Open Letter: Why I, Kevin Kofler, am not rerunning for FESCo

Mon May 3 10:53:20 UTC 2010

On Monday 03 May 2010 02:20:51 Kevin Kofler wrote:
> Hi,
> 
> You will have noticed by now that my FESCo term is about to expire, that
> the nomination period for FESCo just closed and that my name does not show
> up on the list of candidates. No, this is not an accident or negligence,
> the decision not to run for another term was intentional, for several
> reasons:
> 
> * When I ran for election a year ago, one of my reasons for running, and
> also something I made part of my campaign, was that it shouldn't always be
> the same people who are sitting on FESCo. We have a much higher number of
> active contributors than FESCo seats, so it makes sense to see some
> turnover happening. So it would be very hypocritical from me to attempt to
> sit another year on FESCo myself, now that I'm myself a FESCo "veteran".
> 
> * I have never been a committee person and have always hated sitting on
>   meetings. I have done it anyway for a year because I believed it to be
>   important for the good of the project. But I'm really fed up of those
> meetings (I'm feeling burned out) and prefer focusing on more practical,
> less political areas of Fedora. The fact that I don't feel my presence in
> those meetings being of much if any use (more on that later) doesn't help
> either.
> 
> * When looking back at what happened over the year I've been in office, I
> have a feeling that I have been able to acheive basically nothing:
>   - The vast majority of votes were either unanimous or 8-1 against me. In
> both cases, my vote was entirely redundant. Even for more contested votes,
> my vote hardly ever mattered.
>   - Any attempts to discuss those issues where everyone was against me went
>     nowhere. In most cases, people rushed out a vote without even
> considering the real issue at hand and then shot down any discussion with
> "we already voted, we want to move on". In those few cases where there
> actually was a discussion, my position was always dismissed as being
> ridiculous and not even worth considering, my arguments, no matter how
> strong, were entirely ignored.
>   - Basically any proposal I filed was systematically shot down. Even
> things which should be obvious such as:
>     . calling GNOME by its name rather than the generic "Desktop" or
>     . eliminating the useless bureaucratic red tape of FESCo ratification
> for FPC guidelines which just wastes everyone's time and constitutes pure
> process inefficiency
>     got only incomprehension.
>   I have come to the conclusion that it is just plain impossible for a
> single person to change FESCo's ways and that therefore I am just wasting
> my time there.
> 
> * I am very unhappy about FESCo's recent (and not so recent, which were
> what made me run in the first place) directions. The trend is steady
> towards bureaucracy and centralization:
>   - Maintainers are continuously being distrusted. It all started with the
>     provenpackager policy, where every single provenpackager has to be
> voted in by a FESCo majority vote, as opposed to letting any sponsor
> approve people as provenpackagers as originally planned, or just opening
> all our packages to everyone as was the case in the old Extras. From
> there, things pretty much degenerated and we're now at a point where FESCo
> no longer trusts maintainers to know when an update to the packages they
> maintain is stable, instead insisting on automatically-enforced
> bureaucracy which will never be as reliable and effective as a human. The
> fact that we trust our maintainers used to be one of the core values of
> the Fedora community. It has been replaced by control-freakiness and
> paranoia.
>   - All the power in Fedora is being centralized into 2 major committees:
> the Board and FESCo. FESCo is responsible for a lot of things all taking
> up meeting time, leading to lengthy meetings and little time for
> discussion. Many of those things could be handled better in a more
> decentralized way. Power should be delegated to SIGs and technical
> committees wherever possible, FESCo should only handle issues where no
> reponsible subcommittee can be found or where there is disagreement among
> affected committees. In particular, I suggest that:
>     . FPC guidelines should be passed directly by FPC, only concrete
> objections should get escalated to FESCo.
>     . membership in packager-sponsors and provenpackager should be handled
> by the sponsors, with a process to be defined by them (my suggestion:
> provenpackager should take 1 sponsor to approve and no possibility to
