Calling off Beta1

[Jonathan Thomas]

Hello all,

On Tue, May 29, 2012 at 1:35 PM, Scott Kitterman <kde at kitterman.com> wrote:
>
> Why cancel instead of just telling people the need to update to soprano from git?
>
> Scott K
>

I think that most of the discussion here is based off of a
miscommunication caused by the title of this thread. The original post
by Albert does state that the current *tarballs* for beta1 have been
cancelled, while the title suggests that the whole of beta1 has been
called off. This uncertainty, I think, has lead to confusion as to
whether any tarballs at all will be made for beta1, and I think most
of the posts (as well as the resulting tensions) have been based off
the premise that beta1 isn't happening at all.

If I am understanding the original post correctly, the release is
still planned. (But perhaps delayed?) On the other hand, I could be
totally wrong, and the cancelled tarballs could mean a cancelled beta1
release entirely.Some things could have perhaps been communicated
better, and things were definitely exacerbated by Trueg's MIA-ness
over the weekend. I think that there are still some pieces of
information that still need to be known before the situation can be
considered "cleared up":

- Given that only the tarballs for beta1 have been cancelled, will the
release proceed with new tarballs now that we have a soprano release,
or will beta1 not have a release in tarball form?
- If so, when is the new release timeframe for beta1? (I'm assuming
that if new tarballs are spun, the timeframe has changed, since the
tarballs are not currently available via ftpmaster)

I also have a few remarks as to how the situation with soprano was
handled in general. There were bound to be bumps along the way with
this transition, so please don't take these as accusations of blame
against any person or group of people, but more as things to take in
to consideration for the future. (Remember, hindsight vision is always
20/20, etc, and there were external events complicating things...)

Given the nature of the reason for cancelling the tarballs, I don't
think a complete removal from ftpmaster/respin is quite the proper
response for a missing dependency. I can definitely see the rationale
for a delay of the release, but once the issue is resolved, the KDE
tarballs will only be respun from the same tag and will be effectively
identical. Technically, the tarballs are embargoed until release
already, or at least embargoed until a day or so before release to
give distros a chance to have binaries more or less built at the
official release time. I seem to remember Kubuntu catching some flak
for this early on in the KDE4 days for publishing tarballs before they
were ready, so I have a bit of personal experience to this bit. ;-)

The tarballs are still useful to packagers in a practical sense even
in the case of a release delay, especially since any respins will be
based off the same tag. Packaging work is cumulative, so work could
still be done with the old tarballs. I think that in the future
instead of the release team removing the tarballs from ftpmaster, a
note send out to the various m-l's saying "the release has been
delayed, so please remember to respect the existing embargo on the
tarballs" would be a bit more appropriate than the outright removal of
the tarballs from ftpmaster.

Regards,
Jonathan