[mercury-users] Testing with dependent lists?

[doug.auclair at logicaltypes.com]

Dear Ralph, you wrote:

[snip my question on if it's possible in Mercury to have the inst
of one object depend on the inst of another in a pred decl]

>Not to do what you want here.  The subtyping aspect of insts is
>occasionally useful, but it isn't a magic bullet.  Sometimes the
>easiest (and best) thing to do is just add a couple of extra clauses or
>use an if-then-else.

*sigh* Not the answer I wished for, but thank you for your replies.

As to the magic of dependent type bullets, the last FLoC (nee ICLP)
seemed to have more than a few sessions on their utility (in fact, the
ICFP (that Zoltan attended) had a keynote on that topic)) so dependent
types are, oh, I'd say, within 20 years of the mainstream (again, 
*sigh*), and, actually, Mercury goes much further than many other 
general-purpose (and used) programming languages in providing most
of the features that dependent types provide ... so, thank you,
Mercury team, for providing these features.

Sincerely,
Doug Auclair

P.S. Just a hypothetical question: how would one go about implementing,
e.g., functional 'insts' along the lines of my inquiry?  Or, is this question 
a "That's a good PhD thesis" question?  And, if it is, then (waaaay *OT* 
warning) what kind of funding is necessary to get that rolling as well as 
the expected turn-around time into a ROTD?

Yes, I believe in the Aladdin factor.

--------------------------------------------------------------------------
mercury-users mailing list
Post messages to:       mercury-users at csse.unimelb.edu.au
Administrative Queries: owner-mercury-users at csse.unimelb.edu.au
Subscriptions:          mercury-users-request at csse.unimelb.edu.au
--------------------------------------------------------------------------