Target market?

[Ralf Corsepius]

On Wed, 2007-07-25 at 02:02 -0600, Stephen John Smoogen wrote:
> On 7/25/07, Ralf Corsepius <rc040203 at freenet.de> wrote:
> > On Wed, 2007-07-25 at 03:06 -0400, seth vidal wrote:
> > > On Tue, 2007-07-24 at 12:06 -0400, Jonathan Blandford wrote:
> > > > On Mon, 2007-07-23 at 23:38 -0400, Jesse Keating wrote:
> > > > > On Mon, 23 Jul 2007 23:34:47 -0400
> > > > > "Luis Villa" <luis at tieguy.org> wrote:
> > > > >
> > > > > > (Of course, another option is that my desire for a polished desktop
> > > > > > experience may be best met by someone doing a polished desktop spin of
> > > > > > Fedora rather than by having Fedora work on desktop polish at all, and
> > > > > > that Fedora should merely enable that and work to get fixes/polish
> > > > > > upstream where possible.)
> > > > >
> > > > > There really isn't a reason why enterprising folks couldn't make a High
> > > > > Polish Desktop spin of Fedora, or help rel-eng to make it.  It just
> > > > > needs somebody to drive it.  I couldn't successfully drive a Desktop
> > > > > spin for F7, maybe somebody who understands the target better can.
> > > >
> > > > That's what we try to do right now.  I am pretty much coming to the
> > > > conclusion that "High Polish" can't really be done as an add-on.  You
> > > > need the ability to make the whole distro change (eg, what ubuntu does
> > > > to debian).  If we want Fedora to be a competitive desktop, we need to
> > > > make Fedora a desktop.
> > > >
> > >
> > > So if we take this out to it's next obvious conclusion: where does that
> > > leave the upstream consumer of Fedora: RHEL? Or any server-oriented
> > > initiative based around fedora, for that matter?
> > I am quite irritated whenever reading about "servers" vs. "desktop"
> > vs. ...
> >
> > IMO, servers, desktops etc. are "just setups" of one an the same modular
> > basis. They don't necessarily collide.
> >
> > If they do, to me this means deficiencies of the setup tools and/or
> > packaging.
> >
> 
> THe collision is about resources of who si going to focus on what.
IMO, the "who" and the tools they are using are the key

>  The
> terms are used because for the most part, people rolling out large
> number of servers use different timeframes for technology renewal than
> people rolling out desktops. I have a good number of servers that are
> still running RHEL-3 and will probably be running them til 2009 or so.
> My clients want desktops with newer stuff and so RHEL-5 is already
> cramped for rollout. The conflict is that server software wants a lot
> of long term support, and desktop support doesnt.

FWIW: My primary focus/usage is "application development", so neither
"bleeding edge, eye-candy-ladden desktops" nor "stone-age technology
based servers" (hyperboles intented) are much of interest. I need a
compromise between "stable desktop" and "a new, but not bleeding edge
devel-infrastructure" (exactly what Fedora had provided so far).

Both are reasons, why most Fedora competitors are widely non-interesting
to me.

>  WIth a limited
> number of engineering resources.. you have a conflict.
Well, IMO wrt. FE, a lot of these conflicts (and the work-load you
complained about on a parallel thread) are home-made. 

It's lack of efficiency being caused by lack of simplicity of the
infrastructure being used (bodhi, koji, cvs, acls, distribution
packaging (CD/DVDs) ).

As I see it, RH should focus their forces on improving this
infrastructure to give interested contributors room to step in.

Ralf