[net.lang.prolog] Request For Utilities

OKeefe.R.A.%EDXA@sri-unix.UUCP (11/10/83)

From:  O'Keefe HPS (on ERCC DEC-10) <OKeefe.R.A. at EDXA>

     There are lots of Prolog utility files in the <Prolog>
directory at {SU-SCORE}.  With the exception of Barrow's
pretty-printer for terms (and maybe one or two others that
I overlooked, sorry if so) all of them are from Edinburgh
and the only contributions since they were first mounted have
been from me.

     I think the Edinburgh library has some really good stuff
in it, but I do *not* think it adequate for all needs.  There
must be LOTS of useful little files kicking around that could
be cleaned up and submitted to the {SU-SCORE} library.  If you
have a little program that simplifies propositional formulae,
clean it up and send it in !  Someone may have a use for it.
If you've got an implementation of the SUP-INF method, send
it in !  If you've got a program that checks for style errors
in Prolog, send it in !

     The source code you send should be tested, and it should
be possible for someone else to figure out what it's for in
under half an hour, but apart from that anything *useful* goes.
Files in Edinburgh syntax would probably be best, but if you
have something neat you've held back because it's in Waterloo
Prolog, send it in !  Some Waterloo Prolog users read this Digest.
Translations of existing utilities into other dialects would
also be a good idea.

     One little point: complete programs that hack the data
base are ok, but "library routines" that people might want
to include in their programs should avoid hacking the data
base if at all possible (not because data base is evil, but
to avoid conflicts with the user's data base hacking), or if
Ait is not possible to avoid hacking the data base, the
predicates hacked should be listed near the top of the file
and highlighted in some fashion so that the user knows what
he is risking.

     Due to transmission delay, my reply to Uday Reddy's
first message appeared after the second one.  My qualified
agreement with the first message does not apply to the second.