Okay. Sure. You are entitled to an opinion, and your experience differs.
My goal is to make the kernel source easier to maintain and add features
to. Linus controls the marbles though, so his word is final [no, I don't
want to start my own kernel branch]. So I sit quiet, and only stand up
to back-up another [as I did in the message you quoted], in hopes that
not all ideas will vanish into the void.
Linus included his coding style guidelines in the kernel source when
I first brought this up. Whether he did it solely because of me, I don't
know. I'm just glad he did something. [Linus is not unreasonable.]
> >Linus refuses to impose strict style and documenting practices.
> >I understand his views on these, but I don't agree with them.
>
> But Linus has managed development of a working kernel and you have not.
Cheap shot. That isn't nice, or necessary.
> Better (or even some) documentation of kernel design decisions would
> be wonderful. They are obviously not essential for the development.
I never stated they were essential, and yes, it is obvious.
So your point for stating the obvious is ... ?
> You could, demonstrate the correctness of your views by generating some
> good documentation and showing the good effects of this documentation
> on the system. I look forward to seeing your results.
You will be seeing some soon in the loop driver. I've already been
complemented on the style I used, though it is not mine - it's a hybrid.
I don't in any way mean it is perfect, as to me it is just tolerable.
If I recall, Alan Cox described my _personal_ style as being of too low
a density, so I accepted nearly all of his suggestions [forgive me Alan
if my memory is faulty and it was someone else - too lazy to dig in my
mail archives]. We agreed on something we both could live with. A few
others contributed to the style too, but in smaller ways.
If you'd like to see a pre-release, send me private E-Mail (lemme know
if you are in North America too). I always love to get suggestions, even
on coding style.
May the source be with you...
-- Andrew E. Mileski mailto:aem@netcom.ca