Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!ljdickey
From: ljdickey@water.UUCP
Newsgroups: comp.lang.apl
Subject: Re: APL puzzle based on Unix' Escapes
Message-ID: <826@water.UUCP>
Date: Thu, 5-Mar-87 09:56:19 EST
Article-I.D.: water.826
Posted: Thu Mar  5 09:56:19 1987
Date-Received: Thu, 5-Mar-87 20:55:59 EST
References: <29000002@infbs.UUCP> <4000002@hpcvck.HP> <427@net1.UCSD.EDU>
Organization: U of Waterloo, Ontario
Lines: 38

In article <427@net1.UCSD.EDU>, graifer@net1.UCSD.EDU (Dan Graifer) writes:
> In article <4000002@hpcvck.HP> paull@hpcvck.HP (Paul Liebert) writes:

> Thus,
> 
>     {nand}\1 1 0 0 0
> 1 0 0 1 1
>     {comment} not 1 0 1 1 1 

>                               Dan Graifer
>                               graifer@net1.UCSD.EDU
> Disclaimer: Nobody ever listens to me anyways; Why should they start now?


Dan is right.

APL\11 has had a problem with scan for some time and it seems that
nobody is fixing it.  You could check this expression:
 
	-\ {iota} 5

The result should be  1 -1 2 -2 3.
The APL\11 gives      1 1 2 2 3.
 
It is just wrong.

The essential test for correctness of scan is this:
 
		( f \ V ) [N]    <----->   f / N {take} V

The APL\11 gets the right hand side right but not the left hand side.

I think I know the solution, if anyone want to try to fix it.
-- 
 L. J. Dickey, Faculty of Mathematics, University of Waterloo. 
 ljdickey@water.UUCP    ljdickey%water@waterloo.CSNET
 ljdickey%water%waterloo.csnet@csnet-relay.ARPA
 ljdickey@water.BITNET		UUCP: ...!watmath!water!ljdickey