Jon Brandvein

Greg's right that it was an intentional decision to not make the list of available providers on a target programmatically accessible to client code. (Even for native code inside Bazel...

Sounds like the request is for there to be a string-valued equivalent to `label_setting` / `label_flag`. (This seems like a slippery slope to doing the same for all attr types...

What's the resolution w.r.t. ErrorProne? Is it decoupled from the Bazel version, or do we no longer update its version outside of a major Bazel release? Note that even with...

What I mean is the issue title is asking us to decouple ErrorProne, and the closing comment is that we have a policy. So which is it -- did we...

Maybe I'm making a mountain out of a molehill -- If we want to upgrade the ErrorProne version in the future while following backwards compatibility policy, is it technically easy...

> Would it be still OK, to bump EP version in major Bazel releases, like upcoming Bazel 4.0.0 release. That would still break all Bazel users but we wouldn't care,...

+@comius, I don't really know much about stamp so maybe you can comment?

Ping @comius for triage.

> Are you suggesting the truth-value of a type (e.g. list) be described in the section for lists, not under Booleans or the bool(x) function? Or are you suggesting that...

> Every language provides support for the machine's basic ALU operations, and Skylark should too, even if Bazel doesn't need them. Looking at previous times we've discussed this, there's also...