Re: Getting to Topaz (Was Re: getting on a longer release cycled)

[Havoc Pennington <hp redhat com>]

Travis Reitter wrote:
> So, adjusting 1. to "Unique, focused, user-centric benefits" (instead of
> the too-vague "concepts"/"major features that sound cool"), how do you
> (and everyone else) think the plan sounds now?
I think the best shot at this would be to gather a small group that 
agrees on some audience they want to try and do stuff for, and just 
start doing it; I'm not sure how the overall GNOME boat can be turned up 
front, it's probably not possible. The small group would have to be 
prepared for potentially large divergence from the existing 
gnome-panel/nautilus/etc. desktop codebase - they would need to be open 
to doing very different things either instead or in addition, if that 
made sense to provide the benefits to the audience.
> > Unfortunately a lot of people have the IMNSHO insane theory that the 
> > above sort of stuff is "too specific" for something "general purpose" 
> > which is sort of like saying a "hammer" is too specific so our product 
> > should be "a tool."
> The "Big Question", then, is whether Gnome could support many
> significantly different user profiles (like all the ones listed above,
> and many others) _really well_, and not just wandering into the
> "mediocre middle". Is it possible?
In my opinion no, the only way to do lots of these different things at 
once is basically to have a set of distinct codebases. The codebases may 
share code, community, and community infrastructure; that sharing is 
what I'd say "GNOME" should be. Right now "GNOME" = 
gnome-panel/nautilus/etc. instead.
Havoc