Subject: Re: DMA beyond end of isa
To: David Mazieres <dm@amsterdam.lcs.mit.edu>
From: Jordan K. Hubbard <jkh@time.cdrom.com>
List: port-i386
Date: 12/26/1995 07:55:38
> I, too, have 32Mb and an aha1542.  I hacked machdep.c to limit my
> memory to 16 Mb, but am basically swapping constantly.  Can you say
> something about how different OpenBSD is from NetBSD and how hard it
> was to switch?

No insult to the OpenBSD folks intended, but my impression of that
project is that it's supposed to be some sort of "reference library"
of code for general use, not a place to send "end users."  In point of
fact, I'm not even sure how useful as a "reference library" it's going
to be, given that none of us seem to have much interest in looking at
it.. :-)

My advice would be to stick with NetBSD until they get the
bounce-buffering problem worked out or go to FreeBSD.  Both systems
have the *infrastructure* required for answering user questions,
responsing to features and bug reports, several books on the way that
will be hitting the bookstores this spring, etc and so forth.

All of those things aren't particularly important to software
historians and archivists, I'll agree, but of major importance to
anyone actually trying to run a system for day-to-day tasks.  I think
the end-user base is bewildered enough already and needs another "user
oriented BSD" like Linux needs another 12 distributions.

In short, if OpenBSD would like to continue pitching themselves as a
programmer's resources, then I think they're making the best of an
unfortunate NetBSD scism and I'm inclined to say "pax - let's make the
best of it."  If they're trying to fragment the end-user base of BSD
even further then we're all being done a major disservice given that
resources are already stretched wafer thin as it is.  We're supposed
to be "competing" with the 900 pound gorilla that Linux has become,
not one another!

The users can help too, of course, given that they always get the
final "vote" in such things anyway.  Don't bail out on the two OSes
who have done so much good BSD work over the last 2 1/2 years and are
now starting to get really good - help us make them better!  There has
already been a rather tragic replication of effort, and I see no need
to make it worse!

NetBSD resources:
	http://www.netbsd.org

FreeBSD resources:
	http://www.freebsd.org

					Jordan

P.S.  Yes, I have talked about merging the two groups again to even
better serve the "BSD cause" but I guess that too much water has now
passed under the bridge for that to be practical, given the
exceedingly luke-warm response I've received.  Ah well!