Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.6.2.16 $; site ima.UUCP
Path: utzoo!decvax!cca!ima!compilers
From: compilers@ima.UUCP
Newsgroups: mod.compilers
Subject: Re: denotational semantics
Message-ID: <136300031@ima.UUCP>
Date: Tue, 7-Jan-86 11:24:00 EST
Article-I.D.: ima.136300031
Posted: Tue Jan  7 11:24:00 1986
Date-Received: Thu, 9-Jan-86 02:35:17 EST
Lines: 19
Approved: compilers@ima.uucp
Nf-ID: #N:ima:136300031:000:748
Nf-From: ima!compilers    Jan  7 11:24:00 1986


[from ]

   When I read a denotational semantics, I think of it as an
operational semantics written in Scheme (a small Lisp dialect closely
resembling the lambda calculus).  Such games as making up
statically-scoped functions on the fly and passing multiple
continuations are standard coding practice in Scheme, but appear
nearly inpenetrable, I would imagine, to a person trained solely in
traditional algorithmic languages.

   One neat aspect of looking at things this way is that it is
conceivably possible to write a meta-interpreter that can interpret
any language, given its denotational semantics.  (I believe that this
has sort of been done: Peter Mosses' SIS (Semantic Interpretation
System)?)

		-- Richard Schooler
		Intermetrics, Inc.