[palerdot]

> But in their eyes, Elm is not done

I think this is one of the major themes of all the 'Elm dramas' that we have been reading and hearing about. Elm core team should stop recommending it as a stable platform that people can try out (which Elm clearly is not).

[anthonybullard]

I've personally never seem them say "the language is done, you can expect no breaking changes". Hence the fact that it is v0.19

[ziotom78]

I have read several times that they do not consider the language stable. Yet they repeatedly advertise it as «production-ready» [1,2], which is contradictory at best.

[1] https://discourse.elm-lang.org/t/two-experiences-with-elm/91...

[2] https://github.com/gdotdesign/elm-github-install/issues/62#i...

[palerdot]

> the language is done, you can expect no breaking changes

Well ... I don't use Elm in production. But, I'm sure there are people out there who got gaslighted because of this ill-conceived move, and they can better explain about 'production readiness' story that was weaved around Elm.

> Hence the fact that it is v0.19

I also don't understand how can a minor version 0.18 -> 0.19 completly break backwards compatibility even if it is not stable. Well may be it's just Elm. They should have atleast thrown a `deprecated` warning and maintain backwards compatibilty till the next major version.

[yhoiseth]

It's in line with semantic versioning, point 4 [1]:

> Major version zero (0.y.z) is for initial development. Anything MAY change at any time. The public API SHOULD NOT be considered stable.

1. https://semver.org/#semantic-versioning-specification-semver

[palerdot]

> Major version zero (0.y.z) is for initial development

> Anything MAY change at any time.

Thank you for clarification. So, now it begs the question how did Elm team persuaded people to try out their language in production if it is still in initial development and anything MAY change at any time ...