On 04/05/2015 07:02 AM, Bill Page wrote:
>> Eltable(S : Type, Index : Type) : Category == with

> Do you notice anything strange about the documentation of this
> category? S is apparently the "domain" (D) and  'Index' is the
> "co-domain" (I) of the function represented by % referred to
> initially. So the name 'Index' seems like a very poor choice!

True. And I think, we should change the arguments to (Dom, Im) as they
are in EltableAggregate.

> Are there any domains in FriCAS that one might conceivably use as an
> index domain which does not export equality?

I guess not. It's even hard to find a domain in FriCAS that is not of
type BasicType.

In fact, in the long run, I would like to remove non-constructive
functions. Or at least adapt their specification. Equality of streams,
for example, is not computable in general. So Stream(X) should not
inherit from BasicType. We can replace that with another function

  eq: (%, %) -> Union(Boolean, "failed")

whose specification explicitly says that eq(x,y) might run forever or
return "failed", if the respective algorithm isn't able to decide about
equality. But that's for the future since it needs quite some thought
about what other part of FriCAS it affects.

Ralf

-- 
You received this message because you are subscribed to the Google Groups 
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/fricas-devel.
For more options, visit https://groups.google.com/d/optout.

Reply via email to