home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.programming
- Path: sparky!uunet!gatech!rpi!batcomputer!cornell!karr
- From: karr@cs.cornell.edu (David Karr)
- Subject: Re: Programming by Description of Output...
- Message-ID: <1992Dec31.155736.29068@cs.cornell.edu>
- Organization: Cornell Univ. CS Dept, Ithaca NY 14853
- References: <1992Dec30.174004.4168@cs.cornell.edu> <C036t1.KCH@phage.cshl.org> <BEVAN.92Dec31001940@panda.cs.man.ac.uk>
- Date: Thu, 31 Dec 1992 15:57:36 GMT
- Lines: 28
-
- In article <BEVAN.92Dec31001940@panda.cs.man.ac.uk> bevan@cs.man.ac.uk (Stephen J Bevan) writes:
- >Getting back to the sorting example: if the arrays are changed to
- >lists, then a spec is :-
- >
- > sort(List, Sorted_List) :-
- > permutation(List, Sorted_List),
- > ordered(Sorted_List).
- >
- >Suitable definitions of `permutation' and `ordered' are left as an
- >excercise for the reader.
-
- My impression is that most of the work of writing this specification
- is in the spec of "permutation", a smaller amount is in "ordered", and
- the part shown above is trivial in comparison. So the above example
- doesn't make much of an impression on me.
-
- >[...] Someone here did some
- >work on automatically transforming Prolog programs (the thesis
- >specifically covers generation of various sort routines from the above
- >spec).
-
- Good, I figured someone would be working on this. I could be surprised
- by the results.
-
- -- David Karr (karr@cs.cornell.edu)
-
-
-
-