On Tue, Mar 29, 2011 at 10:55:04PM +0800, Paul Wise wrote:
> Cast your mind back 5 years and forward 5 years.
With 5 days of delay, that would have bee perfect!, too bad I added an
off by 1 bug to that count.
> What important opportunities have we missed out on? What important
> things™ have we left by the roadside to our detriment? Which roads
> have lead us astray? Why did we miss those and how can we notice
> similar things in the future?
- We didn't realize early enough the importance of sharing efforts with
derivatives and other distros. Years back (although maybe more than 5
years) we though that we had an inextinguishable source of people
power, which is clearly not the case for any volunteer projects out
there. To fix that there are several ways, one is that of making it
easier to share the work among stakeholders with similar interests.
- We didn't realize early enough the potential of contributions from
non-developers, not only of those with skills that geeks tend to
consider "soft" (e.g. artwork, communication, etc.) but also of those
with technical skills other than packaging (e.g. triaging, testing,
technical doc writing, etc.)
- We didn't realize early enough the importance of being a welcoming
community and the risk that not being one would have driven away
potential contributors to other distros.
- We didn't realize early enough that strict package ownership is a
barrier to contribution. It has some good features, but it needs
monitoring for inefficiencies and easy ways out (e.g. properly done
NMUs).
(Unsurprisingly, some of the above goes in the direction of other topics
I've already mentioned in reply to other campaigning questions ...)
I believe we are starting to realize some of the above, and that we have
also started working to fix them, but lot is still to be done. Hence I
do consider the above part of the challenges we are facing at present.
> What challenges are coming? Are we ready for them? Where should the
> universal OS go tomorrow? What changes could we make now to enable us
> to tackle future challenges?
- The increasing lack of relevance of "local" operating systems, in view
of the increasing outsourcing of computations to "the cloud".
Arguably, this is not a challenge for Debian, but for Free Software in
general. We need to find a role for Debian in the battle for
liberating users from the risks of moving all their computations to
remote machines whose software and data they do not own. One part of
Debian role into that could be to ensure excellent packaging of
federated web apps. Another part is to make it as easy as possible for
"cloud" users to run Debian on the infrastructure of their choice.
- Keep up with the pace of commercial distros. Distros with bodies of
hundreds paid developers can put into use, especially under crunch,
more people/hours than most volunteer projects, including Debian. Will
that mean that Debian will remain behind technical improvements made
by others? I don't know. But a world in which a group of independent
volunteers stays up to par with the biggest companies out there is
something worth fighting for, at least for me.
- Add to our release offering a release which suits the needs of
developers and desktop users who live "on the edge" and need updated
software more often than what the current schedule of Debian stable
releases allow. Debian Testing is one such release---arguably it's the
release who invented the now trendy notion of "rolling releases"---but
it has some shortcomings for the developer and desktop use cases. Some
of the discussions around the "CUT idea" are trying to fix them, but
it's challenging as we need to find good balances in terms of new
developer "duties".
- Become even more dependable:
- internally, to developers, introducing time-based freezes (see
current discussion on -devel)
- externally, to stable users, by introducing distro-wide, long-term
security support for stable (topic on which the security team is
working already)
In general, to face any forthcoming big change, we need to stick more
and more to the well known principles of "rough consensus and working
code", rather than to fruitless discussions and inertia as defaults. We
need more people who dare to propose changes and who are able to show,
with working code, that those changes are viable. Otherwise the whole
(geek) world will evolve around us, leaving Debian behind.
Cheers.
--
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. Capossela .......| ..: |.......... -- C. Adams
Attachment:
signature.asc
Description: Digital signature