In
spinoza1111wrote:
It's astounding, isn't it? You obviously agree, since you cite the
same quotation twice. I don't remember the quote and I don't know the
context, but (remembering that I don't trust you to tell the truth) I
will take the above at face value for now.
Let's look at the implication. Here's Seebs, who has never taken a CS
class in his life, running technical rings round you despite all your
alleged college education. He knows more about the Standard than you,
The Standard is the problem and the less said about it the better.
Knowing it is not good. You admit it failed.
he knows more about C than you, and in fact he knows more about
programming in general than you. You have yet to score one technical
point over him, and he has scored many over you. In fact, in that
regard he reminds me of me.
You say this for the same reason corporate types say it. "Knowing" in
most financial software corporations means "talking with confidence,
and proving your chops by bullying safe targets".
In fact, Seebach gives no evidence for knowing programming. Had he
known programming, he would not have participated in a "standard" that
declares that things cannot be known or predicted. He would have known
that we need programming languages that allow us to sensibly predict
what code will do.
Seebs's technical knowledge is not only formidable but also very easy
to identify, and anyone who seeks to mock it is on a hiding to
nothing.
This isn't at all evident. In my experience in taking computer science
as an undergraduate and graduate student, computer science professors
in computer architecture, and textbooks, will confirm that stacks and
twos-complement are in ordinary use. Schildt confirms both facts, but
Seebach denies them because he learned computer science apparently
from hacking and reading the Standard.
Seebach doesn't give any evidence for knowng WHY twos-complement is
the default choice, that being the fact that we don't want to have two
different zeroes.
It's as if I'd reasoned in 1970 from the local properties of the IBM
1401 (variable-length operands, no general registers) to the
properties of all computers. But reading Sherman's "Programming and
Coding for Digital Computers" I realized that most large computers
used in fact a radically different architecture with fixed word
lengths and registers. But reading Turing, I realized that either
computer could simulate the other given enough time and memory space.
Whereas Seebach has read a grand total of one book, and resembles the
lawyer in Dickens' Bleak House, who prides himself on reading only
case materials in Jarndyce v Jarndyce.
Whereas Seebach believes it to be intelligent to say what C "is not"
because that way you have through the logic of negation a higher
chance of being right, which is what you need when you're basically
bullshitting yourself and others...by randomly attacking Schildt
because you can't get laid (or something).
A Bush supporter, Seebach uses Bush's logic, of making a statement not
easily proved nor easily disproved. Bush said that Saddam Husayn might
have WMD and thousands of servicemen and women died, and, we found
that Saddam Husayn had no WMD. Seebach implies that Schildt is
unqualified because the C standard allows for exceptions to the use of
twos complement and the stack; but Seebach, like Colin Powell in the
UN in 2003, Seebach is unable to coherently identify or name the
exceptions, only to insist on their logical possibility. And when
challenged to do so, Seebach, like Bush, ruins careers and names, even
as Bush saw to it that Valerie Plame, a CIA agent, was outed in
violation of the law when her husband, a former US ambassador,
published facts disproving the "Nigerien yellowcake" justification for
belief in Saddam Husayn's nonexistent nuclear program.
Let's look at this example of Seebach's incompetence and dishonesty.
Herb: free() must only be called with a pointer that was previously
allocated with one of the dynamic allocation system's functions
(either malloc(), realloc(), or calloc()).
Petey: Also specifically untrue. ANSI states that free(NULL) is valid
and has no effect. (Also note that it must be called with a pointer to
space previously allocated, not with a pointer previously allocated,
and that the pointer must not have been already freed or passed to
realloc().)
Herb is imparting a valuable lesson, fundamental to programming: when
you open something, close it, for every right parenthesis there is a
left...like debits and credits in accounting. As a practical matter,
Schildt is saying that if you have a memory bug, check the balancing
of malloc/realloc/calloc versus free. Competent C programmers do this
all the time.
Little Petey's counterexample? Drum roll: free(NULL). As if little
Petey never heard of the special role of the number zero in ordinary
arithmetic, or of the zero element in an abstract group.
Peteywunkle, we KNOW that NULL is different. In fact, do you know that
if it is returned by malloc this is a failure indication that needs to
be tested?
Your words after that are also things we know, written very poorly.
Seebach is worse than my student Otto who with his personal problems
constantly interrupted my C class in Chicago providing insights based
on 20 years of assembler. Seebach is more like the Nazi bullies who
would interrupt classes in the Weimar Republic.
