Re: devfs again, (was RE: USB device allocation)

Nathan Hand (nathanh@chirp.com.au)
Fri, 8 Oct 1999 09:57:07 +1000


On Thu, Oct 07, 1999 at 04:48:15PM -0400, Charles K Hardin wrote:
>
> Ah, the devfs Holy War - before people get to heavily involved in what
> UNIX should be, because DAMN IT, I KNOW WHAT UNIX SHOULD BE AND I AM
> ALWAYS RIGHT rhetoric ... you might consider that other OSes, like IRIX,
> have done similar things with their hardware fs (not the same setup as
> devfs, but similar in concept to create a tree that just has the devices
> that are currently on the system)

Sure, but just because some other UNIX did it, doesn't mean it was a good
idea. The argument should be argued on its own merits.

> this argument is pathetic... a philosophical holy war from a bunch of
> voices with strict opnions defining their arguments and not presenting
> a clear logical argument for why this is such a horrid design to
> introduce in the kernel...

The devil's advocate's position is that the people who want devfs haven't
done a good job of explaining why devfs *isn't* a horrid design. Though I
honestly don't know how it could be argued any better.

One interesting thing is to stand back and try and work out what model of
management we're seeing here. If the people in power listened to the poor
huddled masses and gave us what we asked for, we'd have a republic. If we
instead had a system where the people in power said "we know what is best
so it'll take a revolution to change our minds", we'd have a theocracy.

The real answer obviously lies somewhere in between, and varies over each
argument, but I'm definitely seeing some "High Priest" behaviour whenever
the argument over devfs flares up.

-- 
Nathan Hand - Chirp Web Design - http://www.chirp.com.au/ - $e^{i\pi}+1 = 0$
Phone: +61 2 6230 1871   Fax: +61 2 6230 1515   E-mail: nathanh@chirp.com.au

- 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/