[vlc-devel] Releasing too slow

[Rémi Denis-Courmont]

On Mon, 6 Jul 2009 00:38:01 +0200, Rafaël Carré <rafael.carre at gmail.com>
wrote:
> On Sun, 5 Jul 2009 21:41:49 +0300
> "Rémi Denis-Courmont" <remi at remlab.net> wrote:
> 
>> First, there has been way too little informations on what was going
>> on. Basically, I have almost never known what was blocking the
>> release at a given point in time.
> 
> j-b has sent lists of blocking items from time to time, what kind of
> information would you have needed?

J-B did. Way too late and after I had complained already twice. Not that I
would blame him; he was not supposed to be the 1.0 release manager to start
with.

>> Second, it has been way too slow. Most recently, there have been so
>> few and minor bugfixes, that I really wonder why the release was not
>> already done. It makes me think we are waiting for things that nobody
>> wants to fix, which is utterly pointless.
> 
> Right
> 
>> IMHO, the proper time for a freeze is a month or so after forking.
> 
> Makes sense.
> 
> I suggest scheduling regular IRC meetings during forking to sum up the
> current status, blocking items, and who's working on what.
> 
>> Frankly, I am getting quite demotivated. The lack of release
>> predictability is demotivating, and so are the lack of visibility and
>> timeliness. And it appears to hurt our credibility in the open-source
>> community in general. Either we get someone to do source code release
>> management seriously, or we might as well give up.
> 
> j-b is making the release now, like promised in his mail from Friday,
> so I'm not sure if you had missed the mail, or thought that he would do
> it earlier in the day.

I know. But it took us over three months - a record - even though 1.0 had
far less problems than 0.9. That does not make sense to me. I claim we
could have released more than a month earlier with the same resources. We
have lost a lot of time in the early time of the freeze because nobody knew
what was going on as there was none in charge. Then we lost another awful
lot of time just to stabilize a few platform-specific contrib issues.

Lets say that 1.0 was "special" because of the number... But I think we
really need to revamp the release process next time. By that, I mean we
should have a release manager from early on (even before the freeze), and
(s)he should not be afraid of taking decisions. Decisions *might* be bad,
whereas lack of decisions *will* be bad - so I'd rather risk the first
option. And no, the release manager is not the one who builds the binaries
- (s)he is the one who makes informed decisions about the release and
communicates. (I have to agree with Laurent that we have really lacked any
sense of direction lately)

-- 
Rémi Denis-Courmont