Re: GNU/Linux stance by Richard Stallman

Oscar Levi (elf@buici.com)
Sun, 4 Apr 1999 15:57:42 -0700


On Mon, Mar 29, 1999 at 07:37:00AM +0000, Linus Torvalds wrote:
> In article <7dls0b$uq2$1@alfie.demon.co.uk>,
> Nick Holloway <Nick.Holloway@alfie.demon.co.uk> wrote:
> >
> >ObKernel: the kernel does not fall back to using /bin/sh to execute files
>
> Indeed.
>
> A zero-length /bin/true does not work reliably, and never has. That's
> why it's not zero-length any more (not on any modern system I know of,
> anyway). You need to have the magic "#!/bin/sh" part to make it be
> reliably recognized as a shell script.
>
> However, making it do anything but a simple "exit 0" is horrible.

Duh. There's no need for a help string/copyright notice in a brief
shell script.

> Anybody who really thinks /bin/true should report a version number and a
> help string (or even a copyright notice) needs to get his head examined.

Just did. A ok.

%^)

Who's to say what's in a binary executable? The overhead of
implementing a usage message is negligible. In fact, the only excuse
for *not* putting them into small binaries is...code size?
nope...complexity? nope...efficiency? nope...laziness? Bingo.

Think again Mr T.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/