Newsgroups:   comp.lang.apl
Path: watmath!watserv2.uwaterloo.ca!torn!utcsri!geac!itcyyz!yrloc!intern
From:         eem@ipsaint.ipsa.reuter.COM (Mcdonnell, Eugene E.)
Subject:      genealogy of nonce error
Message-ID: <1993Jan2.182944.12981@yrloc.ipsa.reuter.COM>
Sender: intern@yrloc.ipsa.reuter.COM (Intern via QUADRAM)
Reply-To:     eem@ipsaint.ipsa.reuter.COM (Mcdonnell, Eugene E.)
Organization: Reuters Information Services (Canada)
Date:          2 Jan 93 18:00:02 UT


-----------Message forwarded from IPSA Mailbox-------------


no. 6871877 filed 17.10.58  sat  2 jan 1993
from eem
to   uclapl
subj genealogy of nonce error


Bruce Bathurst asks for more on "nonce error".  

When the APL\360 system was being designed and implemented at IBM Research in 
1966, the implementation often fell behind the design.  The implementers were 
the target of frequent complaints from users inside IBM that certain language 
features described in the documentation (written by the designers) produced no 
result or an incorrect result.  To relieve their frustration the implementers 
introduced a new error, the nonce error, which indicated that the piece of 
syntax attempted would not produce the desired result "for the nonce", with the 
implication that at some future date, when they got good and ready, the given 
feature would appear, to everyone's delight.  After APL became commercially 
available as a more or less IBM product this error was removed, because, as the 
suit and tie types in IBM said, it was tantamount to an announcement of an 
unavailable feature, which was contrary to United States government restrictions 
on IBM, and thus illegal.  

Iverson Software Inc. suffers from no such inhibitions in its J product.  

I once suggested the introduction of a converse to nonce error, the "yore error" 
to indicate a language facility that had been removed (such as monadic 
representation, which in early APL systems gave the binary equivalent to a 
numeric argument, e.g. representation 7 yielded 1 1 1; this facility has been 
reintroduced in J: try #:7), so that a user seeing such an error report would 
know that this facility existed only "in days of yore".  The suggestion was 
vetoed.    


-----------------------------------------------------------
This posting is forwarded from an internal Reuters mailbox.
No statement or opinion contained herein should be taken as
being Reuters policy, or even as being approved by Reuters,
in any way.
