Re: getting on a longer release cycled



Hi,

On Thu, 2006-09-07 at 12:59 -0400, Hubert Figuiere wrote:
> There have been a large pimping of project Topaz, and I strongly believe that 
> this is the kind of goal that would need a longer development cycle for a big 
> leap forward towards world domination.

The counterargument here is always that this development can and should
go on, but should be done on a branch and merged in when it's considered
ready.  This was the idea behind libegg too: once code there was
considered "ready" it would move into GTK+ or some other more stable
library.  It's worked quite successfully for new software developed
entirely outside of GNOME (deskbar and tomboy being two recent
examples).

I'd agree that our release methodology is geared more toward releasing
in stages leading up to a stable release and that we don't have a good
way to release "preview" software that might be going on in a branch,
but we also have a problem that we don't get probably enough testing on
our actual unstable branches leading up to a release either.

We've got a good thing going with our six month release cycle.  Not all
the interesting work we do on our existing code has to happen within
these cycles, though.

Joe




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]