Subject: Re: Drive Numbering...
To: Chris G Demetriou <Chris_G_Demetriou@balvenie.pdl.cs.cmu.edu>
From: Michael L. VanLoon -- HeadCandy.com <michaelv@HeadCandy.com>
List: current-users
Date: 11/16/1995 21:56:32
>> But I think that this is a bad way to look at the world.
>> Honestly, I'm not ready to tell the novice to install NetBSD yet
>> ... not unless I have a few hours to help. I like to config my own
>> kernel, but if we're EVER thinking of making this an easy OS to
>> install, we're going to have to address the everyday use of a kernel
>> that someone just installed from a binary snapshot.
>My point is: in what way does the generic kernel _fail_ for an
>"everday user"? if their disks change numbers, for instance, what
>goes "fatally wrong"?
I have to agree with Chris, here. I fail to see how the generic setup
is bad. In fact, if it were the other way around, I would be uneasy.
Once you start assuming things about someone's setup, is when you
start breaking things for those setups you didn't anticipate.
Especially when you get into machines with multiple disk controllers.
How are we supposed to decide which one is first, and which is second?
There is a simple and elegant method for nailing down disk devices
which works quite well. I don't understand the problem.
And, the argument that novice users are going to be confused is
ironic, at best. Especially since the World's Most Popular OS
(MS-DOS) rearranges partitions arbitrarily when disks are added to the
system, without any hope of nailing them down in a predictable
pattern, unless you're running Windows NT. So, what the world at
large is used to expecting is just what they're being given:
dynamically shifting device identifiers. The only difference is that
ours have a logical pattern. :-)
>perhaps it'd make sense to mandate kernel source installs, and
>automatically build a custom kernel for people...
Ugh...
-----------------------------------------------------------------------------
Michael L. VanLoon michaelv@HeadCandy.com
--< Free your mind and your machine -- NetBSD free un*x >--
NetBSD working ports: 386+PC, Mac 68k, Amiga, HP300, Sun3, Sun4,
DEC PMAX (MIPS), DEC Alpha, PC532
NetBSD ports in progress: VAX, Atari 68k, others...
-----------------------------------------------------------------------------