Newsgroups: comp.lang.prolog
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!news.sei.cmu.edu!cis.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!walth
From: walth@netcom.com (Walter Howard)
Subject: How can I get Prolog accepted?
Message-ID: <X5Ypk0uRWUaS068yn@netcom.com>
Sender: walth@netcom.com (Walter Howard)
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
Date: Sat, 19 Nov 1994 16:01:37 GMT
Lines: 33

My company uses C++ and COBOL for a huge loan tracking system. 

New projects coming up include Loan Origination, which is a program
to select good candidates to be solicited for further loans.

I used Prolog about 10 years ago ( Turbo Prolog ). It twisted my
brain but I now know declarative programming is HOT. Much hotter than
Object Oriented.

Probably when all is said and done what we'll have is a hybrid of
Procedural, OO, and Declarative.

What I want to do is create Inference Engine Objects that can be
called from higher level C++ code. Because of the rule based nature
of Prolog the method for finding good loan candidates should be
easily programmed and easily changed, which will happen quite often.

For this I need source code for a tiny Prolog interpreter than can be
encapsulated as a C++ object. I am using OS/2 so I conceptually
have infinite stack space (virtual).

I would then have to give them a demonstration of how easy it is to
change a rule and note the change in the resultant program output.

Any suggestions.



-- 

+=============================================================+

You may think I'm crazy, but remember, you heard it here first!

+=============================================================+

walth@netcom.com
