Re: No Go To's Forever!

[Brian Rogoff <bpr@shell5.ba.best.com>]

On Wed, 22 Mar 2000, Robert Dewar wrote:
> In article <38D8DFC2.778826FB@cadence.com>,
>   Paul Graham <pgraham@cadence.com> wrote:
> 
> > Why is there no 'continue' statement in Ada?  You've shown
> > that the function of 'continue' is desirable.
> 
> Yes, the *function* of continue is desirable, but on the other
> hand a continue is indeed a goto, so why not say so? That's the
> other point of view, and as you see from the previous post,
> some people find it clearer to use a goto here. We did discuss
> this in the language revision but there was insufficient
> support for this language addition.

I agree with Roger Barnett who wrote that he preferred your "goto" form 
to C's continue since it made explicit the flow of control. Adding
continue to Ada would be a mistake IMO (like adding "use type" was. There, 
get those flames going ;-)

> I guess the burden here is to explain why the goto formulation
> is so horrible. 

Because the great Dijkstra said goto is bad, and therefore it is, at least
if it is spelled that way. 

Dijkstra also wrote that CS students should be kept away from a computer
for their first year and focus on pencil and paper proofs (no problem for
me, I was a Math major) and he probably didn't like Ada anyways so phooey
on him :-)

-- Brian