From: Jim Freeze <jim@...>
Date: 2005-04-28T08:58:15+09:00
Subject: Re: Comments Are More Important Than Code

* Jim Freeze <jim@freeze.org> [2005-04-28 08:42:16 +0900]:

> * Dave Fayram <dfayram@gmail.com> [2005-04-28 08:19:29 +0900]:
> 
> > Neither piece of code is lying. Code cannot lie. The unit test says,
------------------------------------^^^^^^^^^^^^^^^

If you will permit me to philosophize for a minute here.
<thinking out loud>
I think this kind of dogmatism is not good. I understand
where it comes from, but I'm not sure what good it brings
to the table. 
The first question is what is code and why can it not lie?
Code is nothing more than the representation of logic.
Surely it doesn't speak to us when it is sitting on
a diskdrive, a piece of paper or memory. But, it must
be executed through some type of CPU. If the CPU or memory
experiences gamma rays and memory glitches, then I suppose
one could say that the code lied. If the same code
lands the space shuttle one time and crashes it another
because it experienced a hardware mishap, then our 
interaction with the affects generated by the code is 
surely different. This could lead one to conclude that 
one of the outcomes was because the code 'lied'.

That is a simple story to show what that kind of dogmatism
sparks off in my head. :)

Now, for another, even better, story on the subject of lying.
My cube neighbor at work brought this to my attention.
He heard something that was the truth and a lie at the
same time. Wow, how could that be?

He looked up the definition of lie, and one definition was
to mislead or mis-represent. Then he told me of a comment
from a tobacco spokes person who said, "No death has
ever been attributed to smoking." (or something like that.)
He was correct, technically. But at the same time he
was misleading people about the harmful effects of smoking.

Now, what does this have to do with computers. Well, if some
code is being subtle and misleading me, while faithfully and
accurately doing what it is supposed to do (no glitches),
then, I suppose one could say that the code was lying.

Enough of the dogmatism, please? :)

But, then again, I could be completely wrong and you will
show me an important benefit from this mode of thinking.
</thinking out loud>

Jim
(with no offense toward kind Dave Fayrem.)

There are no absolutes in science, absolutely.

-- 
Jim Freeze
Code Red. Code Ruby