Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!ukc!mucs!r4!mshute
From: mshute@r4.uucp (Malcolm Shute)
Newsgroups: comp.arch
Subject: Paper tape vs card structuring
Message-ID: <983@m1.cs.man.ac.uk>
Date: 20 Feb 90 17:07:52 GMT
References: <1639@aber-cs.UUCP>
Sender: news@cs.man.ac.uk
Reply-To: mshute@r4.UUCP (Malcolm Shute)
Organization: University of Manchester, UK
Lines: 24

In article <1639@aber-cs.UUCP> pcg@cs.aber.ac.uk (Piercarlo Grandi) writes:

>[...] the difference between encoded (paper tape) and structural
>(card deck) representations of data. In encoded representations the
>relationships among parts of composite data are encoded within the data, in
>structural representations they are part of the data structure; [...]
>I know of languages that are mostly structural (Lisp) [...]
>[...] most architectures are structural (machine programs are represented as
>graphs), but some could be argued to be encoded (e.g. reduction machines).

But machines which employ String Reduction are generally executing
something which looks like low level Lisp, so by your definition will
be 'card-deck' machines; and machines which employ Graph Reduction must
surely fit also into your 'structured representation' category.
So I am unclear about your definitions here... can you elaborate.

(Incidentally, since all functional and single assignment languages
can be shown to be mere syntactic variants of each other, a machine
which has a low level functional language as its machine code
must have the same attributes as high level functional languages.
I'm assuming that you are talking about functional-Lisp here (pure-Lisp),
of course).

Malcolm Shute.         (The AM Mollusc:   v_@_ )        Disclaimer: all