Eugen Leitl <[email protected]> writes:

> On Wed, Sep 04, 2013 at 04:28:52PM -0700, Simon Forman wrote:
>
>> There is a (the?) universal logical notation being elucidated right now that 
>> seems to me to be very promising for this sort of stuff.
>
> Is it intrinsically massively parallel? If it isn't, it's probably
> not going to go places.

I don't think I've ever seen a logical framework which isn't
intrinsically massively parallel; ie. it seems to me like sequencing is
always an explicit construct, whether it's applicative functors, arrows,
monads, linear/uniqueness types, etc.

There are cases like Prolog and Pure where the evaluation/search order
is defined, but I would still represent sequences of instructions using
something like a list (applicative functor) or tree (free monad),
freeing up implementations to find the elements in parallel whilst
ensuring they're executed in sequence. Maybe I've been using Haskell for
too long? ;)

Cheers,
Chris
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc

Reply via email to