Re: [dwm] Documentation!

From: David Tweed <>
Date: Sun, 18 May 2008 22:09:32 +0100

On Sun, May 18, 2008 at 6:38 PM, Steffen Liebergeld <> wrote:

> I completely disagree with you. First of all, your pen and paper equations
> are used only one time and that is when you do the equation. Afterwards you
> throw the paper away and go with the results. When programming, you have to
> make sure that the code is read- and understandable when you get back to
> it. What makes good code is how fast you can grasp its workings when first
> reading it (when you read your code after a year of not touching it, it is
> like the first time you encounter it). And it is not a good thing to have
> short variables and document them somewhere else. Have you ever heard of

I tend to keep pen and paper equation solutions, because I generally
make some subtle mistake first time.

Try interpreting some relatively complex arithmetic expressions with
the intel simd intrinsics (with long names like _mm_sub_epi32,
_mm_madd_epi16, etc) and see if you can follow what's happening. In
personal code I've got macros that give these shorter names that are
less immediately obvious but make it easier to see the overall
expression (as in "You can't see the forest for the trees".) I tend to
use longer and more descriptive variable names than others
(particuarly at the interface level) but go with short names when
they're being used in program code intensively or in complex ways.

As in all things, applying considered judgement is better than
decreeing an inviolable rule.

cheers, dave tweed__________________________
Rm 124, School of Systems Engineering, University of Reading.
"while having code so boring anyone can maintain it, use Python." --
attempted insult seen on slashdot
Received on Sun May 18 2008 - 23:09:34 UTC

This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 15:40:42 UTC