Fedora.NEXT Products and the fate of Spins

[Adam Williamson]

On Thu, 2014-01-30 at 14:53 -0500, Bill Nottingham wrote:
> Jiri Eischmann (eischmann at redhat.com) said: 
> > >         That being said, as we go forward as Fedora.NEXT, we start to
> > >         see more
> > >         clearly defined divisions between Products, Spins and Remixes.
> > >         Since
> > >         these discussions needed to happen, we (FESCo) felt it was
> > >         best to try
> > >         to move the conversation public.
> > > 
> > > 
> > > Part of the problem I have with this discussion other than the
> > > alarmist subject is that we are discussing the "fate of spins" before
> > > even coming out with concrete products out of the fedora.next
> > > proposals.  That seems premature. 
> > 
> > +1
> > I cannot agree more. We still don't have (at least I don't) have a clear
> > idea how basics of Fedora.NEXT will look. So this discussion is really
> > kinda premature. We should clearly define the products first, then we
> > can discuss spins and the border between them and remixes. From bottom
> > up please :)
> 
> Blame me, I filed the original ticket. My concern was that we're (obviously)
> doing work in the Fedora.next space around:
> 
> 1) the three products (workstation, server, cloud)
> 2) the base, which lives under them in some manner
> 3) environments & stacks, which can live on top of them
> 
> First, I'm making the assumption that those three products are not the
> end-all/be-all of what Fedora can ever be. So we should have a defined
> process for how new products can be formed. And that can tie into some
> things that are now spins, if they want to go that route.
> 
> Second, as we design how we build and ship all of those things above, it
> stands to reason that implementation choices made there have the possibility
> of breaking spins as they exist now.  That shouldn't happen in a complete
> vacuum without warning, so it's worth discussing where things that are spins
> now fit into that process.

Looked at that way it's obviously an important question to answer
indeed.

I've written two posts on this topic and ripped them up for being
excessively negative, but let me try one more time:

I think part of the worry is the tension between having a product which
is clearly focused around the GNOME desktop (please, can we just agree
on this? I'm not saying it's a bad thing, but I don't think the 'but we
haven't picked a desktop!' hedge is contributing to the clarity of
discussion: it is patently obvious that Workstation is going to be
GNOME), and what that means for all the other desktop spins.

If we decide the alternative desktops are a valuable part of Fedora -
which seems to be a popular opinion - how do we fit them into a
Product-based conception of Fedora?

We can have a KDE Product, and an Xfce Product, and an LXDE Product,
but...at that point haven't we just re-invented spins? It doesn't seem
to quite work with the Product conception.

I'm sure there is a solution for this, and I don't want it to come off
as stop energy. I just wanted to try and have a stab at identifying the
source of concern.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net
http://www.happyassassin.net