home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!mojo.eng.umd.edu!disney.src.umd.edu!eng.umd.edu!clin
- From: clin@eng.umd.edu (Charles Lin)
- Subject: Re: AI Programing Language
- Message-ID: <1992Sep15.025411.19354@src.umd.edu>
- Sender: news@src.umd.edu (C-News)
- Reply-To: clin@eng.umd.edu (Charles C. Lin)
- Organization: College of Engineering, Maryversity of Uniland, College Park
- References: <95957@bu.edu>
- Distribution: usa
- Date: Tue, 15 Sep 1992 02:54:11 GMT
- Lines: 47
-
- In article <95957@bu.edu>, robrod@csa.bu.edu writes:
- >
- > I am just starting to look at AI (I am a senior Undergraduate in CS). I
- > read that the new movement in AI programming was to move away from
- > Languages like LISP and Prolog and into Object oriented languages like
- > C++, Smaltalk, etc. Is it worth studying LISP and Prolog anymore?
- > What should someone concentrate on for AI programming?
- >
-
- I'm no expert either, but I think LISP is still the language
- of choice for AI programmers. Since so many people in AI
- learned this language and probably know a lot of the intricacies
- in LISP, they aren't likely to throw what they know away
- just to learn a new language. There are still many features
- in Lisp that I'm sure are harder to carry out in other
- languages. I would personally say learn as many languages
- as you can. While I'm no expert in LISP, programming in LISP
- was a far different experience than programming in C.
-
- In any case, I am guessing that the languages that would
- be useful to you are C, C++, and LISP. C because everyone
- has to know it. It is so widely used that that in itself is a
- reason to know it. C++ since OOPs seem to be getting more
- popular and it's becoming more widely available. LISP,
- since it is still widely used in the US AI community.
- Trends are very slow to come about.
-
- For example, functional language folks will tell you that
- languages such as ML, Miranda, and Haskell are the better
- "functional" languages (and say that LISP isn't even one)
- and that these languages are better for learning programming
- than some language like C. Yet, the success of a language
- has a lot to do with non-esoteric features such as
- availability, having a fast compiler that makes fast code,
- having a debugger, having good references, having useful
- features (such as dealing with files or good I/O), etc.
-
- You probably couldn't convince a functional language
- purist to use LISP, but I don't see the AI community
- dumping LISP in favor of something else.
-
- Now you could be right, since I'm more of an AI novice,
- so I'm curious to hear other opinions on this topic.
-
- --
- Charles Lin
- clin@eng.umd.edu
-