Leo White

> Are there examples that would look nicer if and+ was the zipping one, instead of an alias for and*? (How do I express the zipping product if and+ is...

The difficulty around `and` can be explained as follows. For any binding construct there are essentially two forms: a `bind` form and a `map` form. The first allows the body...

> In such a case the following example given in the proposal [...] would, if I'm not mistaken, fail to typecheck That's correct, this example should have: ```ocaml type unboxed_t...

The feature is easy to implement and maintain, so I don't object. However, I don't think it is the best way to achieve your aims in the medium to long...

I certainly don't think we should stop supporting cmt files. They are occasionally useful for the kinds of things mentioned above. But I would like to get to a point...

> I mean, the previous RFC, if merged, would also roughly divide the size of cmt files by two. I'm not sure that the size is worth it even if...

> OCaml has a high-level semantics where those attributes are ignored Note that this is not entirely the case because the representation needs to be taken into account when checking...

> My understanding is that the part of the build that would need to be duplicated (parsing, type-checking and bytecode compilation) takes only a fraction of the time, compared to...

Copying my response from the original thread, as it still represents my thoughts on this proposal: > All in all, I'm still not convinced by this proposal as compared to...

Could we see the examples from Nomadic labs? I think the details of such examples are important to the discussion. It would also help if someone who thinks that these...