Path: utzoo!mnetor!uunet!husc6!cmcl2!brl-adm!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: comp.lang.c
Subject: Re: Another way (How not to write a loop)
Message-ID: <7296@brl-smoke.ARPA>
Date: 19 Feb 88 21:38:44 GMT
References: <1988Feb11.200149.25172@sq.uucp> <2550046@hpisod2.HP.COM> <23450@yale-celray.yale.UUCP>
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 17

In article <23450@yale-celray.yale.UUCP>, leichter@yale.UUCP (Jerry Leichter) writes:
> To add a little reality:  SPITBOL/360 was done somewhere around 1971.

Yes, and this discussion about how to iterate via floating-point exception
was being conducted in 1988.  My concern was that people not use a "trick"
unnecessarily, when the trick might not work and when the alternative is
better anyway, in the C language, which is what this newsgroup is about.

I mean, I've relied on the CDC 1700's ability to automatically indirect
arbitrarily many times, and on -0 as a special flag, and other tricks,
when programming specifically for that system (and when portable system
programming was practically impossible anyway).  But the vast majority
of programmers are not in such circumstances today.  There has been
considerable progress in understanding the real economics of software
since the 1960s and early 1970s; we have learned what is wrong with
cute system-dependent "tricks" and assembly-language programming for
general applications.