail.
Jens Petersen (juhp)
Andreas Thienemann (ixs) I think that it's the job of the Fedora governance bodies to protect our contributors from unnecessary interference from any possible source and let them concentrate on their chosen tasks. On the other hand, I do consider it extremely bad style to go against the interests of our sponsoring company Red Hat and their own product development. The MP3 codec policy would be a good example for this. But this does not mean, that we have to implement each and every Red Hat engineering policy. There's no product management to review feature requests, there's no 7+ years support policy and we're not planning on being certified for china. No need therefore to implement policies for engineering departments with these aims.
Seth Vidal (skvidal) Hahah. Seriously do we really think there is a Red Hat Cabal that is manipulating things from behind the scenes? There is no Cabal. I started in fedora when I was working at Duke University. I came over to Red Hat almost 2 years ago and I can safely say that there is no cabal. I think the goal of fedora's governance bodies is to keep things from being a non-stop screaming match between the various sects within fedora. The governance bodies provide a deciding factor which helps make decisions semi-final.
Ian Weller (ianweller)
Christoph Wickert (cwickert) One of Fedora's four foundations is freedom and we must never give this up. The recent debate about flags in Fedora showed that we really need a policy that assures the work of volunteers is not influenced by the big business. Fedora is not RHEL and not for sale, so we need not and should not follow every advice from Red Hat Legal dpt. But to tell the truth, I'm not too optimistic that we can lay this down in a policy when looking at the current FESCo.
David Woodhouse (dwmw2) I don't understand the term 'interference'. There is a deep symbiosis between Fedora and RHEL. That is the justification for the _huge_ amount of support that Red Hat Inc provides to the Fedora community. Fedora probably wouldn't survive in anything like its current form without a sponsor like Red Hat, and building RHEL would be a much harder job without Fedora. So we 'interfere' with RHEL, and RHEL requirements 'interfere' with Fedora, all the time. It's a productive and happy relationship. I think the governance bodies might be able to step in and object if the RHEL considerations are ever _detrimental_ to Fedora. But I don't really see that happening -- were we _harmed_ because we got an influx of paid developers making Fedora's virtualisation support rock, for example? The most objectionable part of the relationship, in my opinion, is that we're tied to US law; stupid patents and all. But I don't see a realistic solution to that.
Return to top


4. The Fedora Project suffers from a lack of communication. In 2009 for 8 out of 21 FESCo meetings no meeting minutes were sent to the lists. What would you do to improve communication between the different groups in Fedora and especially between FESCo, Board and the community?
Tom Callaway (spot) I can't really talk to the FESCo management. I know that documentation is not fun to do, but it seems to me like there is room for improvement in that space. On the Board, we've worked hard to take good meeting notes with gobby, and then publish those notes on a timely basis for the entire community to see. We also have regular public meetings, and I think they have been extremely productive. In my role as Fedora Packaging Committee Chair, I've worked hard to ensure that as new guidelines are added (or old ones amended), an explanatory email is sent out to the development community, so that no one is surprised later on at an unexpected change.
Josh Boyer (jwb) FYI there is a FESCo ticket opened to fix this very item. As a member of FESCo, I agree we screwed up earlier in the year in regards to meeting minutes and hopefully we'll come up with a good mechanism to get them written and published in a timely manner soon. In terms of the Board communications, I am not entirely sure what, if anything, more could be communicated. The Board discusses confidential material often, and I think John has done a good job with the meeting minutes so far. If I am elected to the Board, I will keep an eye on this issue and make sure that as much information as can be disclosed is.
Mike McGrath (mmcgrath) I think too many people have their eyes in too many projects. If the FESCo meetings were important to someone, they are free to go to the meetings or log them. Meeting minutes don't force people to pay attention and if a contributor doesn't have time to read the meeting log, it's possible they shouldn't be involved with the FESCo workings anyway. I think more of our contributors should focus on fewer things and focus on the quality of those things, not the number of things they're involved in. Group to group communication can be a problem. There are technical fixes to this but also policy changes. I think there's room for better communication between groups, but I feel more accountability towards the group leaders for communication. Poelstra has done wonders in this respect and the release readiness meetings.
David Nalley (ke4qqq) Having been responsible for meeting logs I can empathize with those doing the work that it's not easy to get done particularly with everything else going on in a meeting. That said such communication is vital, but for openness and just general communication. I've noticed that other organizations and even one of our SIGs uses a meeting bot, to record and publish meetings. That would help alleviate the situation you refer to specifically, but doesn't deal with all of it. I've noticed that the Board has engaged a non-member to handle publishing minutes from meetings, and that strikes me as a good idea, but perhaps not foolproof. As a contributor, I started in early 2006 and kept up with mailing list traffic and was on IRC for certain meetings. Until 2008 I never realized how much really went on in IRC. In some ways I think that's very efficient, in others I think it's unfortunate as it essentially hides a lot of the inner workings from people. Perhaps logging and posting minutes of key channels is the answer, but that's also a LOT of content, and I think it would largely go unread. At a minimum I think groups should be publishing at least informal agendas for meetings to the mailing list.
