*BSD News Article 40832


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!hpg30a.csc.cuhk.hk!agate!howland.reston.ans.net!swrinde!sgiblab!idiom.com!idiom.com!not-for-mail
From: muir@idiom.com (David Muir Sharnoff)
Newsgroups: comp.os.386bsd.misc
Subject: Re: To Merge or Not to Merge *BSD. What does it really mean?
Date: 14 Jan 1995 14:06:21 -0800
Organization: Idiom Consulting
Lines: 67
Message-ID: <3f9hst$q8n@idiom.com>
References: <3enogm$5l7@fw.novatel.ca> <3f15dh$bp@knobel.GUN.de> <3f1h83$hgl@newshost.lanl.gov> <3f253c$es0@agate.berkeley.edu>

In article <3f253c$es0@agate.berkeley.edu>,
Jordan K. Hubbard <jkh@violet.berkeley.edu> wrote:
>
>Indeed!  What some people don't adequately consider when they scream
>"merge!  merge!" is just what the social effects are going to be.
>From the technical side, it's EASY by comparison!
>
>From the people side, it's anything but easy.  Consider that you'd have
>two groups of generals now working together - some of them still very
>bitter towards the other generals, and looking secretly for any excuse
>to tear the other General's battle plan apart.  It's guaranteed chaos
>in the campaign tent, my friends!  Every agreement would take twice
>as long and be hard fought.  The generals would spend so much time fighting
>with eachother that the war could be lost outside and they might not even
>notice.


It is clear that the generals don't want a merge.  Not much the
rest of us can do about that.

At the upcoming USENIX, BSDI, FreeBSD, and NetBSD are all having
BOFs (informal public meetings) and guess what?  They are all scheduled
for the same time slot so it isn't possible to attend more than one.
Says a lot doesn't it?

But, if there isn't going to be a merge, at least the costs of the
division can be lowered.

Fundementally, the split between NetBSD and FreeBSD is an internal
affair.  You're all on the same side if you draw any but the tiniest
of circles.  The problem is that most of the people who are interested
in *BSD draw a circle that encompasses both camps.  This includes
people outside that circle who look in from (say the Linux camp)
and see a group split by artificial divisions and unable to get
its own house in order.   It includes people in the DOS/Windows
world who can ignore *BSD much easier than they can ignore Linux.
It includes users, like me, who have friends in both camps and see
features of both operating systems that I would like to have.  It
includes programmmers (like me, kinda -- just wrote one tiny one
device driver), who would like to contribute to both efforts but
find the incompatability a major hurdle.


Source an binary compatability between the camps is extreamly important.

Many programs port from one bsd to the other with no changes (csh).  Many 
other programs require small changes (inn).   Every place where even a
small change is required should be considerd a bug.   Whose bug?  

The FreeBSD camp should consider it a bug whenever a NetBSD program 
fails to compile or its binary to run.

The NetBSD camp should consider it a bug whenever a FreeBSD program 
fails to compile or its binary to run.

Whenever possilbe kernel interfaces should be made identical.


Part of the success of Linux stems from its lack of ego.  Linux includes
interfaces to make things work.  Doesn't matter too much where the
interface came from.  Used to be that SunOS was the system most likely
to be supported by 3rd-party free software.  Now I think it's Linux.  
*BSD is way below critical mass.  Everything that can be done to make
things easier should be considerd.


-Dave