[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Parted, advice for distributions
From: |
Andrew Clausen |
Subject: |
Parted, advice for distributions |
Date: |
Sun, 27 Jan 2002 07:37:06 +1100 |
User-agent: |
Mutt/1.2.5i |
Hi all,
This email is directed mainly at package maintainers.
First, 1.0 and 1.2 haven't been maintained for some time.
My fault, I guess, for not making this clear. But, from
my point of view, they are unmaintainable. Many of the
bugs are "broken by design". (Most notably: not using
a constraint solver in 1.2 is practically impossible)
In particular, I think Parted should be removed from
Debian-stable, or updated to 1.4.x. Also, Debian
(all distributions) should track releases in the stable
branch. I don't do anything radical in them, and if
you think I am, then complain!
The idea is: bug fixes, improved documentation and
new/updated translations should reach users. This seems to
be entirely compatible with Debian's policies (and if it
isn't, they are broken, IMHO). Am I missing something here?
I have heard many-a-debian-user complain about bugs that
were fixed over a year ago. If you think that being
conservative about updating makes you more stable, you're
kidding yourself.
Second, 1.6.0 is likely to be released fairly soon. If
you are a distribution that doesn't issue updates very often,
it would be prudent to wait a month or two before considering
1.6.0. In the past, serious bugs have been found in
early stable releases. I think this is less likely this
time, because less work has been done on low-level "dangerous"
stuff. OTOH, these are famous last words...
Anyway, this is all largely my fault for not saying this
before. But we're all learning ;)
In future, I'll announce when I intend to stop maintaining
stable releases, and give some warning.
BTW: is there any interest in me resuming maintainence on
1.2.x or 1.0.x? It would be a lot of work... convince me!
Andrew
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Parted, advice for distributions,
Andrew Clausen <=