Hi folks.  For the past 3 years, we've had a significant OpenSimulator release 
at about 4 to 8 month intervals.

As the last major release (0.7.6) was in September 2013, we are due for another, and this seems like a reasonable point. From my perspective, releases drive a very significant level of OpenSimulator adoption. It's also the case that I may have a limited window of time to act as release manager, after which things may get very busy for me for an extended period.

There was some disagreement over how the last release was handled, particularly over the period when I asked that developers take care to leave master stable. However, it's quite hard for me to see how this could be done in another way without significantly more manpower. Even when the release is branched, if the release branch and master diverge significantly (or master is broken) then a lot of the testing done by users is lost - very few people actually pick up and test release candidates.

So unless there is disagreement, I'm going to ask that people take care to keep master stable until the end of April or release, whichever occurs first. This certainly does not mean a halt to development on master, it's just a request that people take a common sense approach to holding off from putting in destabilising or very complicated changes for a while, unless these are to fix significant bugs or regressions.

I'm very happy to hear alternatives, though I would also need people to help if they involve more work. In any case, Robert has very kindly volunteered to look at some of the current regressions but more help is always very welcome.

The current regressions or very major bugs have a 0.8 target in Mantis. As with every release so far, only major regressions are guaranteed to receive some sort of attention, where the answer may still be WONTFIX for this release.


Justin Clark-Casey (justincc)
OSVW Consulting
Opensim-dev mailing list

Reply via email to