Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 17 Mar 2002 02:08:51 -0800
From:      Terry Lambert <tlambert2@mindspring.com>
To:        Hans Reiser <reiser@namesys.com>
Cc:        Chris Mason <mason@suse.com>, Josh MacDonald <jmacd@CS.Berkeley.EDU>, Parity Error <bootup@mail.ru>, freebsd-fs@FreeBSD.ORG, reiserfs-dev@namesys.com
Subject:   Re: [reiserfs-dev] Re: metadata update durability ordering/soft updates
Message-ID:  <3C946B33.888F2281@mindspring.com>
References:  <E16lReK-000C3T-00@f10.mail.ru> <3C910C57.71C2D823@mindspring.com> <20020315065651.02637@helen.CS.Berkeley.EDU> <3C923C91.454D7710@mindspring.com> <1562810000.1016224776@tiny> <3C928D21.404EA11D@mindspring.com> <1714680000.1016298986@tiny> <3C93BBF1.7E8801DF@mindspring.com> <3C946B57.3060403@namesys.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Hans Reiser wrote:
> Oh this is crap.  There is nothing that Chris does in our journaling
> code that wasn't already done in databases for years before this patent
> was issued.  (Chris, while you implement better than they did, I don't
> think your architecture is at all new.)

This is irrelevent to the legal system.

> As for your claiming you don't want to discuss it, this is bullshit,

As a Senior Software Engineer for Novell UNIX Systems Group,
which integrated the former USL in June of 1994, in late 1994,
during which time my primary job responsibilities included
developing an attributed FS for UNIX for use in the NetWare
for UNIX 4.x, I reviewed the DOW patent claims and materials
prior to the filing by the authors.

This event should be verifiable with Bryan Sparks, Gary
Tomlinson, Jim Freeman, Darren Davis, and other people
curently or formerly working for Caldera Systems, in Orem
Utah, and well known to the Linux community.  Actually, I'm
pretty sure Jim Freeman reviewec the same documents.

It is my opinion that the ReiserFS management of preserve
lists probably infringes US Patent 5666532.  It's no secret
that I hold this opinion; I have posted consistently on it
for several years, ever since the 0.2 release of ReiserFS
included Write Ordering for shifted tree items.  I can give
you list references back at least a year, or even earlier,
if I hit my offline mail archives.

It's also no secret that I think it's easy to get around
this by changing from DOW technology to the Ganger/Patt
Soft Updates technology, which would make it a non-issue.


> you are spreading FUD about our product in a potential future
> market for ReiserFS in a manner that could discourage someone
> from paying for the port.  This is extremely irresponsible.
> Don't pretend to be friendly, your actions are quite
> harmful and irresponsible.

Actually, the ReiserFS-Dev list was added to the Cc: line
of a thread on the FreeBSD-FS list by Josh McDonald; I
would just as soon not be quoted in part and out of context
on a list where the entire thread was not archived.  If
I had noticed the addition, I would have removed it from
the "Cc:" list befre replying to his posting.

For FreeBSD, unless you are building a commercial product
based on FreeBSD and negotiate a seperate license, ReiserFS
under the GPL is a no-op, since you could not ship a binary
for FreeBSD that was capable of booting off ReiserFS, due
to license incompatability with the GPL.  This is the same
technicality that keeps the FreeBSD community from supporting,
in a non-fringe way, a port of XFS or JFS to FreeBSD.

FWIW: almost every UNIX vendor to whom you would market the
code has a license for SVR4.2, which includes a license for
use of the DOW Patents, so it's a non-issue for most
potential commerical licensees of the code.


If it makes you any happier, SQUID infringes at least 5 IBM
patents.  When I was employed by IBM, we were forced to remove
it from an unreleased IBM product (they acquired our company
prior to releasing the product) to avoid granting royalty free
licenses to use those patents to anyone who bought a $1500
product and demanded the sources to the SQUID code from IBM
under the terms of the GPL.

Just because a company has a patent doesn't mean it will sue;
if the patent had been transferred from Novell to SCO along
with USL, the problem would be moot.  As it is, I'm not sure
whether or not the license is sublicensable (I assume that
it is), so the shipping ReiserFS on Caldera's OpenLinux
Workstation 3.1 may in fact destroy the enforcibility of the
patent by Novell, in any case.  The people to ask on that are
Caldera; my assumptions are not strong enough for me to take
the risk, so they shouldn't be strong enough for you.

My opinion of software patents is probably lower than yours,
but they are a fact of life in this business.

-- Terry

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-fs" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3C946B33.888F2281>