Mail Archives: djgpp/2004/07/22/03:32:28
Eric Sosman wrote:
>>Sterten wrote:
>>
>> I tried R[99]={770} and then found that only the first value was 770,
>> so I concluded that the other values are undefined and only accidently
>> zero in most cases.
>
> Many years ago I worked on software to drive some
>automated electronic test equipment. One member of the
>team was a sure-fire electronics engineer, who could
>explain to us just why it was important to adjust the
>potential to 5 volts *before* applying the 1200 Hz sine
>wave, and that sort of thing.
>
> He was accustomed to analyzing his circuits before
>applying power to them, and he was much amused and bemused
>by us software types, perfectly willing to throw together
>any old bunch of gibberish and run it through the compiler.
>He characterized our methodology as "Random punches in
>random card columns," (I said it was many years ago),
>"then debug."
>
> Guenter Sterten, you are validating his observation.
>You have, it seems, no knowledge of the programming language
>you are attempting to use. You throw together some half-
>understood bits of C-ish text, stir them until the compiler
>stops complaining, and then try to draw inferences from
>whatever you can observe of the outcome. Your efforts do
>not seem guided by purpose, but by chance.
>
> Now, when confronted by the mysteries of the primate
>genome or the interstellar microwave background radiation,
>your method is indeed the best available. But given the
>existence of instruction manuals -- to wit, dozens of fine
>books explaining how to use the C language -- the choice to
>ignore them is either folly or arrogance. Your experiences
>in trying to get this simple program to run must surely have
>cured you of the latter;
>what will cure you of the former?
what was the former ? the radiation ?
Or manuals (the latter being books) ?
Well, an online text, easily searchable by keywords would help.
I guess this is the usual way how programmers (and other scientists)
proceed. trial and error.
You can't know everything these days and then you forget frequently.
Usually it's easier to try out a doubtful command by writing a
test-program rather than consulting the manual/book, which often
even has some errors.
> Learn C, Guenter Sterten. Get a book: There's no
>shortage of good ones, and even a bad one would put you
>ahead of where you stand today. Your problems stem from
>a complete lack of knowledge of the tool you are trying
>to use; take some time to learn about the tool before you
>chop your leg off in your ignorance.
the disadvantage is, that books are usually _huge_.
Even when small ones were sufficient, authors blow them up
because it's not profitable to see small books.
Basically, I only need a dozen or so C-commands.
And for each of these the corresponding assembly routine.
> Seriously.
>
> There's nothing wrong with being ignorant; it's where
>we all started. But there *is* something wrong with plowing
>ahead in a wilful state of continuing ignorance, when the
>means to cure it are all about you. Use them; you'll feel
>better for it.
you have to choose where being ignorant and on what to concentrate.
Only very few people will specialize in C.
Should we close the doors to all others who just want to learn
"a little C" ?
Why should I learn (and then forget) about those commands, which
I rarely use ?
Guenter Stertenbrink
- Raw text -