Re: Modulesets Reorganization

[Shaun McCance <shaunm gnome org>]

On Wed, 2010-06-02 at 13:40 -0400, Jason D. Clinton wrote:
> On Wed, Jun 2, 2010 at 11:00 AM, Shaun McCance <shaunm gnome org>
> wrote:
>         If we're to allow external hosting for "blessed" applications
>         (to whatever extent we're blessing applications), these rules
>         would go a long way towards helping bridge the gap. But the
>         gap will still be there.
>         
>         It's just more stuff to think about for contributors, and for
>         non-git hosting services, it's another VCS to learn. There's
>         a lot of stuff I have to teach new documentation contributors,
>         and adding another VCS to the mix doesn't help.
> 
> 
> I think there's a misunderstanding that applications would be blessed
> at all. The way this change was proposed in the GNOME Boston Summit
> 2009 and the way the announcement reads, applications are merely
> awesome to the point of having a sufficient level of community
> mind-share, like in other free software communities, or they are
> hosted on our infrastructure (which is really easy to have happen now)
> and easier to contribute to because of that. Remember, any reasonably
> GNOME-related is welcome to be hosted on our infrastructure.
> 
> 
> From the Marketing Team's perspective, we will mention and showcase
> apps which have that high level of "community mind-share". That
> basically means everything that's currently in the Desktop suite plus
> a whole lot more.

This scares me a lot, for a number of reasons:

1) All I've seen are kind of vague ideas of "promoting applications
through our communications channels". Our communications channels
aren't that awesome. Our website is disorganized and often out of
date. We don't have an application showcase.

This is all fixable, of course. But before we just throw away our
finely-tuned release sets in the hope for something better, I'd
like to see very concrete and attainable plans for that something.

2) Translators no longer have any idea what's a priority. How do
we determine what makes a language fully supported?

3) Documentation writers no longer have any idea what's a priority.
Sure, many of us will just work on popular applications (or those
we personally like) regardless of inclusion. But it's really nice
to be able to point people to a definitive list of things that we
consider a priority.

4) Inter-application integration just gets way harder. Yelp can
now use nautilus-sendto. GTG has can talk to Hamster. Lots of
people are talking about integrating with Cheese. It is really
useful to be able to look at a list of applications to figure
out ways you can integrate better. Without a list, we have to
just kind of watch the ethereal space of what was hot on our
site for a few days.

5) Without any notion of inclusion in a blessed list, release
schedules diverge, which compounds the above problems.

6) Who decides what gets promoted? It sounds like the marketing
team alone, if they're the ones doing the promotion. That's not
necessarily bad, and I want the marketing team to take a more
active role. But we'll lose the fantastic whole-community input
we get with our module discussions.

--
Shaun