Re: Thoughts about Kde, Gnome, and Staying Neutral (Was gtk-term widget?)

[Bruce Ide <bruce ide echostar com>]

>... 
> The Point (tm):
> The perfect example of my point is a project near and dear to my heart: The
> Gnome-db project.  I'm a database developer, and the Gnome-db project is all
> about database access.  They had the opportunity to make the world a better
> place.  They didn't.  They're making it worse.  They are GNOME-db, which means
> that the Kde people will not embrace them.  Quite the contrary, Kde will now
> have to unleash kde-db, a similar yet incompatible API.  Gnome-db would have
> been much better off being Gtk-db, or even X-db and leaving Gnome to provide the
> environment.  There is absolutely no reason why their custom widgets
> and APIs have to tie into Gnome.
> 

So write your own, if it bugs you so much. Even at the widget set level
you're going to have in-fighting and in a perfect world, program designs
are set up so that the portions of the work that don't need a GUI are
separate from the portions that do, anyway.

The beauty of this entire development model is if you think something
sucks, you can fork. If your code is good, it'll get used or merged, if
it's not, it won't. And don't take it personally when people elect not
to use your code. People will tend to choose the best tool for their
particular job. The "My way is the right way" attitude works for
Microsoft, but it has no business in this community.

> Oh Yeah, Well Here's Another Point:
> Writing a spec, or an app, or an API because the current one sucks is
> honorable.  Merely duplicating efforts for political reasons is a monsterous
> waste of time.  I do realize that choice and competition make the apps better.
> Absolutely.  Examples:  Gnome/Kde, OSS/Alsa, Redhat/Suse, etc.  Making a carbon
> copy of a program and exchanging the "k" for a "g" in the name is not
> competition, folks.  Case in point Evogellan (Evolution and Magellan).  Why not
> just name them Goutlook and Koutlook?  These guys could kick some serious
> proprietary ass if they would cooperate on "LinOutlook".

As I said, that's the whole point of the community. And NIH syndrome and
programmer ego clashes are no stranger to any development team. However
the right to fork if you don't like the way a project is going is what
makes this community strong. Usually changes and features get merged or
you end up with two radically different products serving different needs
for different people. This process may take a while or it may happen
overnight, but it DOES happen. At least, as long as everyone plays nice
and sticks to a truly open license.

And yes, I'm doing corporate development too, and every so often I do
gnash my teeth at something or other. But I respect how this community
works. Hell, I actually really dig it. I much prefer to work this way.

--
Bruce Ide                                           
bruce.ide@echostar.com
"C has all the power of assembly language combined with all the ease of
programming of assembly language."