From: Matt Blaze <mab@crypto.com>
Date: Sun, 24 Dec 1995 04:45:49 +0800
To: Jim Choate <ravage@einstein.ssz.com>
Subject: Re: CFS and Linux (fwd)
In-Reply-To: <199512231633.KAA04297@einstein>
Message-ID: <199512232014.PAA23147@crypto.com>
MIME-Version: 1.0
Content-Type: text/plain


> If you are going to do it, do it right the first time.
> 
> Second, I really doubt Blaze has a problem obtaining access to computing
> power and a $25 CD should be within his budget.
> 

What, exactly, is your complaint?  I'm honestly confused.

Please see my previous message on this subject for an explanation of
the situation regarding CFS with respect to Linux.

I'm not "refusing to support" anything.  I want to do whatever it takes
to get CFS (and other applications of strong cryptography) as widely
deployed as possible.  I am unwilling to allow supporting CFS to become
a full-time job, however, and I'm not going to apologize for that.

I have a Linux machine, in fact.  But I've tried to run CFS on it and it
seems to work fine.  I've not investigated further because that seems like
a poor use of my time given the large number of Linux experts (who know far
more than I ever will about the vagaries of the various Linux releases) who
have not come up with a satisfactory, general patch that runs on all the
various Linux platforms.  The problem seems to be that some versions of Linux
include an rpcgen that produces non-standard output.  I don't have one
of those versions, however, so I've not encountered this "problem" myself.

Again, if you want to see CFS, or any other software that I distribute, run
on some platform that I don't have, you are welcome to send me patches that
I will happily wrap into the distribution (as long as it doesnt break the
other supported platforms).  Until you do that, you have nothing to whine
about.

-matt