> object or veto, sponsor should take 3 sponsors to approve and objections
> can be escalated to FESCo).
>     . features should get approved by the responsible SIG or committee
> (e.g. FPC for RPM features, KDE SIG for KDE features etc.). The feature
> wrangler should decide on a SIG to hand the feature to for approval, or
> even accept features filed directly into "approved" by the responsible
> SIG, and FESCo would be responsible only where there is no clearly
> responsible SIG, or to arbitrate when a SIG is trying to make a change
> which affects other SIGs without their consent.
>     Unfortunately, these suggestions are falling on deaf ears, in fact I
> filed the first suggestion as an official proposal (as it looked very
> obvious to me, the ratification process is pure bureaucracy) and it was
> shot down (also due to the FPC chair claiming FPC doesn't want this,
> despite at least 2 FPC members having spoken out rather favorably). I
> think a more decentralized approach (in general, not just for FPC
> guidelines) would be more efficient, more democratic, less bureaucratic
> and less corporate and would increase overall maintainer happiness by
> reducing the impression of the "diktat from above".
>   - The prevailing opinion of the electorate of Fedora contributors keeps
>     getting ignored. Feedback on the Fedora devel mailing list is never
> seen as in any way binding, it's often dismissed as noise or "trolling".
> The predominant opinion in FESCo is "you voted for us, now we get to do
> whatever we want", which is flawed in many ways:
>     . It assumes there were true alternatives to vote for instead. This
>       assumption does not look true to me.
>     . It assumes the voters were aware of the positions of all the
> candidates. I'm fairly sure this was not the case. While I appreciate what
> has been done in an attempt to solve this issue (questionnaire,
> townhalls), this has proven by far insufficient to build an opinion on the
> candidates. I think there's a reason representative democracies normally
> work with parties/factions and I think something like that might help a
> lot, depending on what kind of factions would show up.
>     . It assumes representative democracy is a well-working model in the
> first place, especially in its most hardcore form ("now we get to do
> whatever we want"). I believe elected representatives should really
> REPRESENT the people who voted them. I realize politicians aren't doing
> that, but are they really a good model to follow?
>     I believe listening more to the feedback on the devel ML and taking it
> into account during decision-making would reduce frustration with FESCo a
> lot. - The prevailing opinion of Fedora users keeps getting ignored. See
> e.g. Adam Williamson's poll about the kind of updates users expect from
> Fedora, its clearcut majoritarian result, and FESCo and the Board both
> planning to do the exact opposite.
>   - Common sense is just generally lacking, see e.g. the decision that the
> GNOME spin should continue being called "Desktop Spin", despite evidence
> that this is confusing many users, both the ones actively looking for
> GNOME and the ones who want some other desktop. And that's just one such
> nonsensical decision, the one I remember best because this is an issue I
> care much about.
>   I do not wish to stand for such a committee anymore (in fact I probably
> should have resigned much earlier, as I've just been frustrated and burned
> out for more than half of the term, but I didn't because my feeling of
> responsibility was too strong) and, as pointed out before, I feel
> powerless to change anything.
> 
> Therefore, I will stay in office until the end of my term, but I will not
> be available for reelection. I would like to thank the people who voted
> for me last year for their support and apologize to those who would have
> liked to vote for me this time for not giving them this opportunity. If
> you would like a KDE SIG person in FESCo, vote for Steven M. Parrish (and
> vote for Rex Dieter for the Board). But if you want to see the kind of
> change to FESCo I'd like to see, it'll take a faction of at least 5 people
> to make it happen.

Kevin,
thanks for your (Don Quixote's) work in FESCo! You had a lot of supporters 
here in the office! It's sad you don't want to fight windmills anymore but I 
understand your reasons. 

Jaroslav

>         Kevin Kofler

-- 
Jaroslav Řezník <jreznik at redhat.com>
Software Engineer - Base Operating Systems Brno

Office: +420 532 294 275
Mobile: +420 602 797 774
Red Hat, Inc.                               http://cz.redhat.com/