Re: [Caml-list] What is an applicative functor?

[Jacques Garrigue <garrigue@math.nagoya-u.ac.jp>]

On 2011/04/08, at 17:20, Jacques Garrigue wrote:

> On 2011/04/08, at 17:04, Alain Frisch wrote:
>> I'm not sure the benefits of applicative functors are worth the extra complexity they induce, especially if you want to support applicative functors in the same system as well. As far as I understand, applicative functors are useful primarily to give precise types to higher-order functors (which are rare enough).
>> 
>> It would be interested to see how much effort would be needed to adapt large code bases to compile with -no-app-funct (it probably amounts to giving a name to the result of a few functor applications).
> 
> Applicative functors have other advantages, like the fact you can refer to a
> type produced by a functor without having really applied it.
> 
> For instance think of the following functor definition
> 
>   module F(O : Set.OrderedType)(S : sig type t = Set.Make(O).t val union : t -> t -> t end) = ...
> 
> If you want to do the same thing with generative functors, I believe you have to
> pass the result of Set.Make(O) around physically.
> I do think this is a significant weakness.

Another example of that is my conversion-based version of the expression problem.

	http://www.math.nagoya-u.ac.jp/~garrigue/papers/mixmod.ml.txt

With -no-app-funct, most of the code is refused.
Note that -no-app-funct not only refuses paths of the form F(X).t, but also F(X).S (i.e. a functor returning a signature).
Actually, I wonder why it does refuse it syntactically, whereas potentially problematic signatures are very few.

Jacques Garrigue