Should Pinta 2.2 really be Pinta 3.0

[JGCarroll]

Pinta 1.7 was a huge update 5 years in the making, but was still seen as a minor patch despite revitalising Pinta for a lot of users.

2.0 was the GTK3 release. If we use Semantic Versioning, it makes sense, since there was some functionality loss (addins) despite the majority being gains.

2.1 was a nice set of improvements over 2.0 but still on GTK3.

For nothing other than vanity and public relations, would it make sense to rebrand the GTK4 port as Pinta 3.0? From my personal observations of the Linux community, they really like GTK4 and see it as huge wins by itself. The Mac/Windows/BSD communities may disagree, but I can't imagine they'd really be bothered about version numbers.

Through nothing other than human psychology, bigger number = better number; and so although rebranding as 3 this late on might mean nothing from a technical POV; it'd be good for drumming up PR on Reddit/etc.

[cameronwhite]

I haven't thought much about the version number yet, but those are good arguments :)
The port to GTK4 / libadwaita (and switching to a different library for the GTK bindings), along with restoring add-in support, has made it at least as large a release as 2.0 with the GTK3 / .NET 6 port

[Lehonti]

What about increasing the number version when there are major breaking changes in the API? Currently what can be done through extensions is limited by what the API allows, and the API reflects the capabilities of the software, of course (we got a taste of this when I wrote the first non-tileable effect), and in order to expand the things one can do with third-party extensions, the API would basically have to be re-written at some point. Increasing the version number would signal a 'clean break' to both developers and users.