[Pkg-postgresql-public] postgresql 8.2 packaging

[Markus Wanner]

Hi,

Gerfried Fuchs wrote:
>  I thought that way - but there were points raised that weren't that
> clear to me so far and I guess neither to you: postgres propably won't
> maintain 8.3 for as long as we would have to support squeeze. We are
> currently stumbling into exactly this problem with postgres 7.4:
> Upstream released its last update to it - we will have to keep
> maintaining it for another year.

Okay, I'll try to help with that.

> This means upstream work within Debian
> - and this is nothing we are able to do.

Understood. :-(

>  Exactly. It most propably though is rare that it won't work, but it's
> nothing that one really should depend on, it's nothing guaranteed.

Being able to upgrade is nothing you should really depend on? That has
been one of the major features of Debian for me.

>  No, you are actively requesting a solution where you aren't in the
> position to do so and repeatedly deny the reasons outlined to you.

Again, I'm just expressing my opinion and I think I'm free to do that. I
tried hard to express them in a friendly and respectful manner. If you
still feel pushed into any direction, I'm sorry.

>  You are claiming over and over again what Debian should do and Debian
> shouldn't do, and that Debian "prevents" users.

It's been my experience with productively using Postgres on Debian. And
it is my proposal for an improvement of Debian, yes. If you or Debian
don't want to follow that proposal, fine, that's your good right.

> It would boil down to
> exactly that in the end, like it does at the moment for pg 7.4 in etch.

Sure, I just happen to still like that option better, because I'm
thinking it serves the users of Debian and Postgres better (even if
nobody from the Debian community ever provides bug or security fixes;
Postgres does that for long enough, IMO).

>  ... in a way that just sounds like Debian has to do this and Debian has
> to do that, and Debian "prevents" users from doing things.

Well, apparently neither you nor Debian have so far helped users with
that problem. I did.

I've pointed to a problem, offered my help in fixing it and did
everything I could do to help fix it. So I did myself what I
"requested", anything wrong with that?

And yes, not accepting that help is fine, nobody (from Debian) has asked
for my help.

>  I never denied you to maintain your repository, where did you get that
> impression?

Because you seem to deny that there even is a problem with the current
approach, which implies that my work is useless, IMO.

> And if you wouldn't have started with "Debian should do
> this and that" and "I do not understand why Debian ignores" there
> wouldn't had been any problem.

Oh yes, the problem would have been there, even if I didn't point you to
it. Don't shoot the messenger.

> See my first paragraph in this mail,
> maybe this helps you finally to understand.

I understand that Debian doesn't have the resources to maintain
obsoleted major versions of Postgres.

>  Not anymore because I do not want to have to do upstream development
> stuff for it, which it would mean at the end of the Debian support
> cycle. It's even worse for ubuntu's long time support releases.

I realize that and I accept this position. I'm sure you are aware of the
pros and cons of this decision. Nevertheless, I'm in respectful
disagreement, obviously weighting the pros and cons differently, as
pointed out above.

Regards

Markus Wanner