Re: The nature of the Bonobo freeze.

[Mathieu Lacage <mathieu eazel com>]

Miguel de Icaza <miguel helixcode com> writes:

> > I admit to some trepidation, since it's a longer set of changes than
> > the last one that was promised (only the studlyCaps change) but if you
> > promise publicly that this is the final list, I think that would make
> > the Nautilus team happy.
> 
> The freeze we had agreed on was an API freeze, not a bug fix freeze,
> nor an improvement freeze.  That was the extent of the promise we made
> because Eazel insisted on a promise.
> 
> My list includes only bug fixes, and the only API breakage from the
> Nautilus perspective is the few lines that need to be removed from
> your code (streams).
> 
> Then again, as I said, software releases slip as I said before, so I
> am not sure that promising anything in the context of software release
> schedules makes any sense.
> 
> It strikes me as odd to have to `promise' something in the context of
> software releases.  If the software is not ready, I can not release
> the code.  I do not want to see Bonobo 1.0 be another GNOME 1.0, I
> learned from the past mistakes.


- This was true one year ago when you were having fun hacking on bonobo.
This is not true anymore. There is now money at stake both in your own
company and Eazel. And as far as I can tell, this is not small money
(I am not a business weenie so I cannot tell how much but I know the
figures are in tens of millions of dollars which is _a lot_)
Release schedule has thus some importance now. Also, I would like to
point out that delaying both nautilus and gnome 1.4 does no good to 
gnome either.

- It is obvious that major architectural changes will be needed for 
gnome 2.0. As far as I can remember, bonobo was to be part of the 1.4 
platform so that gnumeric/nautilus/evoltuion could use a stable released
version but we were not to try to advertise this as a cool platform to
use for everyone. ie: "we use it, please, do not use it.".

As a consequence, I see absolutely no reason why there could be no reason
not to create a stable/unstable branch of bonobo in cvs because it will 
be broken later anyway. Oh, sorry, I forgot: 1.0 _will_ be perfect.


As as side note, I find myself badly personally fucked up by the current
bonobo release schedule since it seems that bonobo 1.0 will be advertised 
as a stable platform.
My editor might be kind of pissed off to learn that while we have been
targetting gnome 2.0, there is going to be another stable platform
we could have tried to target.


Of course, I guess no one cares about this either...


Mathieu
-- 
Mathieu Lacage <mathieu eazel com>