Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uunet!mcsun!ukc!axion!tigger!raph
From: raph@tigger.planet.bt.co.uk (Raphael Mankin)
Newsgroups: comp.lang.c
Subject: Re^2: Oh noooooo!!
Message-ID: <556@tigger.planet.bt.co.uk>
Date: 12 Sep 89 12:55:18 GMT
References: <7598@goofy.megatest.UUCP> <34566@apple.Apple.COM>
Organization: RT5111, BTRL, Martlesham Heath, England
Lines: 25


Before most readers of this new group were born, Dijkstra wrote an
article called Uncontrolled use of Goto Considered Harmful. It was
		^^^^^^
not called 'Goto considered Harmful'.

It is possible to program without goto provided taht you have a
do-forever and an if available, but it remains true that the goto is
sometimes the best way of doing things.
              ^^^^

Personally, I restrict gotos to forward trasnfers - this minimises the
risk of infinite loops; and to the multi-level break/continue or error
escape cases. After some 20 yewars of programming I have not found
this to cause any problems and the code remains readable.

The issue of use/non-use of goto is not one of theology. What matters
is
	does it work
	is it safe
	is it readable.
Nothing else matters.
-- 
Raphael Mankin
raph@planet.bt.co.uk