Re: The real challenge ;)

"H. S. Teoh" <[email protected]> Wed, 22 May 2002 19:23:39 -0400
Newsgroups gmane.comp.gnu.cons.general
Message-ID <20020522232339.GB14848@crystal>
On Thu, May 23, 2002 at 01:41:18AM +0300, Asko Kauppi wrote:
> 
> Following this cons-or-scons debate, I think the real "challenge" for both
> of the tools is to take over the position of 'make' as the official
> Linux-kernel-distribution-dependency-tool. There's a lot of subdirectories
> and interdependent makefiles in the Linux tree. Seems like a perfect target
> for you guys!

Indeed. If cons/scons can handle the Linux kernel, even if it doesn't get
officially adopted, this would become a major selling point.

> Has anyone been in contact with Linus about this and what's his intensions?
> Gettin (s)cons into that business would give a lot of more publicity and
> general approvement to whatever tool manages to take on that job. Good luck!
[snip]

Good luck indeed. Seeing the current "we're engineers, not scientists"
attitude(*) of the Linux kernel developers, good luck convincing them to
switch the build system.


(*)I.e., "we have a working system in C, why re-write everything from
scratch in C++?" Or, simply, "if it ain't broke don't fix it". Not that I
disagree with this attitude, mind you. Nothing like Linux falling flat on
its face because 2.6 is rewritten in C++ and just does not work because of
countless bugs introduced by the transition. The last thing kernel
developers would want to worry about is a broken build system, when they
could be better spending their time doing real work with the kernel code
itself. But this is just IMHO, perhaps they *will* be receptive to the
idea.


T

-- 
Real Programmers use "cat > a.out".

_______________________________________________
[email protected]
http://mail.gnu.org/mailman/listinfo/cons-discuss
Cons URL: http://www.dsmit.com/cons/