Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 3 Dec 1996 20:44:14 -0500 (EST)
From:      "John S. Dyson" <toor@dyson.iquest.net>
To:        lm@neteng.engr.sgi.com (Larry McVoy)
Cc:        dyson@freebsd.org, lm@neteng.engr.sgi.com, thorpej@nas.nasa.gov, davem@jenolan.rutgers.edu, jkh@time.cdrom.com, dennis@etinc.com, kpneal@pobox.com, hackers@freebsd.org, torvalds@cs.helsinki.fi, lm@relay.engr.SGI.COM, iain@sbs.de, sparclinux@vger.rutgers.edu
Subject:   Re: TCP/IP bandwidth bragging
Message-ID:  <199612040144.UAA02462@dyson.iquest.net>
In-Reply-To: <199612032219.OAA22523@neteng.engr.sgi.com> from "Larry McVoy" at Dec 3, 96 02:19:16 pm

next in thread | previous in thread | raw e-mail | index | archive | help

> 
> My, my.  You sure are sensitive.  When I wrote that, my thinking was
> it was *Linux* that was the bullshit, not FreeBSD.  Because of all your
>
I don't know how I could not have interpreted your statement about my
carefulness or conservativism in giving numbers out as saying it is
"bullshit."  It was obvious that the "numbers" weren't us (I don't normally
give any out.)  Sometimes I can be coerced to talk about them (when
pressured), but it is something that I generally hold back.

Here Larry McVoy's statement:
>
>  You've heard the old saying: numbers
> talk, bullshit walks.
>
Sorry, again I HAVE NOT GIVEN NUMBERS OUT RECENTLY.  I am not talking,
so my position is bullshit...  Very easy to interpret it that way...

> 
> As the old commercial goes: Where's the beef?
> 
If I misinterpreted your position,  I am sorry, but others have
apparently interpreted your statments the same as me...  Please refer
to other irate responses.  I think that your interpretation of the
situation is in error.

>
> Don't fill my mailbox with flames, fill it with test programs (or real
> application suites) that show FreeBSD to be better.
>
Let the user base decide, I study FreeBSD (and once in a while Linux)
to find ways to improve FreeBSD.  I check out Linux to make sure that
we aren't falling behind in any important areas.  I very well know the
differences in performance between FreeBSD and Linux, but it isn't in
my demeanor to want to deal with the issues after putting the info
into public view.

>
>  I can easily believe
> that you have real applications that perform better under FreeBSD - so
> what are they?  
> 
Likewise...  What are the applications that perform better under Linux,
so I can improve FreeBSD?

> 
> I am saying "I believe that FreeBSD does some stuff better than Linux and
> I believe that you can prove it.
>
I believe that it is Linux's job to find out where their perf problems are.
I don't publish my "bogostones" :-) and frankly I don't want to.  As soon
as I quote benchmarks, then I need to supply information about them, then
I would have a responsibility to disclose information that I don't want
to.  Much of the problem in fixing the perf problems in an OS is finding
them...  I have spent time finding problems and performance issues, and
don't necessarily want to give the fruits of that labor away.  I will
release some of my tools to certain groups, if they want.  But it is
my choice.

I freely give away source code that I work weeks on, but I also intend to
maintain the freedom to choose what I give away also I intend to keep it that
way.  (These are my kind-of part of my anti-GPL sentiments.)

>
>  I'd like to see you do so.  I'd like to
> learn what you did that is so great because I might learn something."  There
> is nothing negative, implicit or explicit, in that statement.  Stop looking
> for slams - they aren't there.
> 
I guess that I have had conditioning while dealing with your condescending
attitude in the past.  I simply expect it from you.  It takes time to build
up trust.  Also, if you are so interested in FreeBSD, why do I only hear
from you during flame wars?  This is wierd.

>
> Instead, show up with the data that says "FreeBSD can support 10 zillion
> web hits a day and Linux can't - here is the benchmark that proves it".
> Or whatever else it is that you think is important.  Here's your chance
> to prove how good FreeBSD is.  So just do it.  
> 
I don't care to spread around the results of how much better FreeBSD performs
than ANY OS.  Let's let the users evaluate and decide.  FreeBSD is way way past
the chest beating stage (and sure wish others would quit also.)

But, I am sure if someone specifically posts a totally bogus benchmark comparing
FreeBSD to BOGOos or what not, I'll look into it and challenge untruths or
misleading statements.  I certainly am not going to brag about things
or initiate some kind of chest-beating session.  I don't need to.

I wish that many Linux people could do the same.  Note that I have
another opinion about  Linux advocacy -- wonderful, super high performance
benchmarks on FreeBSD won't convert most of the Linux user base.  They
love Linux, it is an emotional attachment that defies reason and sane
thinking.  Why shouldn't we (FreeBSD) just do the best OS that we can,
and when users that are really in a pickle with an inferior OS, they will
look towards FreeBSD as a possible solution.

John




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612040144.UAA02462>