Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!ames!pasteur!ucbvax!decwrl!sun!pitstop!sundc!seismo!uunet!munnari!mulga!ttey
From: ttey@mulga.oz (YEO Tee Thiam Eric)
Newsgroups: comp.lang.prolog
Subject: Re: The meaning of "declarative"
Message-ID: <2997@mulga.oz>
Date: 8 Oct 88 04:06:06 GMT
References: <152@s1.sys.uea.ac.uk>
Distribution: comp.lang.prolog
Organization: Comp Sci, Melbourne Uni, Australia
Lines: 18

in article <152@s1.sys.uea.ac.uk>, kh@s1.sys.uea.ac.uk (Kevin Hammond CMP) says:
> But Lisp has assignment, I/O and other imperative features. How can it
> be declarative?    Lisp is   functional   yes  (it has   higher  order
> functions), but declarative no.

I believe that "declarative" is a relative adjective when used in the
context of programming languages. When talking about programming languages,
one should talk about the degree of "declarativeness". If one say that
Prolog is a declarative language, one could also argue that there are
features in Prolog which makes it "non-declarative", e.g. the cut.

Certainly, there is a subset of Lisp (pure Lisp) which conforms to our
notion of "declarativeness" as there is a subset of Prolog. Instead of
branding a programming language as "declarative" or "non-declarative"
we should consider the features of the language which encourages the
declarative style of programming.

					Eric Yeo