Dennis Gilmore (dgilmore) Look into having a meeting logging irc bot. that will post minutes for us making them consistent between the different groups within fedora.
Kevin Fenzi (nirik) Sadly, this was a failed attempt at spreading the log taking out. We need to be much more clear who is responsible for the logs and minutes, and I think we should require they be posted the same day as the meeting, if not right after. It's usually much easier to just do them as the meeting happens than try and go back and do them later anyhow. I would be happy to just do them moving foward if that's needed.
Kevin Kofler (Kevin_Kofler) Well, for one we need to ensure that minutes actually get sent out, that's kinda obvious. ;-) See the KDE SIG for an example of how it can work. It just needs one or two people volunteering to do the minutes (it's good to have at least two in case one of them is absent), and another person proofreading them (in KDE SIG, that's usually me). But the communication should not be one-way. Community members often bring up valid concerns with some of the planned decisions, which then get at best dismissed, often ignored entirely. This is very frustrating. Concerns brought up on the mailing list need to be taken seriously.
Adam Miller (maxamillion)
Bill Nottingham (notting) FESCo adopted a policy of rotating the meeting minutes and posting of logs. It appears with the occasional absence from the meeting, or forgetting to do the task, it quickly became completely unclear who was doing the meeting summary, and if it wasn't brought up in the meeting itself, it didn't get done. Obviously, that sucks - it's why I took the time a couple of weeks ago to go back through all the old meetings this year and make sure they all had proper summaries on the wiki. I've also created a ticket for the next meeting to fix this going forwards - if need be, I'll do the summaries.
Jens Petersen (juhp) I guess we need a (rotating) Secretary for the meetings to take minutes or summarise the logs within a day.
Andreas Thienemann (ixs) A simple way of having meeting minutes published is to mandate that each unpublished meeting log means having to offer one round of drinks at the next FUDcon social event. I'm sure that this would either result in no missed meeting minutes ever again or a bunch happy fedora contributors. Either way, it's a win-win situation. On a more serious note: Minute notes should be published regularly. I assume that this not being done is an oversight which could be prevented by having more than one person take over responsibility. Maybe alternating from month to month. But I really like the drinks idea. About the communication between FESCO and the community but also between groups: I think it is important to try to include the contributors in the decision making process of FESCO by inviting comments on tickets to a greater extend than before. Controversial topics such as the flag debate on fedora-devel-list created several hundred mails after the policy was announced but went basically unnoticed during the decision phase itself with only a few people chiming in. Publicizing upcoming decisions and other new tickets better and explicitly asking for input from affected contributors might mitigate the resulting flow of messages on fedora-devel-list and make the list useful again. If this is not enough, it might be necessary to revisit the great mailinglist reorganisation of 2007 and see if we could split up -devel into smaller lists, each with a better signal to noise ratio. Maybe trying to restore the maintainers-list could be a solution. In the end, it's a long process. Many things have to be tried and many will be wrong, but I'm hopeful that in the end we'll find something.
Seth Vidal (skvidal) First I'd question your assumptions about the project suffering from a lack of communication. For fesco meetings I'd just say log the irc sessions with a bot - then there's no need for someone to remember to do it. Tada - problem solved - and using new-fangled 'irc bot' technology that was only perfected about a decade ago. For the board the minutes are much much better than they used to be. Unfortunately the board cannot always disclose everything. It is unfortunate but true.
Ian Weller (ianweller)
Christoph Wickert (cwickert) The noise level on some list such as the devel-list or the ambassadors-list is very high and not all contributors have enough time to follow them closely. In order to make sure all important mails reach the contributors we need to make use of fedora-(devel-)announce more often. For the FESCo meeting minutes I suggest to stick with the current model of rotation of minutes takers. But we need to make sure they get written. If someone forgets to send out the minutes he has to pay a big round at FUDCon. This will also help us with question 9 (more fun).
David Woodhouse (dwmw2) We should definitely make sure we publish minutes more reliably. We've been talking about using Gobby to share the burden of doing that; we'll see what comes of it. The IRC logs are always available, and people can even come and join in the meetings (within reason) -- but that's no substitute for proper minutes.
Return to top


5. What would you do to cleanup and organize the Fedora project packaging guidelines, rules and other wiki pages to make it more consistent and easier for new contributors?
Tom Callaway (spot) Can I have 6 extra hours each day? :) This is definitely something that could use some work, and I would encourage anyone who has a suggestion for cleanup or reorganization (without making changes to the actual guidelines) of the Packaging Guidelines to talk to me.
Josh Boyer (jwb) Encourage the people that take interest in that. I'll be honest and say that at a personal level, I can't see me having the time to focus on cleaning it up myself. However I would certainly want to make sure that those willing to do that work aren't hindered in any way.
