Re: A modest proposal -- We need a patch penguin

[Jeff Garzik]

On Thu, Jan 31, 2002 at 11:03:07PM +1100, Keith Owens wrote:
> On Thu, 31 Jan 2002 03:58:10 -0500, 
> Jeff Garzik <[e-mail]> wrote:
> >On Thu, Jan 31, 2002 at 06:52:55PM +1100, Keith Owens wrote:
> >You are missing a huge point.
> >	You:	"Look Ma, nothing breaks!"
> >	Ma sez: "It's supposed to, silly"
> 
> Hypocrisy, Jeff.  In your previous mail you complained that kbuild 2.5
> was not ready to go in.  When I point out that not only is it ready but
> it can go in without breaking the existing code, then you complain that
> I am not breaking anything.  Make up your mind.

You still do not appear to understand.  Please re-read my last message.

If kbuild was ready to go in, the old build system should go away.

ready != not breaking things, if you are breaking things on purpose.

> >Cleanup does not occur if cruft lives on as "backwards compatibility."
> >You simply promote further bitrot and discontinuity.
> 
> The old code does not live on indefinitely, it gets removed as soon as
> kbuild 2.5 is deemed stable.  Four weeks, tops.

You don't prove something stable with most people not using it.

> >Let's see if I have this right:
> >* You want completely duplicate build and config systems in the kernel,
> >  with all the accompanying headaches for maintainers.
> 
> Only for as long as it takes to prove that kbuild 2.5 is ready.

Not the Linux kernel way.

	Jeff