Re: "Why" Debian Core Consortium ? Why not UserLinux? Why not Debian?
* Ian Murdock <firstname.lastname@example.org> [2005-08-07 17:59]:
> > Do you seriously think that a new organization which hasn't
> > actually talked to Debian at all before being created will help
> > bring some of these closer to Debian proper?
> What does that mean exactly, "to talk to Debian"? The DPL is in
> the loop, plus a dozen or more Debian developers that work for the
> participating organizations. Debian is a group of individuals, so
> is involving some number of those individuals not
> "talking to Debian" in just about the only sense one can do that?
Debian is indeed a group of individuals and they are distributed all
over the world. As such, our official communication infrastructure is
based on the Internet, in particular on mailing lists [*]. This
proposed could be discussed with Debian on the debian-devel lists (for
technical matters), debian-project (for organizational matters) and
debian-private (if it really really cannot be discussed on a public
mailing list). debian-project, which we're using now, is probably the
best venue, but the point is that you should have discussed your plans
before going ahead and announcing stuff to the world. How would you
feel to hear from the press that I started the Progeny Common Core
[*] We're generally not very happy when decisions which effect the
whole project are made in private. Please refer to the "Vancouver"
discussion (starting in March) for a good example.
> I for one would be delighted if this were to become an official
> Debian project, and we are indeed planning to work within the
> existing Debian framework to the largest extent possible. I just
> figured coming in *assuming* any of that as a forgone conclusion
> would be seen as presumptuous.
Obviously, there's no guarantee that your proposal will be accepted
but this mailing list is the appropriate place to discuss it. Maybe
you could explain what this alliance is exactly about, how and which
kind of resources you intend to put forward to it (within the existing
framework of Debian) and why this initiative is a good idea for Debian.
Based on this, we can discuss whether this is something Debian can and
should support and whether it should be an official sub-project.
Anyway, just a few more words to elaborate what my problem with
someone calling themselves a "Debian common core". First of all, let
me emphasize that I'm all for more cooperation. I think there's way
too much reinventing the wheel going on. Just looking at the partners
of this alliance, I see that Progeny has a graphical installer based
on Anaconda, Linspire has their own installer, and Sun Wah wrote a
graphical interface to debian-installer (which they promised to
integrate into d-i; unfortunately this never happened, and in the
meantime, people started working on a graphical d-i front-end which
will probably make Sun Wah's installer obsolete). If you look at my
past record, I've always promoted working together and I haven given
various organizations advice on how they can work with Debian.
So what's the problem? The problem is quite simply that anything
which aims to be a common Debian core claims exclusivity
(intentionally or unintentionally). There really cannot be two common
Debian cores. Debian has always given the same rights to the
community. There isn't one company which has the exclusive right to
sell Debian. Rather, we publish ISOs and everyone who sells those
ISOs can sell "official Debian". There was a new company once which
wanted to be "Debian Press". While it might be nice to have official
books on Debian, we cannot give one company the exclusive rights for
this. Does this mean that no third party can use the "Debian" name in
any way? No, of course not. In the case of CDs, you can call it
"official Debian" if you use our ISOs, or "Debian" if it's basically
Debian with minor modifications. If you'd like to publish an official
book on Debian, we could create a "Debian Press" brand and license it
to various companies - but the rules would have to be the same and
everyone should be allowed to publish under "Debian Press" if they
meet our standards and criteria.
I think that the idea of a Debian core is fundamentally a good one.
However, it would be unfair to give Progeny et al. permission to
create a "Debian common core". After all, why should _you_ be allowed
to create a common core? What if Ubuntu/Canonical or UserLinux (or
whoever) would like to do the same? And no, your alliance being open
to others doesn't solve this problem - what if UserLinux would like to
create a Debian common core based on technology which is slightly
different to yours (or whatever NIH reason people may come up with)?
If there is to be a Debian common core, it can only be defined by
Debian. So if you want to make this a reality, think how you can work
with Debian, what resources you can provide and what's in it for you
(having the common core will make your work with ISVs easier, you will
get customers if you are the driving force behind making this
happen within Debian, etc).