Mike McGrath (mmcgrath) I don't think that's the boards job, so running for the board I don't feel that I'd get involved unless there were a major issue, like packages no longer getting accepted into Fedora. As far as new contributors I feel more documentation is required on the part of the major groups. Podcasts, video casts to welcome people are one idea I've been kicking around. In particular Fedora is so large that people get overwhelmed and have no idea what to expect. I'd like to see the groups spend more time on new people expectations.
David Nalley (ke4qqq) This is a FESCo issue, though watching some of the involvement with FPC and the Docs Project to accomplish the consistency and ease of use has been interesting. Personally I think this is an issue everyone agrees on, and there are already efforts underway (although surely slowed due to the impending release) to get this accomplished and I think it should run its course.
Dennis Gilmore (dgilmore) Move them out of the wiki into a CMS perhaps, or perhaps look at making them into something that can be printed as a book/pdf.
Kevin Fenzi (nirik) Part of the problem with a wiki approach is that things get slowly modified over time and overall organization gets lost. I think in order to do better in this area we need to step back and come up with a plan for reorganizing things.
Kevin Kofler (Kevin_Kofler) I think that should be best left to the wiki team, which is doing an excellent job, together with FPC, which is responsible and holds the access rights for the packaging guidelines. I see no need for FESCo to micromanage this.
Adam Miller (maxamillion) I think the packaging guidelines and rules are in a solid place and anything that would be modified would be just that, modifications. The wiki has made a lot of progress since I started with Fedora but I would like to push for making it a more used means of collaborative knowledge transfer and to make sure to keep it up to date as much as is possible.
Bill Nottingham (notting) What would you do? There's nothing in being a FESCo member that somehow makes them the only ones able to do this. Anyone is free to bring clarifications or edits to the Packaging Committee. It's not to say the pages aren't a bit long, especially considering there are multiple long pages that need to be read. Ideally, we'd make changes to RPM that make many of the guidelines irrelevant or automatically applied. That's the long path, but it's the one that's going to have the most quantifiable effects. Outside of that, it may be worthwhile to have the design time look over them from a more 'new contributor' standpoint to see how they can be improved.
Jens Petersen (juhp) I think that is something for the Packaging Committee to oversee but some consolidation, reviewing and revisions would be good. I would like to make reviewers' and submitters' lives easier with packages having to pass basic review-o-matic first before human reviewers need to look at them.
Andreas Thienemann (ixs) Ask the teams or persons responsible for these pages to see if they should or can be cleaned up? I do consider FESCO more of a managing body. The FESCO members certainly are doing technical work as well and are in a position to drive this work better forward due to their FESCO involvement. But they are not experts at every task and do not have to have a personal involvement in e.g. wiki cleanup. Taking the packaging guidelines, I can certainly see some points which could be written more concise or cleared up a bit but I'm not sure if this is really going to make packaging easier for new contributors. Just removing rules from the guidelines is not the right way as these rules are there for a reason. Each rule either requires the packager to do things a certain way, either to allow a clear understanding of the spec by mandating coding style etc. or to prevent errors and mistakes. Being a good packager means having experience. People gain experience by packaging software and having these reviewed by an experienced and patient reviewer and sponsor. In my opinion the time spent on coaching new packagers is used more wisely than looking for possible inconsistencies in the packaging guidelines. This will help new contributors more effectively. In general, common sense can be very useful when interpreting guidelines. If there are discrepancies, the responsible team, e.g. the Fedora Packaging Committee is certainly able and willing to correct these without FESCO involvement. If they should be unwilling however, an escalation to FESCO is the right step and they will take care of the issue, one way or another.
Seth Vidal (skvidal) I'd make them shorter and, ideally, make the index read less like a laundry list. finally, it would be nice to have them easier to do searches on.
Ian Weller (ianweller)
Christoph Wickert (cwickert) I am not willing to make compromise about packaging quality, so I don't see much space to clean up the packaging guidelines. But of course the presentation of the guidelines in the wiki could be better. We need everything important in one place. The feature and the spins process should be simplified. Drafts and official policies must be marked as such. Cleaning up the wiki might be possible, but I don't really think it's FESCo's job.
David Woodhouse (dwmw2) In general, I want to keep the bureacracy down -- as I already said. I don't have specific changes in mind at the moment; I think we're doing a reasonable job.
Return to top


6. There is a proposal out there to moderate fedora-devel. What should be disallowed? Racist or sexist speech? Profanity? At what point does a complaint against a small project become a personal attack? ie "Your idea is utterly stupid." Should we disallow trolls or posts likely to start flamewars? Who will decide the difference between intelligent debate and flaming?
