Re: libotr 4.0.0 hasn't entered testing after 55 days

On Sat, Jul 6, 2013 at 5:39 PM, Adam D. Barratt <adam@adam-barratt.org.uk> wrote:

On Sat, 2013-07-06 at 17:14 +0200, Thibaut VARENE wrote:
> On Sat, Jul 6, 2013 at 4:39 PM, Paul Wise <pabs@debian.org> wrote:
> On Sat, Jul 6, 2013 at 10:31 PM, Thibaut VARENE wrote:
>
> > libotr 4.0.0 hasn't entered testing yet and I don't
> understand why. The
> > status page reports that it hasn't been built on any arch
> ("missing 3
> > binaries"), but those binaries are no longer built by
> libotr.

They'll stay in unstable until ftp-master remove them. And that
generally won't happen whilst they still have reverse-dependencies.

> The transition looks like it is blocked by some RC bugs:
>
> http://release.debian.org/transitions/html/libotr5.html
> http://bugs.debian.org/cgi-bin/pkgreport.cgi?sev-inc=serious;sev-inc=grave;sev-inc=critical;src="">
>
>
>
> Still don't get it: why are bugs in packages that depend on libotr
> blocking libotr?

See above. They're keeping libotr2 in the archive, and britney won't
consider libotr for migration while that's the case.

> libotr is perfectly fine and I don't see why it should be blocked from
> moving to testing because some packages that depend on it aren't
> properly maintained? Wnat if these packages are never updated to
> libotr5?

Then they should be removed from testing, and possibly from the archive
altogether.

One of the bugs had an update a week ago, so it seems a little unfair to
claim it's "not properly maintained".

Possibly. Except the new libotr was uploaded to experimental and released almost a year ago. So I'm afraid I'm gonna have to maintain my statement :)

> Are we going to punish those who have updated? This is for instance
> preventing the migration of pidgin-otr and irssi-plugin-otr, and I'm
> receiving (angry) mail about it (which is the reason why I looked into
> this in the first place) ;-P

Then you should tell them to help fix the RC bugs instead. :P

I tend to frown upon the undergoing idea in open source that pushdown is okay and that it's always someone's else job to fix things. It seems especially wrong when said someone else is an end user.

You maintain a library with several reverse dependencies, and uploaded a
package to Debian which changed its SONAME. Part of the implied
responsibility of doing that is to help make sure that the
reverse-dependencies are able to cope with the changes.

I've pinged the bugs against the other packages to query progress.

Thank you.

Regards,

Adam

--
Thibaut VARENE
http://hacks.slashdirt.org/