Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!think.com!zaphod.mps.ohio-state.edu!wuarchive!uunet!munnari.oz.au!goanna!ok From: ok@goanna.cs.rmit.oz.au (Richard A. O'Keefe) Newsgroups: comp.lang.prolog Subject: Re: Specifying types Message-ID: <6333@goanna.cs.rmit.oz.au> Date: 14 Jun 91 04:05:27 GMT References: <9106111357.AA29175@ucbvax.Berkeley.EDU> Organization: Comp Sci, RMIT, Melbourne, Australia Lines: 33 In article <9106111357.AA29175@ucbvax.Berkeley.EDU>, POPX@vax.oxford.ac.uk (Jocelyn Paine) writes: > Is there a generally accepted notation for specifying types in Prolog? No. > Are any of these programs in wide > enough use to merit adopting their notation for types as a de facto > standard? No. There isn't even wide agreement on what kind of types we ought to have. For what it's worth, Bristol's new Goedel language uses essentially the Hindley/Milner/Mycroft types, modulo minor syntactic details. > If so, are any of them in the public domain? I have the type-checker > from Edinburgh Tools, but it relies on some oddities of DEC-10 Prolog > (won't run under Poplog for example), and doesn't handle modules. I am unaware of any dependencies on DEC-10 Prolog. It took very little effort to make it run under C Prolog, Quintus Prolog, SICStus Prolog, NU Prolog. If there are any problems, please let me know. Of course it doesn't handle modules. There is even less agreement about modules than about types. (The latest ISO suggestion I've seen is unacceptably complex. I mean by that that if it becomes standard I will join the Pop camp and forget about Prolog; it'll be too hard to use reliably.) As far as I know, the type checker in the DEC-10 Prolog library (what is the relationship between the DEC-10 library and "Edinburgh tools"?) is the _only_ Prolog type checker which has ever been broadcast on the net. -- Q: What should I know about quicksort? A: That it is *slow*. Q: When should I use it? A: When you have only 256 words of main storage.