translatenix

I can't think of a concrete problem caused by having one module per source of truth. I can think of several problems caused by having multiple modules per source of...

Comparing with other languages is a good idea. Ideally we could compare with other nominally typed config languages. > I can't think of a real-world scenario where treating symlinks as...

Could you check if Pkl built from the main branch gives a better error message that doesn't include "Would you mind filing a bug report?"? I'm asking because my `HttpClient`...

FWIW, `CertificateUtils` is gone in HEAD.

I haven't tried, but I'd expect gh to be unaffected. Plain git is definitely unaffected.

Correction: `gh pr merge` does honor these settings. So you'd need to use plain git to intentionally do a rebase-merge.

> it's not really clear how we can add proper syntax highlighting here. Would it be possible to support pluggable highlighters that are told which renderer is used?

My understanding is that accessing `bar` from within a type constraint isn’t allowed because it isn’t `const`. If you don’t get an error for that it’s probably a bug.

`foo.bar.baz.qux {...}` looks very similar to expressions `foo.bar.baz.qux` and `(foo.bar.baz.qux) {...}`, which complicates parsing and reading. I think a different syntax should at least be considered.

> Actually, the (...) in object amends syntax disambiguates that entirely. Distinguishing `foo.bar.baz.qux {…}` from `foo.bar.baz.qux` requires arbitrary lookahead. It’s also difficult to parse for a human, especially in the...