Tom Callaway (spot) I think that some people on the Fedora mailing list have gone beyond respectful and reasonable discussion/disagreement in their comments, and unfortunately, were creating a negative environment for people to work in. I proposed that our moderation policies be based around the idea that we should all "be excellent to each other", even when we disagree. I think when people start to make personal attacks against individuals or projects, that goes over the line. You can get the same message across with: "I feel strongly that the idea that you have proposed is not good for Fedora, because foo and bar and baz." as opposed to: "Your idea is F*%$&^% stupid, and only a *&(^%& would think it was good." We'll have to wing it as we go along, but I'm hopeful that this new moderation policy will help people to think twice before they decide to say something that they may later regret.
Josh Boyer (jwb) There is a balance to be struck. Criticism of the project is often one of the best ways to help us grow, but there are times where it can go too far. And I'm not sure you can really quantify and codify what the boundaries are. Frankly, I think even the specter of moderation is likely to help. It can be surprising how often a gentle reminder to be polite can work. I also think that as a community we should trust our elected leadership to do the right thing. If they aren't, and there are egregious examples of abuse by the leadership, I'm fairly confident that they can't be censored from the internet completely and they'll be pointed out. I am not worried in the slightest about this.
Mike McGrath (mmcgrath) I'm against censorship in any form. If trolls want to be trolls, people can ban them on their own via mail filters, etc.
David Nalley (ke4qqq) Wow that's a lot of questions - I'll be honest and say I really dislike the idea of moderation, and do so for several reasons. Censoring some or all 'bad speech' is a zero sum game. And it's always going to infuriate someone. The worst offense that censorship brings is waste. Some of the most talented, hardest working people in Fedora are the same people who are likely to have to moderate, what a waste of precious cycles. At initial blush, I'd almost guess that something akin to greylisting might be employed once a thread reaches some arbitrary number (100 is the number sticking in my head). So essentially once a thread hits that arbitrary number, messages for that thread are rejected for 36-48 hours. This doesn't directly solve the problem of people not 'being excellent to each other' but does make the medicine a bit less painful, and hopefully automated. Every instance of 'bad behavior in recent memory seems to have occurred in massive threads that can't seem to die. Perhaps that's a stupid suggestion, after all it's 1am, and I'm still up working on these questions. Realistically there is no good solution to people behaving badly in a volunteer community, you are essentially left choosing the difference between multiple evils.
Dennis Gilmore (dgilmore) I think there is no right or wrong answer here. people being rude or argumentitive need to know that its not there right to be that way with others. I think sometimes we need to think if what we are about to send is something that would be considered appropriate if that person was in front of us. a little more thought about our actions is a good thing.
Kevin Fenzi (nirik) I am thinking we should not do moderation on that level. Instead we should make fedora-devel a better place by becoming more technical and usefull. When someone makes a post thats an attack or other, we should stop responding to them. I think we can lead by example here, as well as stopping feeding the trolls and non productive folks. I note that since this proposal, the list has been a good deal more tame. Is it the threat of moderation? People realizing they were not adding to the discussion? Or just more techincal on-topic posts? I'm not sure, but I think moderating is a pretty slippery slope.
Kevin Kofler (Kevin_Kofler) It's kinda hard to moderate a mailing list because a mail can't be deleted or edited once posted (unlike on a web forum). (I don't think running all mail through manual moderation before it gets posted is an option.) I'm not sure it's worth it, as the real problem the fedora-devel-list has is not the tone of discussion (there have been a few abusive posts, but those were far from the majority), but the volume. And reducing the volume without blocking legitimate discussion is kinda hard because it's not always obvious whether a given post is on topic or not. And I don't see why things like "Your idea is utterly stupid." or posts which merely contain the "f-word" without insulting anybody should get moderated, I think those are perfectly harmless. Racism, sexism or direct insults towards a person (not their software) are clearly not acceptable and can be moderated, I just don't think this alone will reduce the list's volume in a significant way.
Adam Miller (maxamillion) I think something like this should be handled similar to how most irc chat rooms are handled. The community should elect or appoint (depending on how the structure is pefered or setup) moderators and those moderators' judgement should be trusted. If in the event an user feels they were not in the wrong after being banned, they can go infront of the Board (or alternatively a Mailing List Committee)
Bill Nottingham (notting) The Board's provided policy actually spells out the answer to most of your questions here (what's disallowed, who is the arbitrator, etc.). Honestly, I think that policy is a good start, and doesn't need any tweaks yet.
