Re: Herbert Schildt, author of The Complete C++ Reference (NOT C Unleashed) rehabilitated on wikipedia

spinoza1111 said:

On Jan 20, 6:17 am, Richard Heathfield <r...@xxxxxxxxxxxxxxx> wrote:
spinoza1111 said:

<snip>



Since "Herb's guidelines" recommend a particular interpretation for
i = i++, we know they're useless. And you still don't understand what
"undefined" means.

It means that you didn't do your job.

No, it means that C doesn't try to dictate every possible outcome of
every possible construct. To do so would have made C much less useful.
For example, if C ruled that any pointer not pointing to an object (or
function) was required to have the value NULL, then the implementor
could probably find a way to do this, at the expense of slowing the
machine down

Oh, we mustn't slow the machine down to a crawl...let's get those
wrong answers out the door boys.

Well, if I want wrong answers, I know who to ask. But I have a better idea.
Let's get the *right* answers out the door, in a decent timeframe.
Sandboxes have their place, but no value is added to a correct program by
constant nanny-state bounds-checking. If the code is right, the extra
checks won't help, and they will very likely get in the way. And if the
code is wrong, the proper course is to fix it.

This is why I've actually seen phrases such as "today is more
important than correct" in management hymnals.

Correctness is important. If we don't have to get the right answer, we can
have the program run in zero time and take zero memory. But speed can be
important too. If a program's task is to produce an important daily report
that is used for making important and timely decisions, but takes 25 hours
to run, what's the point in having the program at all?

Like it or not, runtime controls are the only sensible regime.

During debugging, yes. Production code should not need such props.

What about his error that bytes are 8 bits? Or that main() is
declared void? Or .... And you admitted an actual one in another
posting.

Bytes are 8 bits for working programmers.

So what about people who write DSP code? Are you trying to claim that
they are not programmers? Or that they are not working?

As far as I can tell, there's just as much stupidity in that type of
code and it seems to me, as a lay observer, that its biggest problem
is that in a natural field for applying OO, retardo-geeks insist on
their rights not to learn anything new, and hold companies to ransom
by insisting on using dope C. The game is ending as the work migrates
East.

You failed to answer the question. You do that a lot. Presumably you hope
nobody notices. But people do notice.

--
Richard Heathfield <http://www.cpax.org.uk>
Email: -http://www. +rjh@
Google users: <http://www.cpax.org.uk/prg/writings/googly.php>
"Usenet is a strange place" - dmr 29 July 1999
.