Arnaud Spiwack

So building `base` and `nofib`? There will probably be a few eta-expansions to do along the way. Would that be acceptable? If so, then yes, we could do it.

Ouch. I'm sorry. Your question requires more than a 5min answer, that how it ended up slipping my mind. I'll give my thoughts by tomorrow (Saturday) night.

Ok, I'm back. Yes, your point is a very good one. I was, ironically, thinking in terms of mutable data structures, in which case there is no problem of going...

> One thing that doesn't feel right with your proposal is that `merge` is also unsafe and `rebuild` is exported. > So going back to find and canonicalize: I wouldn't...

(This is getting very off topic, but this is very neat! I'd love to see you demonstrate it to me some time)

I've got to say that I really like the idea of separating, at least in part, the notion of namespace from the notion of modules (some may remember me championing...

> Is the idea of associating a name space to inductive types part of the proposal? I'm rather an outsider these days, but for the record I'd very much like...

Something like this is very welcome. The current syntax is not ideal (and often confusing). I don't know whether it should all be purely inferred, as here, or if there...

You should probably explain in your CEP while you want to roll your own library rather than using an established one like [cmdliner](http://erratique.ch/software/cmdliner).

Just to be clear: I don't want to say that cmdliner should be the solution. I'm saying that, if you are not using it, the CEP document should explain why...