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

Re: request for Technical Committee ruling on Bug #109436



On Fri, Aug 24, 2001 at 04:30:18PM -0500, Branden Robinson wrote:
> I'd like the Technical Committee to:
> 1) Tell me exactly how how rename and/or reversion my package;
> [...] Furthermore, if the Committee agrees with
> Jason's mail "My thought", I need an answer from the archive maintainers
> on 1). [...]

I'd recommend changing the version number to

	xfree86 (4.1.0+1-3)

with

	xfree86_4.1.0+1.orig.tar.gz
	xfree86_4.1.0+1-3.diff.gz
	xfree86_4.1.0+1-3.dsc
	xlibs_4.1.0+1-3_i386.deb

etc. It's the least obnoxiously different name I've seen (as compared
to suffixing .dfsg-free, eg), it's the least effort on anyone's part
(compared to changing the source package name which'd require other
source changes and manual processing on the part of ftpmaster when it
got uploaded), and it can disappear entirely when there's an actual new
upstream tarball (as opposed to introducing an epoch), and it doesn't
really seem likely to confuse anyone ("oh, that's xfree86 4.1.0 with
some minor difference"). There also seems to be some precedence for
that sort of versioning, see for example alsaplayer 0.99.36+1-1 or
mozilla 2:0.9.3+0-1.

All you need to do is change the name of the .orig.tar.gz
so it's something different. If you want to make it
"xfree86_4.1.0.ftpmaster.sux.orig.tar.gz", well, I'm sure we can live
with that.


Personally, I'm still not seeing the objection to changing the version
number, at all. The technical reasons why we don't want to overwrite
files have been listed numerous times, and, as far as I've seen there's
only be aesthetic ("but the upstream version number *isn't* 4.1.0+1, it's
4.1.0!") and essentially bureaucratic ("there ought to be a procedure
for this, even if there's no need for it"), neither of which seem like
they should override even the limited technical objections that've been
discussed. More worrying are the remarks that make this issue seem overly
political and personal ("The archive maintainers and I are deadlocked.
I cannot think of a solution that involves neither them yielding nor
me." and "Sounds like that makes it unanimous against me."). I don't
think that's any basis on which to discuss this issue.

For reference, the .orig.tar.gz overwriting thing has come up on -devel in
the past, see:
    http://lists.debian.org/debian-devel/2001/debian-devel-200101/msg01522.html
for example. There's nothing personal in either the initial rejection,
nor that we're not making any special exceptions.

Also for reference, the current ftpmaster team is:
     * FTP Archives -- <ftpmaster@debian.org>
            member James Troup
            member Michael Beattie
            member Anthony Towns
            member Ryan Murray

Personally, I'm ambivalent about the tech ctte's authority here. OTOH,
since this does appear to have gotten bizarrely political and personal,
getting the best technical solution (read: a new X uploaded with a fixed
version number ;) looks like being best served by having the tech ctte
adjudicate on it, rather than having more rounds of bug 109436 being
closed and reopened, it might be better if ftpmaster just delegates their
authority to the tech ctte on this. In any event, James Troup speaks for
all of ftpmaster in that sense. Getting caught up in the constitutional
aspects of the issue rather than the technical ones is just stupid, IMO.

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)




Reply to: