On Sun, Feb 04, 2001 at 01:18:31AM -0800, Chris Waters wrote:
> On Sun, Feb 04, 2001 at 02:28:09AM -0600, Sam TH wrote:
>
> > Count of AbiWord bugs in Debian BTS: 59
> > Number forwarded to AbiWord developers by maintainer: 1
>
> Hey, at least the Debian BTS is public. I have my own private list of
> bugs for Abiword, and it's none of your business what's on it! :-)
>
> Seriously, you should complain to the maintainer. But the fact
> remains that we need to have the bugs in our BTS for our own purposes.
Well, I wholly support having bugs in the Debian BTS. It's just that
the odds of people submitting them to *both* systems is <0. And given
the choice, I'd rather have them in the abiword bugzilla.
>
> Anyway, I only see twenty bugs against abiword, *two* of which have
> been forwarded, at least three have to do with the Debian packaging,
> and four are wishlist. I also see five closed bugs. I see two bugs
> against abiword-common, both Debian-specific, two against
> abiword-expat, one of which is Debian-specific, and one against
> abiword-xml. That doesn't even come close to adding up to 59, even if
> I grant all the Debian-specific ones *and* the closed ones. Where are
> you looking?
>
Well, I counted the archived bugs too, since none of them were
forwarded. And I didn't count any of the abiword-* bugs. And
although 2 of them claim to have been forwarded, I don't think one of
them ever actually got to us.
> Also, at a rough estimate, at least five of the real bugs against
> abiword in our BTS are the same issue: bad non-iso8859-1 handling.
>
Which is also fixed upstream, in a release that's been out since
Christmas. Maybe the rest of you do things differently, but all we've
ever gotten from gecko is that one bug, although there are patches in
the Debian source, which I have incorporated after finding them
there. So you see why I don't have much faith in the system.
Granted, this is better than some. Suse ships a significantly patched
AbiWord, and I don't even know who the maintainer is.
> > So, I wouldn't be advocating change if I thought the current system
> > had a chance in hell of working.
>
> Have you tried? Have you asked? Have you yelled and screamed?
>
Well, I try to avoid yelling and screaming whenever possible. It
rarely improves things.
> We got flamed left, right, blued and tattooed not too long ago by the
> xemacs folks for exactly the *opposite* problem not too long ago.
> Apparently we can't win no matter what we do. Well, I'm sorry, but we
> have a QA process too, y'know, and it's subverted by not getting bug
> reports just as badly as yours is.
>
Well, I want us *all* to get bug reports. And when I get out the NM
queue, I'll want to get the bug reports for my packages. But right
now, a significant percentage of AbiWord bug reports are never seen by
the developers.
Insert plea for integrated, distributed, all-powerful BTS here.
> Bottom line is that Debian package maintainers are *supposed* to
> forward bug reports. I know that I do. If someone isn't then he
> isn't doing his job, and he should be chastised, and if he persists,
> then maybe he shouldn't be maintaining so many packages (or
> something). But nothing's perfect -- heck, some days, I discover bugs
> in random programs, and then forget to report them to *anyone*.
> Claiming that our whole system is broken, just because it's not
> perfect is unreasonable, even if you have been suffering more than
> your share of our imperfections recently.
Well, I'm glad to hear that it works better for other people.
>
> Let's try to work this out like grownups, ok?
Sounds like a plan.
sam th
sam@uchicago.edu
http://www.abisource.com/~sam/
GnuPG Key:
http://www.abisource.com/~sam/key
Attachment:
pgppnaSHDRsJb.pgp
Description: PGP signature