Jens Petersen (juhp)
Andreas Thienemann (ixs) The temporary moderation of people is not only a proposal, it was passed by the current FESCO. I'm uncertain if it is really going to improve fedora-devel-list. I'd suggest to let some time pass and then review the decision. See if the tone on -devel-list got friendlier, see how many messages/people got moderated, see if the enacted moderation seemed reasonable in hindsight. If the answer is not positive, then it's time to abolish the rule and look for something better. In general, I'm very wary when it comes to moderating a mailinglist and I'd like to see any such policies withdrawn sooner than later. Regardless where you draw the line to moderating a message, it is censorship. And censoring other opinions is a really bad idea. It's not only wrong on a moral level, it also cultivates groupthink. Challenging the way we are doing things or thinking about issues is very important. If the tone is not what you'd expect, mention that. It usually helps. And if it doesn't there's always the ability to filter your mail locally. Any filtering should ever be done at the client as it's your own blacklist and it's not imposed by a third party. To come to the second part of the question, where to draw the line: Again, I'm a strong proponent of using the skill called common sense. I understand that it is becoming increasingly rare (compare <http://filepile.dicp.de/common_sense.jpg>) but it is a much better way of dealing with the issue of where to draw any line than trying to define it in rules. The problem with fixed rules is that someone will find a way around them. This would mean implementing another rule to fix the earlier omission or create an exception and as can be seen by governments all over the world: layering rules on top of other rules doesn't really do much good. So much for the general idea, applying this to mailinglist moderation in case it is absolutely necessary (I wouldn't know why this should be the case though): I do believe that the decision on what to block and what to let through should be left to the moderators. It is my hope that they will act using their common sense. As this cannot be guaranteed, I would implement a defined process of appealing such a moderation. If there are too many valid complaints, we clearly need a better moderator. Checks and balances it is commonly called.
Seth Vidal (skvidal) I think the board has already decided this. There will be one and we will have a group of folks working to moderate behavior that is way beyond the pale. Moderation will temporary provided the person settles down. Profanity is not at issue. It's personal attacks and truly aggressive behavior that is the problem.
Ian Weller (ianweller)
Christoph Wickert (cwickert) When I first read of this proposal on the board's list I thought it was a joke, but obliviously it is not. Mike said it straight to the point: "Oppression sucks. Censorship sucks. Board members spending their time on jerk patrol sucks." On the other hand I do see the problem very well, but I think we all need to apply common sense here. If we started moderating people or even unsubscribing them, we will run into bigger problems, because there is nobody to define when a factual discussion ends and personal attacks begin. We are trying to solve a social problem with a technical solution, which is doomed to fail. We are lucky to have a large community and IMO it is perfectly legal to have different point of views. I'm sure we all love Jef 'best quotes ever' Spaleta statements, they are the icing on the cake.
David Woodhouse (dwmw2) I dislike moderation, and I don't see the point. If you don't want to read the flamewar, don't. Doesn't your mailer have a 'delete thread' button? The only thing I'd ban is posts lacking a correct In-Reply-To: or References: header associating them with the thread to which they're replying :)
Return to top


7. Please give three examples of other boards or communities you have participated in and the positive differences you made there.
Tom Callaway (spot) I'm currently a seated member of the Fedora Board, and I have previously served on FESCo. In addition, I'm leading the Fedora Legal effort, and helped to build the Licensing guidelines that help ensure that Fedora remains committed to Free Software. I'm also the chair of the Fedora Packaging Committee, and have helped create the Fedora Packaging Guidelines.
Josh Boyer (jwb)
  1. Rel-eng: I have generally tried to help rel-eng overall, focusing lately on updates pushing and powerpc.
  2. FESCo: I've been on FESCo for quite some time now, and I can't think of a singular instance of making a huge impact. I'm OK with that, as FESCo is really about the long term impacts.
  3. Fedora user: I've been a Fedora user since FC1 and have tested and helped with a number of bugs along the way. And yes, this counts. The Fedora user community is one of the most important communities we have :).
Mike McGrath (mmcgrath) I've never been involved in another board and my first community experience was with Fedora. I've committed my time to it. I've communicated with many other projects, like SuSE for example and smolt. The key is to find common goals and work towards them. Be polite, empathetic, and understand there are multiple ways to do something. Don't let the perfect be the enemy of the good.
David Nalley (ke4qqq) • NA Ambassadors - Along with Clint Savage, John Rose, Brian Powell and many others helped drive a resurgence in the NA Ambassador Community. We went from leaning (some might say leeching) on Red Hat to produce our swag, send RHT speakers, etc to truly driving the schedule, doubling and in some cases tripling our appearances at LUGs, BarCamps, and conferences, filling speaking slots with community members rather than contributors from RH who are paid to speak. We also greatly improved our swag efficiency and handle making all of it. UCLUG - Steering committee member. Along with others pushed a revitalization that in some cases increased our attendance by an order of a magnitude. Creating content that appealed both to new users and graybeards alike. Pushed multiple community outreach programs. Southeast Linuxfest Foundation - Secretary of the Foundation - organized (well still in the process really) and help build a regional Linux conference in the Southeast that appears will successfully happen on June 13th. Blue Ridge Amateur Radio Society - multiple term vice president - led community outreach with non-technical groups as well as government and other radio groups. Grew organization to the largest of its kind in the state, and putting on the largest amateur radio fest in the state.
Dennis Gilmore (dgilmore) OLPC, Spacewalk
Kevin Fenzi (nirik) Well, I am quite active in #fedora trying to help users there. I have also taken to trying to help out people on fedoraforum. I am also active in some of my local LUG groups (nclug.org and lug.boulder.co.us).
