*BSD News Article 64056


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!newshost.telstra.net!plaster.csdc.toshiba.com.au!inferno.mpx.com.au!goliath.apana.org.au!news.cs.su.oz.au!harbinger.cc.monash.edu.au!ns.saard.net!news.camtech.com.au!calypso.bns.com.au!not-for-mail
From: mike@calypso.bns.com.au (Michael Talbot-Wilson)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: God Damn partition crap!
Date: 23 Mar 1996 21:45:55 +1030
Organization: Calypso & That Jazz
Lines: 83
Message-ID: <4j0mhb$r6i@calypso.bns.com.au>
References: <4hqav8$kmo@nntp.interaccess.com> <3140F2B6.41C67EA6@freebsd.org> <4i1qp4$bmj@nntp5.u.washington.edu> <4i50q6$64i@uriah.heep.sax.de> <4i727m$1hl@nntp5.u.washington.edu>
NNTP-Posting-Host: calypso.bns.com.au
X-Newsreader: NN version 6.5.0 CURRENT #3

kargl@troutmask.apl.washington.edu writes:

>     J Wunsch wrote in article <4i50q6$64i@uriah.heep.sax.de> :
>>
>>kargl@troutmask.apl.washington.edu writes:
>>
>>> Are you (or another core team member) going to ...
>>
>>Why do you (and that's not just you, Steven) always expect the core
>>team members to do all the work???

>so I thought he would be the most logical person to strip out the
>unnecessary parts to produce a USABLE tool quickly.

>>
>>We've got 14 core team members, but more than 50 committers.  Despite
>>of this, everybody else (yes, you, you, and also ... you! there) can
>>contribute something to the system, as long as he can find one of the
>>committers to feel responsible for his `product'.
>>
>>Did you ever get the idea that in particular most core team members do
>>already know how to setup a disk, and thus have much less need for
>>such a tool than many other people?

I can really feel for the original poster, and I feel this kind of
response is offensive.

I had the unpleasant experience of puzzling over the unintelligible
disklabel manpage when I had a spare partition on what was otherwise
a FreeBSD 2.0.5 system, and I wanted to enable FreeBSD to use it.  It
seemed that unless you hired a computer science graduate who had done
a course in BSD (i.e. someone had helped him, so he was in a position
to help you) this was simply impossible.   The knowledge was passed on
by word of mouth, it was legendary, it was not intelligibly committed
to writing.  You were lucky if someone you knew had saved mail from
someone who knew.

I would like to be able to feel unalloyed gratitude towards the people
who have made it possible for me, for little or no money cost, to
obtain a Unix operating system.

But when I see people writing about how they are too busy writing
code to write documentation for the software they have published, 
when I contempate the waste of time when perhaps 100 or 1000 people 
(including me) spend 6 weeks each attempting to discover what the 
core team members and seasoned BSD hackers know already, how to do 
a thing that should only take a few minutes, any such gratitude 
disappears.   In their view their time is important, and the time 
of all others, in their opinion lesser mortals, lamers, is unimportant.
They are wrong.

I hope I have misunderstood Jorg Wunsch where he seems to say, and
say sarcastically, that it is unnecessary for the FreeBSD team to 
make available some simple means of installing a partition (in lieu 
of documentation) because they themselves already know how to do it.

It is possible to answer that this is only BSD, that it has long been
documented, if you know where to look.  Well, the disklabel man page
is one place to look, and it is a disgrace.  More than that, the
ready availability of free versions of BSD means it is in effect offered
to people who can have no knowledge of whatever traditional
documentation exists for BSD.  Their only encounter with it is via
FreeBSD.  They need to find the documentation there, and if FreeBSD
has failed to generate a supporting community such as gave rise to
the Linux HOWTOs it falls to the "core team" to write that documentation
or, at the very least, not pass insolent remarks on the subject.

It remains a paradigm of FreeBSD documentation that the /sys tree
is absolutely devoid of it, except in one deeply hidden and incongruously
named file.  Any other distribution would have a README in the top level
directory, but in the FreeBSD kernel source, forget it.  So little would
be required, just two or three lines mentioning the existence of a
program named config and a file it has to read.  Is it seriously
suggested that someone else should write and contribute those two or
three lines because the core team are too important and too busy to
document their own work?

It is so easy and safe to do on Linux the thing that is it virtually 
impossible and seems perilous to do on FreeBSD.  To the person who wrote 
"SHIT SHIT SHIT": toss FreeBSD in the shit can.  It has certain merits 
and attractions, but it is inessential.  There is an alternative which 
is simply free of the maddening frustrations and undocumented gotchas of 
FreeBSD.