Re: RC bugs in packages maintained by this list
On Thu, Oct 18, 2001 at 04:26:06PM +0200, Christian Kurz wrote:
> On 18/10/01, Colin Watson wrote:
> > Are you willing to maintain it,
>
> No, since I'm currently happy with the number of packages that I
> maintain. Also I'm tracking the OpenSSH development and so far it worked
> for me, so that I don't have a need to use ssh2.
>
> > or at least make it suitable for release?
>
> Is that the new way of threading people who speak up against a removal
> request for a package?
Show me the threat. Otherwise, please calm down and stop being a twit.
I'm merely trying to avoid the situation where people say "keep it" but
nobody wants to keep it enough to fix its bugs so that we *can* keep it.
Two of the RC bugs in ssh2 have been open for more than a year, causing
it not to have been included in potato, and no-one seems to have cared
enough so far.
A package can be as worthy as it likes, but *somebody* has to want to
maintain the thing. QA maintenance, while certainly better than the
situation before the QA Group existed, is almost by definition not
acceptable for a package in the long term. If it is, then there's no
real reason to have normal maintainers at all.
> Also why did you ignore my hint in which case a removal maybe
> acceptable?
I didn't. I was talking about the case in which the openssh maintainer
reckons that there's a good reason to keep ssh2. In that case it needs a
maintainer.
> > We're only contemplating its removal because it's orphaned and buggy.
>
> Oh, so because it's orphaned we shall remove it? Great, when do you ask
> to remove more packages from this list `grep-available -FMaintainer
> -sPackage packages@qa.debian.org` which contains 225 packages with the
> same reason? And yes, I'm pretty sure that some packages in that list
> have also bugs and can therefor be called buggy, so that you reason for
> removal is also fulfilled by them. So I plan to see more removal request
> from you for packages which match the criteria that you defined above.
> Otherwise this looks for me like someone seeking an very simple and easy
> solution for handling release-critical bugs in packages maintained by
> the members of this list.
I look forward to seeing your proposed modifications to
http://qa.debian.org/documentation/qa.html/ch-rules.html which modify
the text that says that, after one month in which nobody has volunteered
to maintain an orphaned package, "they will be withdrawn from the
unstable distribution". I'm absolutely certain that a package which
hasn't had a maintainer upload since February 2000 and which has RC bugs
over a year old is well outside the limits where that document says
(correctly, as far as I'm concerned) that we're supposed to seriously
discuss removing the thing.
Yes, if I see low-quality packages maintained by QA that nobody wants to
maintain for real, I will (a) fix them if I can, (b) look for a proper
maintainer who can fix them, and (c) if none can be found ask for them
to be removed. (Sometimes the prospect of a package being removed can
prompt somebody to step up to maintain it, so even the last resort (c)
can be useful.) I won't be the first. Do you think that leaving crappy
packages around increases the quality of Debian?
If we don't have the guts to remove occasional packages that are "so
buggy that we refuse to support them" (to quote Policy), then "Quality
Assurance" is not the correct name for this list.
Now, I suppose I've just bitten down hard on your attempt to start a
flamewar. I apologize to the rest of the list. Can we get back to fixing
packages now?
--
Colin Watson [cjwatson@flatline.org.uk]
Reply to: