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

Re: Some proposals



FWIW, I like all the ideas of "freshening" releases and a central source
repository, but wanted to comment on the proposal to stop releasing
altogether.  There are a fair number of us out here who are sysadmins by
day and count on Debian "stable" to keep our wits (and to look good in
comparison to our non-Debian peers).  For example, I manage change in
production environment by using stable + security + a very limited set of
backports (only when they are absolutely required, and then every box gets
them).

As an aside, when it does come time to upgrade, that's when Debian really
shows its one of its strengths, actually being upgradable, unlike most
operating systems, where it ends up being less work to scratch the box and
reload.

So I see those "service packs" as great tool for keeping desktop users
more current, but still see the need for formal releases.  Without them,
the archive is always in flux and it's impossible to get a snapshot where
everything works, and (at least for the most part), works together.  This
is important for production shops.  Without it, you're taking away the
advantage of a widespread testing, essentially shifting the burden of
integration testing onto each site.

Put another way, in a production shop, I need to be able to build multiple
boxes that are precisely identical, perhaps several months apart.  If
there is no Debian stable, then I end up "freezing" my local mirror at
some point in time and performing that testing I mention above in-house.
The lack of a formal release also makes it difficult to talk about
supporting Debian with other admins at other sites.

All that being said, perhaps "server Debian" should become a flavor...

Cheers,
tony

  tony@mancill.com     |  A word to the wise is...
http://www.debian.org  |  often enough to start an argument.  (fortune)


On Wed, 26 Feb 2003, Jaldhar H. Vyas wrote:

> Seeing as everyone else is throwing in their two currency units, here
> are some ideas i've been kicking around.
>
> Proposal:  woody service pack 1
>
> Get Adrian Bunks woody backports, the gnome 2 backport , kde 3.1 and any
> other widely used updated software, do a little testing to make sure all
> the dependencies are in harmony and stick it on a PGI enabled CD.  It
> should be enough to mollify 95% of the critics.  It would be nice if it
> could be done under official Debian auspices but there is no reason some
> enterprising user couldn't do this themselves.
>
> Proposal: put all Debian code in a repository and make all maintainers
> responsible for all packages.
>
> No more waiting for absentee maintainers to fix trivial bugs in their
> packages.  The BSDs do it and manage to put out regular releases.  There
> may need to be levels of privilege i.e a core team but I would argue there
> is a de facto core Debian team right now anyway.  As long as there is a
> good communication and a clear audit trail for when thngs go wrong I think
> it could work.
>
> Proposal:  Stop releasing altogether
>
> sid would be the only Debian distribution.  Other companies/interested
> groups of people would be responsible for putting out finished products on
> whatever schedule suited them.  The debian projects role would be to
> produce the packages, and provide infrastructure (BTS, policy, installer.)
> Does this this sound unworkable?  Look how many specialized Knoppix
> knockoffs are beginning to bloom.  Look how many people try Linux From
> Scratch.  And most longtime Debian users upgrade continuously without any
> thought to releases anyway.




Reply to: