Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!princeton!caip!seismo!harvard!h-sc1!h-sc2!moews_2
From: moews_2@h-sc2.UUCP (moews_2)
Newsgroups: net.lang.apl
Subject: Re: Nested arrays
Message-ID: <957@h-sc2.UUCP>
Date: Fri, 23-May-86 00:05:33 EDT
Article-I.D.: h-sc2.957
Posted: Fri May 23 00:05:33 1986
Date-Received: Sun, 25-May-86 17:48:20 EDT
References: <371@water.UUCP> <834@bentley.UUCP>
Reply-To: moews@h-sc4.harvard.EDU(david moews)
Organization: Harvard University Science Center
Lines: 24

In article <834@bentley.UUCP> kwh@bentley.UUCP (KW Heuer) writes:
>...
>
>I find this one easiest to understand.  Enclose always returns a scalar of
>(new) type "tree"; arrays are still homogeneous but the scalars in a tree
>array may disclose to any object; the "standard scalar" of type tree (the
>filler value for overtake or expand, corresponding to numeric zero and
>character blank) is the enclosure of an empty tree.  This is the "grounded"
>tree model....
>
>Karl W. Z. Heuer (ihnp4!bentley!kwh), The Walking Lint

   Uh, but what exactly is the "empty tree"?  If we define

   <scalar> ::= <number> | <character> | enclose(<array>)

   <array> ::= n-dimensional array of (<scalar>)'s, n >= 0

then there IS no empty tree --- every enclosure has an array inside it.  If
you want a "default" scalar of type tree, I think you'd be better off
reintroducing ONE object x0 satisfying x0 = >x0, and letting that be the
default value.   (Is this perhaps what you meant?)
--
David Moews        moews@h-sc4.harvard.EDU     ...!seismo!harvard!h-sc4