home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.ai:4787 comp.compilers:2109
- Newsgroups: comp.ai,comp.compilers
- Path: sparky!uunet!world!iecc!compilers-sender
- From: dyer@airplane.sharebase.com (Scot Dyer)
- Subject: Re: Different Strokes for Different Folks (Was: Assessing a language)
- Reply-To: dyer@airplane.sharebase.com (Scot Dyer)
- Organization: NCR/ShareBase Corporation
- Date: Thu, 7 Jan 1993 19:06:35 GMT
- Approved: compilers@iecc.cambridge.ma.us
- Message-ID: <93-01-036@comp.compilers>
- References: <93-01-016@comp.compilers> <93-01-018@comp.compilers>
- Keywords: design
- Sender: compilers-sender@iecc.cambridge.ma.us
- Lines: 39
-
- eifrig@beanworld.cs.jhu.edu (Jonathan Eifrig) wrote
- > For each style there is (or should be!) an appropriate
- >language. Prolog makes search-based programs easy, but nobody
- >would want to write a numerical analysis package in it. Functional
- >languages like pure ML are great for manipulating quasi-static data
- >structures, but I would hate to write a network-flow algorithm in it.
- >
- > Surprisingly, there hasn't been much work in developing
- >heterogenous programming environments, to support a sort of "mix and
- >match" approach to programming. Such tools would go a long way to
- >alleviating the language holy wars, I think.
-
- Some new AI* languages (such as POP-11 and Trilogy) take this approach.
- Both supply the programmer with the tools of the Functional,
- Relational/Logical and Imperative programming paradigms. To the best of
- my knowledge an object-oriented mix-and-match does not yet exist.
-
- Part of me finds this approach pleasing, but another finds it awkward. It
- would be nice to be able to do prolog-y things in a prolog-y way, ml-y
- things in an ml-y way, lisp-y things in a lisp-y way... etc. Where does
- one stop? A language which allowed these new language forms to be defined
- on top of it would be most aesthetically pleasing, but might not allow for
- sufficient optimization on the part of the compiler...
-
- -- Scot
-
- P.S. I don't think trilogy is commercially (or otherwise) availible yet...
-
- ==============================================================================
-
- * Don't ask me what an AI language is, or what AI is. AI is probably best
- defined as that which cannot currently be done. IMHO, Every AI success
- becomes another conventional programming trick. (Dynamic Memory, AI,
- Heuristics, etc.) By doing it, we "explain away" the intelligence. I
- call these "AI Languages" only since I think it's mostly "AI people" (a
- more ambiguous term) who use them or are interested in them.
- --
- Send compilers articles to compilers@iecc.cambridge.ma.us or
- {ima | spdcc | world}!iecc!compilers. Meta-mail to compilers-request.
-