On Wed, Aug 08, 2001 at 06:54:51AM -0500, Jared Johnson wrote:
> On 08 Aug 2001 21:08:15 +1000, Craig Sanders wrote:
> > On Wed, Aug 08, 2001 at 04:01:17AM -0500, Jared Johnson wrote:
> > > Doing an NMU without the maintainer's consent and without the
> > > maintainer being MIA is one of the more evil evil evil things a debian
> > > developer can do.
> > why? what's so evil about that? debian isn't a turf-war game of "mine,
> > mine, mine".
> debian isn't a turf-war game specifically because debian developers
> respect each others' imagined jurisdictions.
And because we've got a fundamental goal that's meant to be more
important than all the petty bickering, viz getting out a high quality
free operating system.
NMUs are okay as long as (a) you get it right, and (b) you send a patch
with the changes you made to the BTS. Everything else is just being
extra nice.
> http://www.debian.org/doc/developers-reference/ch-nmu.en.html
> Do I even need to point out that this makes it painfully clear that it
> is off limits to do an NMU without the maintainer of the package giving
> explicit permission or else being MIA for a long time?
The developer's reference isn't policy. ``It contains ... generally
agreed upon best practices.'' Personally, I think it empahsises the
wrong things about NMUs, but whatever.
> > being maintainer of a package doesn't mean that it is your PROPERTY, it
> > means that it is your DUTY.
> And according to the debian constitution debian developers have the
> right to do this duty if they're able. They don't have the right to
> usurp each others' duties without consent or neccesity, even if alot of
> sid users think it's a good idea.
All this talk of rights and duties is really way off track. If you
can see your way to improving Debian, you should just do it. Just try
not to get in the way of other people doing what they can (like, say,
Darren appears to have done to Aaron by telling him not do do NMUs, or,
like, say, telling the maintainer that you're about to NMU one of his
packages because such-n-such a bug needs fixing).
> > IMO, it's sitting on a package and doing nothing with it that is evil.
> Myth isn't sitting on any package, he's developing a package. Not
> uploading a suboptimal package is not evil.
A suboptimal package has already been uploaded. Myth may not be able
or willing to make any improvements on it right now, but Kitame appears
to be able to. As long as Kitame's not going to screw up Myth's future
uploads, why shouldn't Kitame upload? Who exactly loses out?
> FTP masters probably would not so much agree with this; and I'm glad,
> because I also disagree with it. It's basically hijacking the package
> and sticking a different name on it, with the added bonus of creating
> problems. The debian constitution gives debian developers the right to
> make technical and non-technical decisions regarding their work.
mozilla.deb would be Myth's work in this case, and mozilla-kitame
Kitame's. Changes to one wouldn't anything to do with the maintainer
of the other. The reason not to do it is because it detracts from
the user's experience (much like, hypothetically speaking, separating
the packages related to a particular desktop onto a different server
somewhere would), because it bloats the archive (affecting the amount of
money donors have to spend on bandwidth and disk space to host mirrors),
and because, technically, it's completely redundant.
Cheers,
aj
--
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.
``_Any_ increase in interface difficulty, in exchange for a benefit you
do not understand, cannot perceive, or don't care about, is too much.''
-- John S. Novak, III (The Humblest Man on the Net)
Attachment:
pgpQv6mBWukwY.pgp
Description: PGP signature