[Numpy-discussion] Going toward time-based release ?
Sun May 11 23:12:50 CDT 2008
On Sun, May 11, 2008 at 7:59 PM, David Cournapeau
> I would like to know how people feel about going toward a time-based
> release process for numpy (and scipy). By time-based release, I mean:
> - releases of numpy are time-based, not feature based.
> - a precise schedule is fixed, and the release manager(s) try to
> enforce this schedule.
Just as a data point, Brian and I just discussed proposing the same
idea to the ipython crowd. We've been in semi-stalled mode for a
couple of months (life hit us hard and at the same time, and we just
had too many balls up in the air at the same time). I'm offline this
week, but in a few days we'll float the idea by the others on the
ipython list to see what they think.
I should note that the ipython circumstances are very different: numpy
has a vastly larger team and is not in any way stalled. Hence
consider the above just an anecdotal reference of a 'cousin' project,
the specifics of numpy are different and I understand that.
FWIW, I'm actually +1 on the idea for numpy as well, but my vote
doesn't count because I don't really contribute almost any code
More information about the Numpy-discussion