FESCo Meeting Summary for 20090424

[William Jon McCann]

Hi Jesse,

On Wed, Apr 29, 2009 at 1:50 PM, Jesse Keating <jkeating at redhat.com> wrote:
> On Sun, 2009-04-26 at 12:10 -0400, Matthias Clasen wrote:
>> On Sun, 2009-04-26 at 08:23 -0700, Jesse Keating wrote:
>> > On Sun, 2009-04-26 at 12:13 +0200, Kevin Kofler wrote:
>> > > I must say that, while I do believe users need to have access to the sound
>> > > card controls exposed by ALSA, the precedent this decision sets worries me.
>> > > It should ultimately be the GNOME Desktop Team's decision to decide what
>> > > goes onto the GNOME spin. I definitely wouldn't be happy if FESCo started
>> > > dictating the contents of the KDE spin.
>> >
>> > I don't see it as dictation.  I see it as reaching a compromise.  FESCo
>> > didn't even vote on it until after mclasen agreed to put the mixer
>> > there.
>>
>> Of course, having wwoods threaten to revert the whole VolumeControl
>> feature has nothing to do with it...
>
> Given that wwoods isn't even in FESCo, that'd be a neat trick.  But
> isn't that the very definition of compromise?  Two sides each have
> desires that neither side can agree upon.  Each side makes concessions
> until they meet somewhere in the middle?

OK, so here's the problem with this.  Compromise sometimes works well
when trying to form rough consensus, or for finding the least
disagreeable position for the most people - for politics maybe.
However, it does not work well as a technique for designing products
that people love to use.  When you find the middle ground between
great and awful you don't find good - you find mediocrity and malaise.
 Please read http://headrush.typepad.com/ for more on this topic.

We've been designing compromised products for far too long.  We need
to acknowledge that we can't please everyone all the time and need to
make choices.  We need to resolve to be great - it doesn't just happen
by accident.

Another thing we have here is a classical case of design by committee.
 This "compromise" dictated by FESCO is a design decision.  A decision
that affects the look, feel, behavior, and experience of the desktop.
Firstly, we should be very sceptical that *any* committee can make
great design decisions (see
http://www.codinghorror.com/blog/archives/000321.html).  Secondly,
this particular committee is simply not qualified to make design
decisions for the desktop - you are all well meaning and very good at
your particular field but frankly not the people that should be
designing our experiences.

In this particular case, there is nothing wrong with telling some
people either: a) you may have to wait for the next release in order
to get the optimal experience or b) go to this wiki page and follow
the instructions to make it work on your somewhat atypical
system/setup.  We simply cannot avoid this if we continue to use a
time-based release process.  There will always be cases where we don't
finish everything we'd like to.  The answer cannot be - let's
compromise our vision and create crappy products.  That is, it cannot
if we ever want to recover marketshare and mindshare from other
distributions (not to mention other/better OS's), and make something
kick-ass that we can be proud of.  The only other option is to move to
a feature and/or goal based release process where you release when
things are done.  However, I'm not sure this is the right thing to do
at this time.

I know you work really hard and care a lot about Fedora, and I'm
picking on you a bit because we all respect you, but honestly the
opinion you expressed is part of what is wrong with Fedora today.  We
have a lot of work to do to make Fedora truly great - and one of the
first and most important things is to make sure we have the mindset
that will allow us to achieve greatness.

Thanks,
Jon