Re: focus! (was Re: Focusing on innovation re: mono, python et al)

[Havoc Pennington <hp redhat com>]

Sankarshan Mukhopadhyay wrote:
> Havoc Pennington wrote:
>
> > My first-order answer is that GNOME thinks of itself as "making a 
> > desktop" - even though the _reality_ is that the larger GNOME 
> > community/ecosystem is doing way more than that, and that the larger 
> > tech industry is doing still more.
>
> Would you consider junking the concept of "GNOME as a desktop" in favor
> of "GNOME as an application development programming context" or would
> think that the slicing should go deeper ? Namely, percolating the idea
> right down to the level where applications are developed around GNOME
> core (assuming the segregation of GNOME core and GNOME extras). From the
> earlier mail that you posted it would appear that you favor a shift of
> GNOME from a software development paradigm to a more personal/social (if
> I may) context. Wherein the *who* assumes greater importance while
> releasing GNOME rather than *what*.
>
> As it stands I don't see an aggressive movement towards (re)doing the
> GNOME messaging - its happening but its taking its time probably because
> of the distribution centric messaging that goes for GNOME. Perhaps the
> time is really there to start talking more about the context in which
> GNOME figures in every day computing rather then the concept where GNOME
> provides applications (cool as they may be) but in no ways do emphasize
> the stuff GNOME is supposed to do.

I tend to think explicit platform-building sucks (vs. accidentally 
making a platform in the course of making something useful).

What I'm advocating is something like this goal set:

GNOME big picture: Bring a 100% open source computing environment to the
                   general public.

Subprojects we have already -

   GNOME Thin Desktop:
       Create a manageable, secure, simple, gratis desktop for computer
       labs worldwide.

   GNOME Technical Desktop:
       Create a fun, hackable, rapidly-changing work environment for
       programmers, administrators, and tech enthusiasts.

   GNOME Server Console:
       Create a command center for enterprise server operating system
       administrators.

Stuff we have already, but not labeled GNOME, much of it never will be, 
but by way of example:

   GNOME "One Laptop Per Child":
       Create a simple social environment for experiential learning.
       (or whatever, would not presume to know how to state this one)

   GNOME Maemo:
       I don't know their "concept" or target audience, but I can
       imagine something like -
       Create a "newspaper replacement" device for coffee shops,
       the kitchen table, riding the train to work.

   GNOME [anything - look at the full breadth of the tech industry!]

In other words, associate the names/brands/teams with 
specific-benefit-to-specific-audience statements.

Many of these things would share code, or even almost all code. That's 
what I'm saying about not splitting by codebase.

There's a natural "platform" which is just "stuff many of these 
subprojects happens to use"

There's are also other "platform" meanings, like "stuff the desktop-ish 
subprojects recommend to desktop-ish ISVs"

Havoc