Propose releasing as 1.0

[rossabaker]

We are due another release with #288. I propose releasing as 1.0.0:

• sbt-typelevel-0.7.7 has breaking transitive changes, so we should bump at least the minor
• Early semver is a pain in the ass, and unlike the main project, there's no compelling reason to cling as tightly to deployed version numbers here.

[danicheg]

I have no strong no against it. Although, it feels ordered, as it is aligned with the core project's zero-based versioning.

[rossabaker]

Fair enough. I just want a tradition of breaking changes in dependencies are breaking changes in the project. (One we violate in core with respect to the non-JVM platforms, out of deference to the dominance of the JVM and how many things are built on top of us already!). I'm happy with 0.18.

[armanbilge]

I just want a tradition of breaking changes in dependencies are breaking changes in the project.

I do as well. Maybe this is something we should discuss in sbt-typelevel?

sbt-typelevel-0.7.7 has breaking transitive changes

Maybe?

• The dependency bumped is not a transitive dependency (i.e. a dependency of sbt-typelevel), at least with my understanding of transitive. The dependency bump is in a tooling dependency for the downstream build that uses sbt-typelevel.
• The bump occurs in an configurable/overridable setting. A user of sbt-typelevel can restore it to the original value.

[rossabaker]

Maybe this is something we should discuss in sbt-typelevel?

Yes, I should have opened there. I didn't see a board there, but I could have opened an issue, and ... there is a board there. My fault. 🤦

The bump occurs in an configurable/overridable setting.

This is a good distinction. I was looking at failed internal builds, and version numbers that could be to blame. If we can configure our way out of it, then I'd say we don't have to bump. Should we? Bumping forces everything downstream to bump, which is more consequential in sbt-typelevel (which many things extend, public and private) than sbt-http4s-org (maybe a "leaf plugin")? But not bumping led to a patch release that left a trail of red X's. Can argue that one credibly either way.