Kevin Kofler (Kevin_Kofler) I haven't been in a steering committee yet, unless you count KDE SIG as one. The largest non-Fedora community I have been (and still am) involved with as a developer is KDE: I saved Kompare from bitrot and removal in KDE 4.0.0 by completing its KDE 4 port at the very last moment. (It only took me one night to do the port, I would have done it sooner if I had realized that nobody else was working on it.) I am now Kompare's official maintainer. I also use my commit access from time to time to upstream fixes for issues discovered during Fedora KDE packaging. Other team projects I participate in include TIGCC [1] (of which I've written a sizable portion of the code) and TiEmu [2] (where my main achievement was integrating GDB and the Insight GDB frontend into TiEmu to allow debugging C programs from within TiEmu). As you can see, I'm coming from the development side. However, I don't see this as disadvantaging or disqualifying me: I believe a clear understanding of the subject matter is more important than experience with sitting on some unrelated committee.
Adam Miller (maxamillion)
Bill Nottingham (notting) I've been involved in Fedora since before the beginning, including serving on the Fedora board itself. I'd hope that I've made a positive difference.
Jens Petersen (juhp) I started contributing to XEmacs and Emacs (I created find-func.el). I started the "Haskell Fedora" packaging project which was merged to fedora-haskell sig. At Red Hat I have been contributing to input-method development (reworked gimlet for iiimf, scim lang menu, etc). I have contributed patches to quite a number of upstream projects.
Andreas Thienemann (ixs)
  • Management board/CEO of a not for profit Internet Service Provider. Implemented new processes to improve technical service quality, increase service/product offerings and increase the efficiency of the technical staff. Result: The ISP is still in business while nearly all comparable non-profit ISPs in the German market have failed. Added benefit: I did this all having lots of fun and even learning a few new things. \o/
  • Teamlead during a project to wrestle control of the communication infrastructure for a residential accomodation (~500 appartments) away from the non-performing communications provider. This comprised a multi-year drive to rescind two multi-decades contracts, coming up with contingency plans as well as a long term alternative and convincing stakeholders of supporting this alternative. Result: Out with the old, in with the new, self-hosted communication infrastructure which is cheaper than before and performing way better.
  • Delegate to various European gatherings about defining better Drug/Tobacco policies and lobbying governments into adopting them. Result: Lots of new friends all over Europe. The laws in question are unfortunately still not fixed. Added benefit: I learned how to lead people, and to better stay away from too much politics. It's better for one's mental health.
Seth Vidal (skvidal)
Ian Weller (ianweller)
Christoph Wickert (cwickert) I have joined the LXDE community and foundation recently and it seems like I'm the most active distro package maintainer. This is a success for both Fedora and LXDE as the large interest at Chemnitzer Linuxtage showed. Two important community members from LXDE have joined Fedora afterward and even the lead developer switched to Fedora. As a packager and ambassador I have managed to recruit other contributors for Fedora and some of them have become quite important for us, e. g. Sebastian Vahl, who is doing an amazing job in the KDE SIG. I'm interested in politics and member of several NGOs. Please forgive my for not going into details, I don't want people to judge me by things that don't belong here.
David Woodhouse (dwmw2)
Return to top


8. Simple question: Among the two, do you prefer Gnome or KDE?
Tom Callaway (spot) For most things, I prefer GNOME, although, I do keep both installed on my primary workstation.
Josh Boyer (jwb) I don't prefer either really. I find likes and dislikes with both. This is not a political answer either. On my machines, I typically run the default Gnome install, but use some KDE applications such as k3b and amarok. I find that I reinstall machines (for testing!) too often to really bother changing defaults or being tied to a particular environment.
Mike McGrath (mmcgrath) KDE.
David Nalley (ke4qqq) vi - ohhh wait that's the next question. At the moment I am using Gnome, though I have been one of the KDE-faithful for most of my Linux desktop 'career'.
Dennis Gilmore (dgilmore) KDE
Kevin Fenzi (nirik) Xfce. ;)
Kevin Kofler (Kevin_Kofler) KDE, of course! :-)
Adam Miller (maxamillion) In the current state of Gnome and KDE, to answer the exact question: KDE But having said that, I prefer Xfce over either of them.
Bill Nottingham (notting) I use GNOME. Why do you ask? (As litmus issues go, it seems somewhat silly.)
Jens Petersen (juhp) I use GNOME, but I am not a heavy desktop user - I mostly live in Emacs, browser, virt-manager, and xchat-gnome.
Andreas Thienemann (ixs) Simple question, simple answer: I don't. Longer answer: I'm not the typical desktop user. I'm only using x-terminals, a graphical webbrowser and an IM client. If there'd be a good browser for the framebuffer, I'd probably be most of the time perfectly happy without X11. As such I do not really care about the installed desktop as long as it's not getting in the way of my work too often. Currently I'm using Gnome as it's the default on Fedora but I have quite a lot of KDE developers among my friends and have been mentioned in their credit file as well. Maybe I'll give another window manager a try some time, a friend and fellow Fedora contributor has been all the rave about a tool called awesome.
