Gabriel Scherer

I don't think this is enough, `OCAMLPARAM` is fragile and it is not clear when it is the correct design to use it. I would support a PR to introduce...

I find it weird that build system's implementation would need to be modified because a compiler suddenly gained the ability to use colors on certain cases. The compiler is trying...

Thanks for the precise recommendation: it sounds reasonable and I can follow it. If you think that an `OCAMLBUILD_COLOR` environment variable is a good design for ocamlbuild, I suppose that...

Yep, the correct cascading order is "env < cli < file-local settings".

I would let the code speak. I think the most natural place to test `OCAML_COLOR` is the `should_enable_color` test in utils/misc.ml. If you do this, then it will be called...

It is not clear to me why you say that this falls under the same issue as the comment in #185. Here the ideas is to add new options in...

We improved the printing of cycles in #44 already, but this PR was actually suggesting a hint in the specific case of module self-reference, where you mistakenly write `Foo.bar` within...

The error message is produced by ocamlbuild, which asks `ocamldep` for dependencies and complains that there is a cycle; I don't think that ocamlc is invoked on `foo.ml` at all.

To summarize: at the time where we first attempted to make `-use-ocamlfind` the default, we conducted the change rather poorly: we tried to make it the default in the same...

I feel that there is some consensus on changing the default, but I don't want to do this for 0.11, we have enough potentially-breaking stuff to deal with.