I am missing subtypes on the abstract level mainly because I would like
to reduce overgeneration by restricting syntax rules (as constructor
functions which are total on their domain) to suitable subcategories.
На 18.02.2015 г. 9:55 "Peter Ljunglöf" <peter.l...@heatherleaf.se> написа:
> Personally I would prefer subtypes to dependent types:
>
> - after 15 years of GF, almost no grammars make use of dependent types, so it's probably not extremely useful in practise
I agree but this is a major backwards incompatibility.
--
---
You received this message because you are subscribed to the Google Groups "Grammatical Framework" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gf-dev+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
На 18.02.2015 г. 10:05 "John J. Camilleri" <jo...@johnjcamilleri.com> написа:
>
> Interesting point Peter.
> The C runtime still doesn't support dependent types, does it? So in some sense, this break with compatibility has already started.
Not really. They are just not implemented yet because I don't need them for the translator. This is also why the Haskell runtime is still used for the GF shell.
--