On Jun 8, 2014, at 5:10 PM, Jed Brown <[email protected]> wrote: > Blaise A Bourdin <[email protected]> writes: >>> We’ve been over a year since a release which is too long. >> >> That’s fair. >> >> As a matter of fact, if the releases are not as far apart, skipping one is >> not a big deal. > > I think we should move to a faster release schedule. At least once > every six months from here out.
You’ve been advocating this for a while and I don’t disagree; so why hasn’t it worked out? Do we really have to state publicly a release date and then force it upon ourselves to make these regular releases? Is it worth it? I’m just asking, I’m not opposing. Barry > Releases are easier that way and there > would be less motivation to try to cram more features in before a > deadline. Though this means that a somewhat higher fraction of users > might have an older version of PETSc (and it might be worth our while to > do bug fixes against an older 'maint').
