I love this idea so much and I 100% agree that this is what we need to do but I’m not 100% clear on how best to do that. I do think that this ties into the repo clean up / organization that we’re doing.
It feels like we’re in a bit of a transition point from “we’re missing a bunch of basic features and core things aren’t right so lets move fast and add them” to more of a “hey things are pretty good now, let’s take our time and make sure we aren’t regressing something critical to add a feature that is more of a ‘nice to have’ than a ‘need to have’”.
I think that we have basically been doing an adhoc version of branching releases, but all in one branch. Basically we get a good stable version then we change things and break a bunch of stuff and then the next two version or so we refrain from changing anything other than bug fixes to get back to stable 1.12 was a good stable release and I’m hoping that 1.15 is as well.
We need to do a much better job of clearly labeling those and acknowledging what they are.
Our big bottleneck is testing. Because any change truly does have the potential to break everything else a LOT of testing is needed to make sure that things are good which takes hours. I would guess that it takes about 20+ hours of testing for me to feel good about a release which is a lot of time.
I have been posting basically branches for testing here: Interstitial Firmware Releases - #571 by dlang
but that could be a lot more organized.