*BSD News Article 48656


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!simtel!swidir.switch.ch!scsing.switch.ch!news.belwue.de!fu-berlin.de!news.mathworks.com!newsfeed.internetmci.com!news.uoregon.edu!news.bc.net!felix.junction.net!okjunc.junction.net!michael
From: michael@okjunc.junction.net (Michael Dillon)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: FreeBSD social event Sept. 2/3
Date: 15 Aug 1995 00:40:13 GMT
Organization: Okanagan Internet Junction, Vernon B.C., Canada
Lines: 93
Message-ID: <40oqdd$8en@felix.junction.net>
References: <DD6HuH.HEA@bonkers.taronga.com> <40mb4i$aei@masala.cc.uh.edu> <40mmgh$cuo@agate.berkeley.edu>
NNTP-Posting-Host: okjunc.junction.net

In article <40mmgh$cuo@agate.berkeley.edu>,
Jordan K. Hubbard <jkh@violet.berkeley.edu> wrote:
>In article <40mb4i$aei@masala.cc.uh.edu>,
>Woody Jin <wjin@hermes.cs.uh.edu> wrote:
>>I suggest that the name FreeBSD be changed to a better one.
>>I understand that most FreeBSD users are unix experts, and they understand
>>what is going on.  However, majority others are not, and they don't bother
>>to be.
>
>Sorry, Woody, but that's just too impractical.  We've already been
>through this before, and all that the various anti-FreeBSD folks
>could say was that we needed "a better name."  When pressed for
>actual SUGGESTIONS they either shrugged and said "I don't know,
>something better than FreeBSD" (really big help, thanks) or suggested
>something even worse, like "OpenBSD."
>
>People who pick an OS based on its name will always be people
>who pick things for the wrong reasons, and I'm not really much
>inclined to go out of my way to support people with such poor
>judgement.  What would be the point?  You think we want to deal
>with the kind of questions we'd be likely to get from someone who picked
>"WowBSD" just because it had a cool name?

I think there is a middle ground here that is worth exploring. It is 
related to the middle ground between people who hate NT/VMS registry and 
those who want GUI config. Fundamentally, there are two camps here and 
there always will be two camps. Over time the ease-of-use camp is likely 
to dominate in terms of sheer numbers while the under-the-hood camp is 
likely to be doing the majority of development on the system

So it would be good to find a middle ground that can resolve the struggle 
between these two groups. I believe that this can be resolved by moving 
towards two official FreeBSD distributions over time. The first one will 
be the FreeBSD core distribution with no GUI config and minimal apps and 
frills. For instance, the core would not include Xfree, nor PERL, nor 
TCL/Tk, nor INN, etc... It would be directed to the set of users who roll 
their own systems and add the pieces they want or need. Person X will add 
MGR as their GUI, person Y will put on Xfree and OpenWin, Person Z will 
have Xfree and mwm. This core distribution should retain the name FreeBSD.

But I think there is a lot to be said for developping a second fuller 
distribution of the system that is focussed more on the masses with 
ease-of-use as a major goal. This distribution would include "standard" 
components. Xfree with fvwm and TCL/Tk would form the GUI environment. 
There would be no twm, mwm, olvwm support at all. Perhaps PERL might not 
even be included. Or if it is included it will be configured for dynamic 
loading and will come with the GIF, Curses, and Tk modules. The exact 
layout is something that would need to be decided over time. This 
distribution could very well have a new name like Sumatra or Visual 
Gateways or something.

A user who chooses the Sumatra boot diskette would be presented with a 
full GUI install from the start (probably text-mode GUI to begin with), 
while a user who picks up FreeBSD will get something a little more malleable.

Under the hood, of course, Sumatra would merely be FreeBSD with a 
standard set of packages added on and and integrated GUI admin system 
that is as powerful and as flexible as the NT/VMS registry. Hold those 
screams of anguish! The fundamental concept of NT's registry is that 
there is a single admin utility that can administer everything. They have 
chosen to implement this with a single binary admin database but there is 
no reason why this should be so. FreeBSD could very well have a central 
admin utility that adminsiters the exact set of existing config files in 
exactly the same places. This means that when your neighbour asks for 
help with his Sumatra system, you can plow in under the hood and fix 
things the way you know how, and your neighbour can look in the registry 
utility to see what you have done. It would be an interesting challenge 
to build a scriptable general purpose admin utility that can handle DNS 
zone files and INN's newsfeeds file and /etc/netstart including comments 
and commented out sections of real config stuff, but I don't think it is 
an impossible task.

>I'm fully aware that marketing is important, but rather than spend
>months debating a new name for FreeBSD I'd much rather put the energy
>into _improving the product_.  Sorry, I'm just kind of silly that
>way.. :-)

Good point. An easy to use O/S with a fancy name is no good unless it is 
built on a strong core. FreeBSD is well on the way there, but it's not 
done yet. However, the skill set required for much of the existing work 
is not the same as the skill set required for DESIGNING and building an 
easy-to-use killer O/S package. There is room here for a parallel track. 
A nice side effect of doing it that way is that the parallel track either 
gains developpers and proceeds, or it fades away due to lack of interest 
while not interfering with the core development track.

It's not like this hasn't been done before. Look at SCO UNIX and Open 
Desktop or Open Server. There are still people buying plain SCO UNIX but 
the marketing effort is focussed on the other named products.
-- 
Michael Dillon                                    Voice: +1-604-546-8022
Memra Software Inc.                                 Fax: +1-604-542-4130
http://www.memra.com                             E-mail: michael@memra.com