Re: [gnu.misc.discuss,gnu.emacs.gnus] Free software: Packagers vs Developers
Hi,
>>"Adam" == Adam Di Carlo <adam@onshore.com> writes:
Adam> Regarding Manoj's treatment of the upstream maintainers and practice
Adam> of being a good maintainer, etc, I am quite sure, Manoj, that you have
Adam> done your utmost. One minor point:
Why thank you kindly, sir.
Adam> Manoj Srivastava <srivasta@debian.org> writes:
>> The soon to be released Gnus is a major rewrite, and there are
>> surprises. I used the unstable version to field test teh new version,
>> and it took a couple of releases to work out the new dependencies and
>> conflicts. I still stand by my relese of Pterodactyl Gnus into
>> unstable.
Adam> My issue here is that if the upstream maintainers, licensing
Adam> notwithstanding, have expressed the desire to *not* have a
Adam> certain version of software in wide distribution, then their
Adam> wishes should be respected.
Adam> Now I'm not saying that they did or did not make this request; I'm
Adam> just saying that if they did, IMHO, their request should be respected.
To the best of my knowledge, Lars has said nothing on
this. Some people have objected, true. Unlike those, actuall Debian
users asked for the new version (which has built in MIME
handling). The stable version of Gnus in Slink was released about a
year ago, which is a major change for Gnus.
The standing rule on development version of Gnus is that it is
software under development, and it can make you lose data. (The
README says it is a Beta version). I would consider releasing such a
version into a release as wrong, since the expectation is that things
in stable do not beak. I would not consider releasing every
development version either (for the record, of the 95 releases made
since last august, precisely one has been uploaded into unstable).
The release uploaded was in constant use by me for over a
month, and I was also watching the mailing list to look at the
problem reports. There has been no bug report, or any report of any
problems with it.
Several variables did change a lot, and one needs to look over
the configuration before being able to use the new version
(specifying news/mail sources has changed). This was mentioned in
the Changelog, and the release announcement. If you are not reading
either, and you can't be botherd to lookj at the docs and update your
config, you should not be following unstable.
As I said, I made a judgement call, and I stand by it.
Adam> If all you want to do is get the dependancies right and have a few
Adam> people test it, then you could upload it to experimental or a private
Adam> area.
The package maintainer generally gets to make judgement calls
like this. You are second guessing with less knowledge of context.
manoj
--
A hard-luck actor who appeared in one coloossal disaster after
another finally got a break, a broken leg to be exact. Someone
pointed out that it's the first time the poor fellow's been in the
same cast for more than a week.
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/%7Esrivasta/>
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
Reply to: