Subject: Re: history, design, or both?
To: Dan Debertin <airboss@nodewarrior.org>
From: Michael Richardson <mcr@sandelman.ottawa.on.ca>
List: tech-net
Date: 11/03/2001 11:07:18
>>>>> "Dan" == Dan Debertin <airboss@nodewarrior.org> writes:
    Dan> Allow me to say from the get-go that I realize that this question is
    Dan> flame-bait. It is not intended to be such -- the question came up in a
    Dan> technical discussion, and none concerned had a technical answer. That
    Dan> said, what I want is a technical answer, not something that belongs on
    Dan> -advocacy. Okay? Good.

    Dan> To wit, then, what I want to know is why *BSD is using the underlying
    Dan> hardware as the basis for the names of network interfaces, i.e. xl0,
    Dan> ep0, rtk0, etc. I know there are historical reasons, and that almost
    Dan> every other UNIX out there (except one :) does what *BSD does, but it
    Dan> _seems_ better from a usage & automation (scripting) standpoint to
    Dan> have all ethernet interfaces called simply "ethX". Why should userland
    Dan> have to care about what the hardware is?

  There are some technical advantages to it.
  I see no reason to call things "ethX". There are so many different kinds of 
"ethernet"-like things that that other OS does not call "ethX", and there are
is the continuous problem of "ethX" devices getting renumbered.  
  If we were going to change, then we should change to:
"net/MAC" or something like that which was slot *and* vendor independant.

]       ON HUMILITY: to err is human. To moo, bovine.           |  firewalls  [
]   Michael Richardson, Sandelman Software Works, Ottawa, ON    |net architect[
] mcr@sandelman.ottawa.on.ca http://www.sandelman.ottawa.on.ca/ |device driver[
] panic("Just another NetBSD/notebook using, kernel hacking, security guy");  [