Seth Vidal (skvidal) I don't really care. I need about 4 applications to get my job done. If both Gnome and KDE died in a horrible code disaster I'd just open my terminals and a firefox window and move along. I use gnome for the most part b/c I have a set of keyboard shortcuts already defined for it and I'm used to how the window popups behave. I'm certain that I could make kde behave the same way in about a day if I had to. I'm positive I could do the same with xfce or lxde or whatever-de is out there. I think both gnome and kde are great. I just don't care that much anymore.
Ian Weller (ianweller)
Christoph Wickert (cwickert) None. I use Xfce and LXDE. (Smart answer, isn't it?)
David Woodhouse (dwmw2) I am unqualified to answer this question, because I haven't used KDE for about ten years. I _used_ to use KDE, but switched to GNOME about that long ago. I'm increasingly frustrated with what I referred to above as the 'GNOME disease', and Evolution makes me want to bash my head against the wall almost daily. I'm almost tempted to switch back to KDE to see if I hate it any less.
Return to top


9. Do you think a Fedora Foundation is still worth pursuing? Why or Why not?
Tom Callaway (spot) I do not think it is worth pursuing at this time. The problems that originally caused us to stop pursuing that option still exist:
  • For patents, OIN does that job for us, and Red Hat is already a member so Fedora gets those benefits.
  • For legal coverage, Red Hat Legal are experts in the fields that Fedora needs coverage in. A independent foundation makes it far more difficult for us to directly leverage their expertise.
  • For financial reasons, a US non-profit cannot take the majority of their funding from one source, and it is very unlikely that Fedora could be self-sustaining in the near future. Our hosting and bandwidth costs alone are very significant. I also do not feel that Red Hat has interfered with Fedora in any significant way, and that the Fedora Board remains the primary decision maker and governance body for Fedora.
Josh Boyer (jwb) I have no insight to the original pursuits so I'm not very familiar with the problems that were encountered. It would be a bit silly of me to state an opinion without having some amount of knowledge on the issue.
Mike McGrath (mmcgrath) I wish we had a Fedora Foundation though the details of that are fuzzy to me. It would be convenient to have a separate legal entity to Red Hat.
David Nalley (ke4qqq) I do think it is. The real benefit is the efficient handling of money. A number of initiatives within Fedora hinge on our being able to deal with money, and those initiatives (such as the Fedora store) are routinely blocked as legally Fedora is effectively a non-entity, or at best the spawn of Red Hat. I don't think, unless things changed dramatically, that there is a possibility for a US-based non-profit entity, but there are a number of different strategies that might be available. Fedora EMEA e.V. makes tremendous use of it's organization. There are a number of thorny issues, from financial, to trademark, to even the relationship with Red Hat that need to be resolved and researched. That said I think that continued exploration of this matter is important, and could be a boon for Fedora in many ways.
Dennis Gilmore (dgilmore) I think it would need to be evaluated if we were going to get resources donated on the same level as Red Hat does from other companies. Until we have that level of investment then its extra overhead for little gain
Kevin Fenzi (nirik) No. I think it would not help any, and might loose us support from our biggest supporter.
Kevin Kofler (Kevin_Kofler) It might be worth pursuing a Fedora Foundation outside of the US if that allows us to avoid patent restrictions. But assuming it wouldn't (which is likely because Red Hat might still be liable for contributory infringement), I don't think a foundation would change much, so in that case the answer is "no".
Adam Miller (maxamillion)
Bill Nottingham (notting) In the US? Almost certainly not; the funding requirements alone make it pretty much a non-starter. Having a Foundation elsewhere depends on the cost/benefit ratio; does the ability to accept cash donations really enable that many more things that couldn't be done now that would justify the legal expense, the trademark agreements, and other costs? I'm not convinced yet. (Then again, this is sort of an odd question for FESCo.)
Jens Petersen (juhp) Maybe sometime but currently it does not look a priority to me.
Andreas Thienemann (ixs) The word from Red Hat was, that the Foundation will not be further pursued as it would actually be detrimental to the current involvement of Red Hat in Fedora. As far as I know, circumstances haven't changed, so I do not see any reason to further pursue this idea. If an organisation is needed to help organizing events or act as a legal body for contracts etc. I had the impression that non-profit associations such as Fedora EMEA e.V. are doing a fine job as is and mostly alleviated the need for a Foundation. Anyway, I do believe that this question would be better posed to a board candidate then a FESCO candidate.
Seth Vidal (skvidal) only sometimes. It's a world of pain - but it could be good to unhinge certain obligations. I'm not settled on it and I doubt seriously it will ever happen.
Ian Weller (ianweller)
Christoph Wickert (cwickert) If possible yes, but I personally would not invest to much work here. I like the model of regional leadership and we already have some local entities that are doing very well. I think we should rather follow that approach.
