*BSD News Article 36999


Return to BSD News archive

Newsgroups: comp.os.386bsd.misc
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!uwm.edu!spool.mu.edu!sgiblab!pacbell.com!amdahl!netcomsv!netcom.com!jmonroy
From: jmonroy@netcom.com (Jesus Monroy Jr)
Subject: Re: Nailed down to 386bsd or linux, now which one?
Message-ID: <jmonroyCxqI6J.Loq@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: TIN [version 1.2 PL1]
References: <jmonroyCx10D8.3yu@netcom.com> <378hjr$c4@knobel.gun.de> <jmonroyCxG277.1ML@netcom.com> <37k59n$14m@u.cc.utah.edu>
Date: Sat, 15 Oct 1994 21:54:19 GMT
Lines: 82

Terry Lambert (terry@cs.weber.edu) wrote:
: In article <jmonroyCxG277.1ML@netcom.com> jmonroy@netcom.com (Jesus Monroy Jr) writes:
: ] Andreas Klemm (andreas@knobel.gun.de) wrote:
: ] : Jesus Monroy Jr (jmonroy@netcom.com) wrote:
: ] : :
: ] : : 	Again, no benchmarks (or stats) are available for
: ] : : 	any of the above.  Numbers generated depend heavily 
: ] : : 	on equipment used and application (program) design.

: [ ... benchmarks from a German magazine ... ]

: ] 	Worthless benchmarks and you know it.
: ] 	I could publish daily reports on how fast 386bsd
: ] 	is in comparision to say, Novell servers.... doen't
: ] 	prove diddly...
: ] 
: ] 	At best the magazine you're quoting says that
: ] 	"when they did the test, with their machines
: ] 	under thier conditions.... they got xYZ results."

: I know of no benchmarks which are not themselves similar statements.

: What benchmarks (since it was your original demand that prompted Andreas'
: posting) would *you* find acceptable?

: Please define them in terms of what they do rather than the results you'd
: like to see them produce (ie: "benchmarks showing 386BSD 1.0 as better
: than all other OS's" is an illegal condition).
:
	As you know, Terry, benchmarking is a science.
	To be complete on this, a technical committe would
	be the best thing.  Certainly some of the older
	traditional benchmarks (I.E., Livermoore loops) make some
	sort of statement, but for a fair complete benchmark
	I would consider no less that 200 tests all weighted 
	and all running on the exact same machine.

	Of course, this is quite a silly proposition, but
	then again my full solution would be by committee.

: ] : BSD is a standard. So there's a lot of documentation around in 
: ] : a very good quality. 
: ] 
: ] 	If you think a bunch of contorted thirty year old
: ] 	man pages based on outdated code is documentation,
: ] 	you must be kidding.

: I don't speak for all of us, of course, but many of us are contorted
: thirty year old men, so this is appropriate documentation.

: Seriously, things like the online documentation on NeXT and Solaris and
: UnixWare suck just as badly and are frequently less usable.
:
	Exactly why you see "/documentation" on my tag line.

: ] : Manual pages are in general more complete and more up to
: ] : date than in the rapidely changing Linux environment.
: ] :
: ] 	I guess your planning on using stone knives 
: ] 	and bear skin clothing for a while.

: You haven't pointed at better tools, only criticised existing ones.

: This is not constructive.

: What existing systems do you consider to be above the quality level of
: "stone knives and bear skin clothing"?
:
	I think that the WWW enviroment is a good model.
	definitely a hypertext type model is most appropriate
	in my book.

	To give you a better answer, I will post later this
	weekend and I have for an improved text (ascii) 
	editing enviroment.


-- 
Jesus Monroy Jr                                          jmonroy@netcom.com
Zebra Research
/386BSD/device-drivers /fd /qic /clock /documentation
___________________________________________________________________________