Re: Two new projects in the CVS and 5th Toe

[Alexander Larsson <alexl redhat com>]

On Wed, 2003-09-03 at 14:17, Murray Cumming wrote:
> On Tue, 2003-09-02 at 21:21, Ole Laursen wrote:
> > I have two projects for which I would like to continue development
> > within the GNOME CVS:
> > 
> >   Hardware Monitor - http://www.cs.auc.dk/~olau/hardware-monitor/
> > 
> >   A better monitor applet for the panel than one in gnome-applets. It
> >   supports combining a given viewer type with any monitored device,
> >   and supports lm-sensors.
> 
> I haven't tried it, but I wouldn't see the point of using GNOME
> resources for an alternative to part of GNOME unless we were thinking
> about replacing that part of GNOME.
> 
> >   Monster Masher - http://www.cs.auc.dk/~olau/monster-masher/
> > 
> >   A mash'em-up action game for GNOME. There's really not a lot more to
> >   add. Mash monsters and save the world of the gnomes!
> > 
> > Both projects are mature and stable and written in C++ with gtkmm,
> > gconfmm, libglademm etc. I believe both are well-written and quite
> > easy to maintain. Both applications are closely tied to GNOME.
> 
> This is just my own opinion, but I don't think GNOME's resources should
> be used as a sourceforge. There is a random bunch of stuff in GNOME's
> cvs, but I think anything new should be something that we think might be
> part of the GNOME platform or desktop one day. It's hard to think about
> that for a first version of something. 

I totally disagree with all of this. Some minimal resources on
cvs.gnome.org costs us nearly nothing. Most projects like these are
small and don't generate a huge amount of traffic, so the cost to the
gnome project is very low. However, the gain of goodwill from developers
is immense (in fact, its the only thing of lasting value the gnome
project has). 

If someone comes with an app using the gnome platform and wants to have
it in gnome cvs so that other people can help him work on it thats great
for him, and he gets to feel like part of the gnome community. If we
tell him "no, your app doens't look like it has a high enough quality",
or "we already decided to do that in another way, we're not interested
in your work", then we've alienated a developer for no real reason, and
if there is *anything* we need in the gnome project it is new
developers. New developers become used to the gnome code and the gnome
organization, and in the extension some of them become core gnome
hackers, driving the project forward.

If my first app in gnome cvs (dia) wouldn't have been allowed and I
would have gone elsewhere, would I today be part of the gnome community?
What about havocs gless? Maybe it wasn't a great app, but does the cost
of having it in gnome cvs somewhere outweigh the value of what havoc
contribute to Gnome? Can we really afford at any cost to drive away
developers? 

Basically, I think this is a very dangerous route to take, and we should
take great care before changing a policy that has worked well for us.
(We never ever denied someone a cvs module that I know of.)

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 Alexander Larsson                                            Red Hat, Inc 
                   alexl redhat com    alla lysator liu se 
He's a suave guitar-strumming assassin with a winning smile and a way with the 
ladies. She's a cold-hearted cigar-chomping magician's assistant living 
homeless in New York's sewers. They fight crime!