David Woodhouse (dwmw2)
Return to top


10. Should kernel module packages be allowed in Fedora?
Tom Callaway (spot) Nope. I think that there are many reasons for this, but the primary one is that it encourages people to not get that code merged into the upstream Linux kernel, where it truly belongs.
Josh Boyer (jwb) No. The code belongs in the upstream kernel, where it would get into Fedora by default.
Mike McGrath (mmcgrath) I have no opinion on this except that if the module is good enough for Fedora, it should ultimately be sent upstream.
David Nalley (ke4qqq) This is a FESCo issue and not a Board issue. I personally wouldn't want to see kernel modules in Fedora's repos.
Dennis Gilmore (dgilmore) no they should not, we should get them into fedora's kernel package and upstream. If there is little to no chance that a module will be acceptable upstream i dont see why it should be allowed in fedora as an out of kernel rpm module. keeping a module in sync with fedora kernels is alot of work and leads to many broken deps and frustrated users.
Kevin Fenzi (nirik) No. If they are not good enough for the upstream kernel, why should they be good enough for our users ? I think kmods have their place in rpmfusion and other 3rd party repos, but I hope someday they will all be gone and no longer needed.
Kevin Kofler (Kevin_Kofler) Yes (assuming the license of the module is compatible with the kernel's GPLv2 license). There should be no restrictions other than this and the restrictions applying to all Fedora packages. We allow plugin packages for any application and I don't think anybody would seriously consider banning them (I certainly won't!), so I don't see why the kernel has to be special. The counterargument that coordinating the updates with kernel updates is hard is not a good one because having the modules in Fedora would allow using grouped pushes in Bodhi, whereas the current ban relegates the modules to third-party repositories where coordination with kernel updates in Fedora will always be a problem.
Adam Miller (maxamillion) No. They are difficult to maintain and the Fedora kernel contributors time is better spent on upstream kernel components.
Bill Nottingham (notting) No. It adds to the kernel maintenance, doesn't push the proper message for getting drivers upstream, and never stays up to date with the current kernel (causing users to get 'stuck' on older kernels). Drivers actually headed upstream should be added to the main kernel packages; drivers not actually headed anywhere upstream are a bad idea to include.
Jens Petersen (juhp)
Andreas Thienemann (ixs) That would be topic which should be decided together with the Kernel maintainers as it falls under their responsibility. The last time I checked, they were pretty much set against the idea of carrying out of tree modules. Thus I think there is no need to revisit this issue, nothing has changed. Furthermore, I do not think there's a need for Fedora to ship kernel modules and compromise our follow-upstream policy: RPMfusion does a pretty good job shipping kernel modules for a variety of needs. This third-party repository explicitly does follow our own packaging guidelines so any fedora packager can easily start creating kmod-packages which will work fine on Fedora. Problem solved.
Seth Vidal (skvidal) no.
Ian Weller (ianweller)
Christoph Wickert (cwickert) Yes. IMO it was a mistake to remove them, but the final word i snot yet spoken. A decision needs to involve FESCO, FPC and the kernel guys.
David Woodhouse (dwmw2) Absolutely not. If it's not good enough for upstream, it's not good enough for Fedora.
Return to top

Notes

Some experiences and notes that might be helpful for future Candidate Questionnaires:

Workflow:

  • aim to collect answers in private and publishing them in one go -- then candidates have no chance to look at the answers from other candidates; that sounds like a minor detail, but seems to have helped a lot to encouraged the candidates
  • to help with that only add candidates answers to the official answer documents/pages that were handed in before the deadline
  • prepare the questions early and have them ready and public in the wiki early -- ideally at the beginning of the nomination period or something like four days or one week before the end of the nomination period; set a tight deadline for handling in the answers like "end of nomination plus two days". That should make sure the whole election process doesn't take to long and give everybody that nominated enough time to sent the answers -- sure it's tight if people nominate late, but that's their fault ;-)
  • to help with that make sure the deadlines are known before the nomination period starts
  • seems a lot of people liked the OpenOffice table for comparing the results we had in the past, as it's possible to easily hide candidates and answers/questions you are not interested in; to reduce copy'n'paste (and thus reduce error potential) consider to put such a basic table with the answers (one per row) into the wiki that candidates (one per column) need to use to hand in the answers; then all that is needed is to merge the rows with the answers into one table, which is quite easy; from that table it's easy to export them into many other formats without too much work

Questions:

  • "open ended" questions obviously are highly preferred
  • someone needs to review the questions and merge similar questions into one and remove others to make sure it are not to much questions (something like 16 to 20 likely should do)
  • some of the old question were quite good and universally;
  • some people dislike questions like "do you prefer Gnome or KDE" that are/should be mostly irrelevant for Fedora as whole and the position the candidate is nominated for, but others people really like the answers as they give a impression about the person itself. So a few of those question are good and definitely acceptable, but "few" is likely important