Mail Archives: djgpp/2001/06/28/13:45:11
On Tue, 26 Jun 2001 16:27:54 +0300 (IDT), Eli Zaretskii
<eliz AT is DOT elta DOT co DOT il> sat on a tribble, which squeaked:
>It's all in the Intel manuals; please read them.
Or what? You'll flame me? I sure *hope* that participating on this
newsgroup without being ostracized, flamed, laughed at, or otherwise
looked down upon doesn't demand one buy copies of big expensive
manuals whose contents you got the optimizing compiler to avoid having
to worry about. In fact, I'll seriously consider leaving if it becomes
apparent that participants are expected to have copies of any kind of
non-free reference material whatsoever. (And there's the additional
question of "expected by whom?" -- who exactly decides these things?
Since it's an unmoderated group, I'd say nobody does. Knowledge of
English and some mixture of knowledge of, and willingness to learn, C
would seem reasonable to expect here, but little else.) I'm also
concerned with the evidence that I have seen recently that asking a
certain class of "silly" question will provoke a lambasting. The
reason our society is so full of media junkies who believe what's
spoon-fed them and are so afraid to think for themselves that we are
heading down a spiral of government by micromanagement is because of a
similar attitude that people are exposed to in early schooling. People
perceive this, especially at that early age, as being *punished* for
not knowing something. Several immediate effects are: some people grow
up scared to ever ask questions and appear "ignorant" and consequently
they are not only incompetent at some things, but (and here's the true
wrong) unwilling and in some way unable to change that, so they have
to be led around by society and government or they become lost. (Seen
The Matrix? The bit about people so dependent on The System that
they'll fight to defend it?) So you have this whole class of people
who don't know a lot of stuff and won't ask. Meanwhile, they look to
the people in authority, who seem to know all kinds of things. But
these are either no more knowledgeable, just do a better job of hiding
their ignorance (or they spectacularly fail at the same, e.g. Dan
Quayle), or they got knowledgeable by asking questions. Of course, the
typical person doesn't realize this, and sees the world as divided
into two groups: everyday citizens and Experts, the latter somehow
being gifted with a better ability to learn, which let them get the
Almighty Ph.D. they can only dream of, or whatever. Yes. This is the
public view. The reality? Anyone can be that knowledgeable -- everyone
but the physically brain-damaged could get a Ph.D. given the time and
inclination, if it weren't for the effects of an early conditioning
against asking "silly" questions. Of course, when nobody will ask
questions but some basic knowledge is needed to make it in the world,
someone has to go out there and drill some basic stuff into everyone's
heads -- thus the modern public education system, which bombards the
students with a potpourri of useful information, trivia, and
propaganda, and then grills them for how well they've memorized it,
without even considering whether they *understand* the material, since
of course the vast majority of them don't. Then there are those few
people whose curiosity to understand is so strong they ignore the
strong social conditioning against appearing ignorant and ask
questions, or risk being seen actually reading a book in a situation
other than reading an assigned curriculum text book while physically
on the premises of their school and at gunpoint. They wind up being
scientists, mathematicians, hackers, and so forth. They wind up the
real experts out there eventually. Of course, this doesn't really
optimize things. With most people part of the proverbial ignorant
masses (which need not even exist -- it's not that some people are
truly unable to be more than one of the ignorant masses, it's just
that most people are conditioned against the behavior that would lead
them out of that trap), they can't distingusih between experts and
crackpots. Imagine you were a con artist in America. Dollar signs
appear before your eyes! Now imagine you were a con artist in a
country where everyone actually is encouraged to ask questions, not
rebuked when it's something they "should already know" (e.g. "why is
the compiler reserving 0x18 bytes?" or "Some modern architectures
concern about alignments as large as 256 bits?!"), and thus able to
think for themselves. Uh-oh! Time to find honest work or flee the
country!
Okay, the above is slightly hyperbolic, but when I see a forum that
ought to be largely, if not entirely, composed of hackers start making
people feel stupid (whether that is the intent or not) for asking
"green" questions and making the odd typo or mixing up their number
bases from time to time (and programmers do, after all, regularly have
to juggle three or even four of them -- can you honestly tell me
you've never mixed up your number bases, not even *once*? I suspect
the honest answer has to be either "I don't remember" or "I've mixed
them up at least once"...) -- well, it bothers me.
The only stupid question is the one you didn't ask. (Or research by
other means. Or, in a newsgroup, the one that you did ask that's in
the newsgroup's FAQ... :-))
>Due to various data
>prefetch penalties, they recommend that the stack be aligned on
>16-byte boundary. I didn't analyze the code produced by GCC too
>deeply, but it might be considering the possibility that the stack
>frame is only aligned initially to 4 bytes, since greater alignment is
>normally not guaranteed, especially if you mix code produced by
>different compilers.
--
Bill Gates: "No computer will ever need more than 640K of RAM." -- 1980
"There's nobody getting rich writing software that I know of." -- 1980
"This antitrust thing will blow over." -- 1998
Combine neo, an underscore, and one thousand sixty-one to make my hotmail addy.
- Raw text -