Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!ukc!slxsys!bae-st!adam
From: adam@ste.dyn.bae.co.uk (Adam Curtin)
Newsgroups: comp.lang.apl
Subject: Re: outer product
Message-ID: <1990Aug1.073227.4636@ste.dyn.bae.co.uk>
Date: 1 Aug 90 07:32:27 GMT
References: <16bc25c18b1226ab2c5c@canremote.uucp> <1990Jul30.202542.16254@eng.umd.edu>
Reply-To: adam@ste.dyn.bae.co.uk (Adam Curtin)
Organization: British Aerospace (Dynamics), Stevenage
Lines: 23

In article <1990Jul30.202542.16254@eng.umd.edu> stripes@eng.umd.edu (Joshua Osborne) writes:
>In article <16bc25c18b1226ab2c5c@canremote.uucp> elliott.chapin@canremote.uucp (ELLIOTT CHAPIN) writes:
>>I have figured out a way to extend the outer product to any
>>binary function.  Is this old hat?
>
>Some versions of APL aready do.  I think APL2 lets you use any dyadic function,
>including user defined ones.

It's old hat. In Dyalog APL, a defined function is not distinguished from a
primitive function or a derived function (built from other functions using the
compose operator). Dyalog also has _external_ functions (gateways to APs) which
also have full rights. This technology circa 1982. Recently, Dyadic has added
defined _operators_, which are syntactically equivalent to primitive operators
and truly boggle the mind.

Adam.
-- 
A. D. Curtin				Tel  : +44 438 753430
British Aerospace (Dynamics) Ltd.	Fax  : +44 438 753377
PB 230, PO Box 19, Six Hills Way,	Email: adam@ste.dyn.bae.co.uk
Stevenage, SG1 2DA, UK.  		UUCP : ...!uunet!mcvax!ukc!bae-st!adam
	Arse-covering  :   <This disclaimer conforms to RFC 1188>
	Fun-to-be-alive:	"My other car is an FJ1200"