Strategy for ST4

[Thom1729]

The current version runs on 3.3 and doesn't have any of the ST4 features. Obviously, a future version will run on 3.8 and have all of the ST4 features.

Is it worth having an intermediate version that runs on 3.3 but relies on ST4 features? The main reason I lean toward “no” is that this would effectively require two new major versions — sublime_lib 2.0 for 3.3 on ST4 and sublime_lib 3.0 for 3.8 on ST4. I'm not sure it's really worth it, but I could be persuaded.

[FichteFoll]

Is it worth having an intermediate version that runs on 3.3 but relies on ST4 features?

Unlikely. We haven't added new features in a while and there hasn't been a high demand either. I say we break compatibility as soon as we see a good reason to do so and then only apply bugfixes to the last minor release branch that supports ST3 & 3.3.

That is of course assuming that we get momentum towards and a base for ~~dependencies~~libraries in 3.8, because there would be no reason for us to abandon 3.3 if 3.8 isn't usable.