[sylpheed:37263] Re: Debian 12 released with two RC bugs in Sylpheed

Ricardo Mones <[email protected]> Fri, 12 Apr 2024 20:09:01 +0200
Newsgroups	gmane.mail.sylpheed.general
Organization	Debian
Message-ID	<[email protected]>
Hi José,

On Sun, 7 Apr 2024 13:26:49 +0200
José Luis González <[email protected]> wrote:

> Hi,
> 
> Debian 12 was released with two Release Critical bugs I filed on May
> 20th 2023 (#1036424 and #1036388) on Sylpheed about issues that I
> found on stable, and remain, with Debian 12 released later on June 10th
> 2023.

These were not RC, and they were not even closed by me. But I guess you need
some basis to build your case upon ;-)

> The maintainer accumulates a lot of bugs for the package, doesn't take
> care about almost all, 

Yes, package has bugs, upstream doesn't provide a fix, even ignores patches
sent for git: https://www.sraoss.jp/pipermail/sylpheed/2023-March/007115.html

So, do you expect Debian developers to fix them all?

> and when I filed a RC bug because the package became unusable to me he
> downgraded severity to important claiming it was just a Gmail issue, when
> it didn't seem so, even if it was just happening with Gmail.

Seems you talk now about https://bugs.debian.org/1036799

As explained that the package is unusable with your mail provider doesn't
make it unusable for everybody. Believe or not there's other providers where
it works fine. Please, read how Debian severity works:

https://www.debian.org/Bugs/Developer#severities

> I wanted to point you to this bug number to
> provide records, but couldn't find it neither opened nor archived. The
> supposed solution at the time for it was to upload 3.7.0beta1, when the
> existing version was 3.6.0, and the issue magically disappeared without
> explanation from him. I discovered he uploaded later another beta
> (3.8.0beta1), which was included in Debian 12. As far as I recall,
> 3.7.0beta1 got into Debian

Yes, that was unfortunate. As never had had a problem with betas I happily
uploaded it and, against all bets (I was expecting more betas a rc and a
final release), it ended in stable. 

So I guess you can burn me in your pyre for trying Debian users to have the
latest version of Sylpheed, I'm guilty.

> 11. He even claimed at the time that Sylpheed was too old and so
> troublesome and useless and was considering removing it from Debian
> just because of that.


Too old? Never said that. I maintain packages way older than Sylpheed :)
Let me quote it for you from the very bug 1036799 message 28:

| Yeah, I understand the situation is not funny for you. Fact is upstream
| has still not dealt with this bug despite the time passed (months), so
| there's no patch I can apply to current version. 
| 
| This fact joined with the increased slowness of upstream development
| during the past years makes me wonder if it's still worth to maintain
| Sylpheed within Debian.

And clock's still ticking and no news from Hiroyuki since january 2023.

Given the bus factor of this project, I'm starting to fear the worst :((

> I want to know why Debian 12 was released with those two Sylpheed RC
> bags, report the incident to you all, know what to do with the
> maintainer and kindly request that someone better at the job takes over
> Sylpheed maintainance, or otherwise I will become a Debian developer
> and package it myself.

Explained in previous paragraphs. Anyway, pretty please, you're welcome to do
so. The moment you become a DD you can take over maintenance of the sylpheed
package and remove my address. I'm just assuming co-maintenance is not an
option with such an awful maintainer like me, but I'm also open to that.

Believe me, I do joke a lot but this is completely serious, no problem on my
side, really. You can start at https://nm.debian.org/

There's other two people in the Sylpheed team BTW, but the more the merrier:
https://salsa.debian.org/groups/sylpheed-team/-/group_members

Anyway, in the meantime and since you seem to expect DDs to fix upstream
bugs, you can get some practice and send me patches fixing them ;-)

I'd be more than happy to upload a new version with your fixes, really.

> There are earlier precedents of me filing a RC bug on Sylpheed, with
> the bug getting unattended, he raising a bad excuse that it was
> inexistant, and the package caming up later with a newer version with
> the issue solved and me making the mistake of thinking I was wrong
> about the bug existing and needed to be filed, and (me) closed the bug,
> most likely when it still remained in stable (this I don't remember
> perfectly at this time).

Oh, the precedents, here they are :) I'm sure you'll be a great maintainer,
better than me for sure, and with good excuses, no doubt. Eager to see it.

BTW, for the sake of completeness and keep you fully informed, if you feel
the need, in Debian you can invoke the TC to override any decision from a
developer: https://www.debian.org/devel/tech-ctte.en.html

> I even have no doubt that what he packaged to stable (bookworm)
> currently has at least one back door that is not credible at all is in
> upstream, showing up with the spell checker marking some words in this
> email as wrong after initially turning up as correctly spelt, namedly
> "caming" and "mistakingly".

I wonder what are you talking about here. A backdoor in sylpheed package?
Where is it?

concerned,
-- 
 Ricardo Mones
 http://people.debian.org/~mones
 «Don't take life seriously, you'll never get out alive.»
signature.asc (application/pgp-signature, 833 B)