by redmail.netbsd.org with SMTP; 20 Feb 1999 06:41:06 -0000
	by c44023-8.lvrmr1.sfba.home.com (8.9.1/8.9.1) with ESMTP id WAA01800;
	Fri, 19 Feb 1999 22:40:33 -0800 (PST)
	(envelope-from clovis@home.com)
Message-ID: <36CE58E0.372BA6D0@home.com>
Date: Fri, 19 Feb 1999 22:40:32 -0800
From: Frank Warren <clovis@home.com>
MIME-Version: 1.0
To: davidw@debian.org
CC: netbsd-advocacy@NetBSD.ORG
Subject: Re: NetBSD/Linux 'distribution'
References: <19990219211709.A25512@debian.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

David Welton wrote:
> 
> Hello,
> 
> Some people have brought up the idea on the Debian (Debian is a Linux
> distribution, see www.debian.org for details) lists of extending our
> 'distribution' to different kernels.  For instance, there is a group
> working to make a Debian GNU/HURD distribution.

Debian has, for the benefit of the NetBSD group, an excellent
reputation, much
exceeding that of RedHat and Caldera, for coherent userland
distributions.  And,
as one who has been trying to get something useful out of NetBSD beyond
writing
yet more C code, NetBSD could use more "stuff."  Users don't use
computers just
to write more C code, Richard Stallman's opinions and license to the
contrary.
Mostly, users want to use it, and using it means that one has to have
"stuff."

It's not hardware support, but "stuff," that put Windows where it is
today.  With
lots of "stuff" and a competitive hardware market, Microsoft rolled into
dominance
easily.
 
> Someone brought up the idea of using most of the current Debian
> userland tools on top of a BSD kernel, which of course sparked
> discussions about which one might be most appropriate (at this point,
> I should note that no one has actually *done* anything about this, and
> I sort of doubt it will ever be more than speculation).

I would welcome as much stuff as Debian has to offer personally.  Mind
you, though,
that I am not a member of the core team and have no standing.  Indeed, I
am a 
disgrace in that I am writing this with FreeBSD instead of NetBSD.  I
have no luck
getting "stuff" to run on NetBSD, and am still not sure why I can't get
emacs
to work on NetBSD 1.3.2 (which severely limits its usefulness for coding
C
programs which, lacking "stuff," is the only use to which I can put it).
 
> What I was curious about was what sort of reaction this might engender
> in the NetBSD community, should something similiar come to pass.

The remaining folks must speak for themselves but, bluntly, if NetBSD is
to
grow and prosper, it needs stuff in a bad way, as does FreeBSD.  I would
remind
the NetBSD folks who look unkindly on all of this that an OS without
applications
is, for 99 and 44/100ths percent of the world, about in the same class
as a
computer without an OS.  It's not useful to them, and rather pointless.

If NetBSD is not to fall down the same rathole that ate and ultimately
destroyed
System V, there needs to be users, and for that, there needs to be
"stuff."  Like
FreeBSD, OpenBSD and all Linux variants, NetBSD has a powerful enemy
named Bill
Gates.  The very existence of ANY OS besides his, or users on ANY system
besides
his, is something he is out to stop.  Without "stuff" these alternatives
will
ultimately stop themselves.

The cool reception of kde is prototypical of what killed System V and
SCO, and
will ultimately kill all the BSD and Linux variants if it gets the upper
hand.

If we want to be popular, we will have to learn that most users aren't
UNIX hacks
and don't want to be.  They don't like thin tools, thin clients, and
want some
hand holding.  And they absolutely despise command lines.  Bill Gates
comprehension
of this in 1984 when he first saw the MacIntosh is what led to his
dominance.  UNIX's
lack of comprehension of this is what nearly drove it to extinction. 
Except for
the Internet, BSD and Linux would both be quite dead by now.

This may be our last clear chance to get things right in terms of
winning a share
of the hearts and minds of enough users to be visible, much less
contenders at
being anything like a mainstream set of tools.

Frank Warren
clovis@home.com