[Nix-dev] Typing nix − funding campaign

[Peter Simons]

Hi Théophane,

 > The inference algorithm that is going to be used is no trivial at all
 > to implement.

my understanding now is that your thesis adviser has invented (or
significantly contributed to) a new type system, and the purposes of
your project is to test that particular system in practice using Nix as
a case study so that you can gain insights into the nature of that type
system and find possibilities to extend it.

In other words, there is no way the outcome of your research will be
something like "Hindley-Milner works great for Nix and all you need is
the following minor changes to the language to allow for explicit type
annotations". That's not going to happen, because you'll never examine
whether Hindley-Milner would, in fact, be sufficient for our purposes.

Is that a fair description of the reality?

Best regards,
Peter