Re: GNOME Enhancement Procedure

[Havoc Pennington <hp redhat com>]

Dan Winship <danw ximian com> writes:
> On 19 Jun 2001 13:25:49 -0500, Chema Celorio wrote:
> > > > Maciej Stachowiak <mjs noisehavoc org> writes: 
> > > > > First, even though I think we do need some kind of more formal
> > > > > process, this one sounds a bit too bureaucratic for my tastes what
> > > > > with all the rules and regulations.
> > 
> > I second Maciej here, this is not how Free Software works.
> 
> You're right. The way Free Software works is that once a project reaches
> a certain size, you have to have one of:
>     * A benevolent dictator (or dictators) whose decisions everyone
>       accepts
>     * Massive infighting and backstabbing among developers
>     * A fork into two smaller projects
> 

I really think there are free software projects that run fine with a
procedure like the one I've mentioned. Python, Apache, W3C, Debian,
etc. gcc has a steering committee, I'm not sure of the details. Owen
says even Perl has something like this these days. Anyway, basically
there is precedent.

Even for GTK we do a lot more posting things for review and tracking
things in Bugzilla than ends up happening for GNOME as a whole.

So I'd add a fourth bulletpoint, basically. ;-)

For bullet a) we have no reasonable candidate IMO. Everyone is
company-affiliated, even if they weren't, GNOME is far larger and more
complex than the Linux kernel. I don't think one person can get it
right.

b) is just not going well. We simply must have a way to make decisions
and we must have people talking to each other.

c) is suicide.

So, we need d). The question is, what are the details of d), what form
will it take - and there's lots of room to get that right or wrong,
but also to evolve it over time.

> > Things should get discussed and your voice shall be heard by how big
> > your contributions are
> 
> Yuck. "It's better to be prolific than it is to be correct."

Indeed. Conservative is good in many areas.

I don't know - I'd rather have fewer features go in, with some decent
consensus/review/QA on the features that go in, instead of having
twice as many features where half of them turn out to be
unmaintainable, stupid, or whatever.

The problem with GNOME at the moment is NOT lack of features.

Havoc