GNOME leadership [was Re: So what do people *except* me want from the foundation?]

[Luis Villa <luis tieguy org>]

[Apologies in my lack of interaction in this thread; I've just started
studying for the bar and have just moved to a place with no internet;
the combination has left me pretty brutally offline all week, as I
will be for most of the rest of the summer.]

2009/6/2 Jason D. Clinton <me jasonclinton com>:
> On Tue, Jun 2, 2009 at 7:25 AM, Dave Neary <dneary gnome org> wrote:
>>
>>   What do you expect from the foundation?
>
> Leadership. I want there to never be another DVCS mutli-year long flame war. The only reason it ended is that Red Hat has the people, servers and bandwidth to JFDI for something of that magnitude. That worked that time. But Red Hat shouldn't be forced in to taking three-four employees off their other responsibilities to prevent GNOME from tearing itself apart. We need a way to make authoritative decisions in a healthy way and then to share the responsibility of making it happen without giving the appearance of back-room dealings or rule by fiat.
>
> Consensus building and making travel happen (to affordable locations) are the only two things I want to see the Foundation doing.

[Note that you're talking about two different things when you're
talking about leadership and consensus-building/decision-making
apparatus. I will address leadership here.]

The useful question is probably not 'why is the Foundation not
leading' or 'how could the Foundation lead'; I think probably the
better question is 'why is no one from the community leading'?
(Because you really don't want leadership by elected committee.)

The Foundation may be part of the cause of that. It certainly has
sucked up time and energy from a lot of dedicated people, helping
contribute to burnout without much to show for it. I am not sure how
to avoid that, though. Is the existence of the board hurting in other
ways? It seems some people look to it for leadership, which might hurt
the emergence of other possible leaders. This was not the case when I
started working on GNOME, but may be the case these days- dunno.

Our historic inability to stop bikeshedding is probably part of the
problem; leadership when people are nipping at your heels all the time
is not fun or interesting.

Our supporting corporations have helped contribute to this, probably;
they've hired our best and brightest (good) and made them work on
corporate priorities (less good) or work in private or on projects
related-to-but-not-really-of-GNOME (even less good). That saps
momentum, energy, etc. We've never really had something equivalent to
TLF or Transmeta or Google, allowing our best and brightest to work on
GNOME 90-95% as they saw fit, allowing them to be leaders. All the
partner companies have at times allowed things close to this (Jeff at
Ubuntu, Miguel at Ximian, several RH folks over the years, etc.) but
nothing like kernel has had, or even (as best as I can tell) as KDE
has had the past 4-5 years with aseigo. Maybe a good question to ask
would be 'why is this'?  Is it that the corporation's goals aren't
closely aligned enough with ours? Is is that we've burned them out,
and they've then let the companies pull them away from GNOME? Is it
that because of the bikeshedding (or other reasons) they just think
that dealing with GNOME is more trouble than it is worth, so they
won't invest in our leadership? (Moblin's development pattern suggests
this might be the case.) Many theories to explore here.

At any rate, I agree completely that we need some strong leaders to
develop in GNOME. But the Foundation is not the place for it. I think
the right question is 'why have leaders not come from other sources?
what can the Foundation do, if anything, to help other leaders emerge
and get the support they need to do their work?' I have no easy
answers to either of these, though.

=====

Two more marginal observations on leadership in the Foundation context:

First, historically, the Foundation doesn't lead or make final
decisions because the Foundation and the board were very explicitly
told by the community, through our charter, *not* to lead. The board's
role was infrastructure, support for existing leaders, and
communication with sponsors. This was primarily, as best as I can
tell, in order to prevent the Foundation's sponsors from exerting
control/leadership of the community through the Foundation- if they
were going to do it, they were going to have to do it the
old-fashioned way, by hiring hackers ;)

I'm not sure this argument makes sense anymore, but everyone involved
should understand this is the single biggest reason why the Foundation
does not provide leadership. (It does not explain the failure of other
sources of leadership.)

Second, leadership in a volunteer/multi-corporate context is a tricky
problem. The git thing is a perfectly good example of this. If we'd
had a magical leader who wisely sat down and said 'we will use ____',
then what? I'm highly skeptical that magical 'leadership' could have
done much better than what we ended up with- they still would have had
to convince a lot of skeptical people, choose to work either with or
against certain corporate partners, and find the resources to do a lot
of hard, unpleasant work. This sounds familiar, no? It is of course
true that having the mythical strong, decisive leader might have
cleared some of this up very slightly faster, and allowed them to use
some 'us against them' rhetoric to marginalize the other side more
quickly, but you can only go so far when that leader doesn't have
firmly committed resources at hand, especially if the leader wants to
not be a manipulative dirty SOB.

Luis