[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

on the role of debian among its derivatives

[ oops, this mail came out longer than I expected, sorry ]

On Fri, Aug 27, 2010 at 07:14:19AM +0200, Christian PERRIER wrote:
> As a side comment: we should maybe some day stop faking ourselves by
> just seeing Ubuntu as "Yet Another Derivative". It is by two or even
> more orders of magnitude the largest Debian derivative.

I personally don't think we are faking ourselves in doing that; in fact
I don't think we're doing that at all.

> It was indeed really interesting to see (at DebConf) the funny way
> that all talks about such topics were mentioning "the Debian
> derivatives" when, indeed, all that they were talking about was
> Ubuntu. As I told to Matt Zimmermann later on in a private discussion,
> this sometimes sounds like ack'ing that Ubuntu is indeed very special
> as a Debian derivative (because of its prevalence) is something we are
> really scared of.

This is a very interesting comment, so let me explain my POV and why, in
the specific example of my DPL talk at DebConf10, I've spoken generally
of "derivatives" rather than of "Ubuntu" specifically.

I'm very well aware of the relevance of Ubuntu as our most popular
derivatives. Interesting discussions during last DPL campaign (which is
still one of the few moment in which we discuss "Debian politics" at
large) suggested that Ubuntu has something like 10 times the users of
"pure" Debian. There are huge potentials for Debian in that: a potential
of users which (indirectly) benefit from our work, a potential of people
power that can contribute back to work originating from us, and also a
potential of new contributors which discover us and our principles via

That is exactly why, gathering input from the Debian community, I've in
the past few months approached Canonical and the Ubuntu community
(e.g. going to their conferences, regularly discussing with Canonical
people, etc.), in an attempt to improve over past grudges and try to
figure out how we can reduce to the very minimum the "viscosity" of
patch flows in both directions.

I consider all this to be evidence of the fact that we are not faking
ourselves "ignoring" Ubuntu, belittling them, or anything like that. We
know they are there, we are glad they bring our work to such a larger
public, and we are actively working with them in the interest of both
distros and Free Software in general.

At the same time, In doing all that we should not consider Ubuntu as a
special case, as that would be a mistake. Ubuntu is currently one of a
kind in term of users, but assuming it will be the case forever is
risky. After all, who among us foresaw something like Ubuntu coming 10
years ago? I surely didn't, but Debian was there. Debian will be there
10 years from now as well, while I have no idea how many others Ubuntu-s
will be there or what would have happened to the "current" Ubuntu by

So the point (or at least my point) is rather understanding what is the
role of Debian in all this, obviously remaining loyal to our principles.
On one hand I think we should keep on playing the "concurrency" game and
preserve/increase our user share, because we think they are better
served by a distro based on our principles than by others. On the other
hand, derivatives are there and Ubuntu has shown that they can be *very*

I wish that all Debian-derivatives out there (we have plenty!) can be as
successful as Ubuntu, because I find that to be in the interest of Free
Software. We can help them doing that by *generalizing* the relationship
we have, or we are building, with Ubuntu. As a tiny teeny example,
that's why we have a "derivatives front desk" rather than an "Ubuntu
front desk". If you look at the discussions there, you'll find two

1) activity comes mainly from Debian and Ubuntu people (no surprise)
2) plenty of other small derivatives contributors showed up, said
   "hello", and are now probably lurking with interest

The benefit of the generalization is that other derivatives can learn
from the Debian-Ubuntu experience (from both positive and negative
aspects) ... and do better!

If we start considering Ubuntu as a special case just because it's the
most successful derivative and, more importantly, if we start
*communicating* and thinking that way, I believe we risk to hinder the
growth of other derivatives and I'm sure I personally don't want that.

Bottom line: learn from the Debian-Ubuntu relationship, and generalize.

> I don't see handling Ubuntu in some special way in DDPO (as proposed
> in this discussion) as such an horrendous thing. Couldn't we simply
> ack the fact that having a bug report rate that is 10 times higher
> than the one in "core" Debian is making it special in some way?

Now, this is tricky as it's where principles clash with reality. The
column in DDPO is surely technically useful, for all of us who would
like to benefit from Ubuntu-originated feedback and work. The fact that
the column exists is just great.

At the same time I don't see much of a benefit in having it enabled by
default. After this discussion, I've checked my DDPO settings to notice
that I've been using from years a DDPO cookie who stored my preferences,
so I doubt I'll ever have to care again bout the default view. This is
probably the case for 99% of regular DDPO users.

So I must admit I find all this technical discussion a bit useless: what
are the benefits of having that column visible *by default*? Just make
sure it is evident that columns are customizable and how preferences can
be saved (cookie/bookmark/whatever). In fact, there's a clear usability
problem in DDPO as the number of columns grows; the best default view is
probably one where the shown columns are kept at the very minimum as
that, in turn, will also invite more eagerly users to explore which
columns they can enable.


Stefano Zacchiroli -o- PhD in Computer Science \ PostDoc @ Univ. Paris 7
zack@{upsilon.cc,pps.jussieu.fr,debian.org} -<>- http://upsilon.cc/zack/
Quando anche i santi ti voltano le spalle, |  .  |. I've fans everywhere
ti resta John Fante -- V. Caposella .......| ..: |.......... -- C. Adams

Attachment: signature.asc
Description: Digital signature

Reply to: