Re: To Richard Heathfield: enough's enough

From: Edward G. Nilges (spinoza1111_at_yahoo.com)
Date: 12/30/03 

Date: 30 Dec 2003 08:14:50 -0800

Randy Howard <randy.howard@FOOmegapathdslBAR.net> wrote in message news:<MPG.1a5a691d38c479e989a54@news.megapathdsl.net>...
> In article <f5dda427.0312291335.1b33bc2@posting.google.com>, spinoza1111
> @yahoo.com says...
> > My point was that at the scale of user entry of propositional logic
> > statements on a console, the effect of using strlen in the for loop
> > was close to the Y axis.
>
> If you had intended arbitrary length restrictions, you should have
> enforced them in your code. If you knew of the potential problem
> with larger strings but couldn't be bothered to hoist the strlen
> out of the loop, then you should have at least put in comments to
> the effect that it would be a good idea for large strings, and in
> point of fact even good on small strings, as you have no idea
> when that code might make it to an incredibly slow embedded processor
> and re-used with terrible results even using relatively small strings
> (or not as there is no warning, apart from the implied one when read
> critically by actual programmers that know enough to look for such
> things.

What nonsense. And what's interesting is that these rules aren't
followed in the enterprise software systems written by gnomes who
nonetheless have made them in a Scholastic fashion a dead set of
strictures which allow them (in the context of corporate downsizing)
to continue to fantasize that programming is a profession insofar as
programmers refuse solidarity with each other or with working people
in general.

>
> > Yes, if you use strlen in a for loop, the poor design of the for loop
> > causes trouble when you scale up.
>
> No, your lack of knowledge of the semantics of the C for loop caused
> the problem. If you want to run PL/I, IBM still has a Windows version
> available, I believe it sells for <$130 US. But guess what, I suspect
> if you had posted it in PL/I, and called LENGTH every time through
> your loop, it STILL would have been wrong. Or, are you claiming that
> references a variable initialized outside of the loop is <= the execution
> cost of calling the LENGTH function every time through the loop?
>
> *DING* Is the light bulb coming on yet? I guess that language must be
> flawed too. It couldn't possibly be that you're just clueless, right?
>
I think I've made it clear that it doesn't bother me if a computer
recalculates, not only in a reliable but also in a transparent way, a
needed value. You see, that's what computers are for.

I think I've made it clear that I long ago disabused myself of the
(absurd) notion that a computer is like a car, down which you travel
"fast" some highway to Wrong Answer, AZ.
 
> Psst! It's okay to admit it, we already know and we'll understand,
> promise.
>
Who is this we? You are just a loser or a thug with a wireless card,
at best.
 
> Your inability to admit that the problem was your lack of knowledge of
> C was the actual cause of this and just move on is causing these
> long threads, for which you try to blame Richard instead. If you'd just
> say "you know what, I made a mistake and I realize there is a better way"
> it would be over and done with, but you're not grown up enough to do so.
>
Yeah, I was saying there is a better way. It is of course constituted
in OOD and typical of OOD, with its close coupling of data and code,
to have shorter strings in the state, where microefficiency
considerations mean less.
 
> > But this is just one more problem in the C language which passive-
> > aggressively walked away from the knowledge that was accumulating in
> > the development of Algol and PL/I to say "leave us alone with our
> > existing knowledge base."
>
> I guess a lot of languages that came later (in fact almost all of them)
> did the same thing. You've already admitted that you've manually
> added support for PL/I constructs from the 60s to VB, so it must be
> an evil language too, by your warped thinking. I tend to agree in
> the case of VB, but for entirely different reasons.
>
I haven't talked about good or evil languages. I have talked about
thugs and losers with modems, and this is what bothers people in this
thread. But if the shoe fits...
 
> > It created a tribalized Bantustan in which male knowledge could not be
> > challenged because the limits of knowledge were defined by C itself,
> > and in which today's C programmers have only to be "right" about C.
>
> As they say down South, "It's your lie, tell it like you want to".

As they say down South, let's Lynch somebody instead of examining why
we're such a bunch of born losers.