Re: ARC patent

Greetings,

I would like to contribute my $.02 to this issue.

I speak as not a lawyer but as someone tho worked
one and a half year in a patent bureau and even
got a certificate from WIPO (http://academy.wipo.int/
those who interested may attend the course too, it
is free).

First, the whole point of USPTO's publishing patents
which are pending is to get it publicly reviewed and
collect objections before final decision. So, those
of you who live in US file and objection based on
"USENIX File & Storage Technologies Conference
(FAST), March 31, 2003, San Francisco, CA" mentioned
by Jan Wieck. Filing and objection should be not
be too expensive though you may need help of
professional lawyer form a patent bureau co compose
a solid objection.

I will call my old friends from Patent Bureau tomorrow
to get a professional advise on this matter.

Second, a pending patent is not a granted patent,
one is not infringing anything by distributing
technology based in a pending patent. As soon
as patent is granted AND "Cease and Desist" letter
form IBM is received removing offending code, removing
offending versions from download and and notifying
customers to upgrade to a new version is sufficient.
I am not sure about CVS, apparently it need to be
cleared out too.

A vaguely similar issue happened between Pixar,
the developer of Renderman and Exluna the
developer of BMRT, a free (but not open
source) raytracing 3D renderer. Pixar sued Exluna
for willful patent infringement. Exluna released
a new version of BMRT - 2.6 without offending
technology and ensured that version 2.5 is removed
from all mirrors. For quite a lot of time
-and even now- one of the most valuable things
a 3D designer may own is a copy of BMRT version 2.5.
Exluna was intended to defend themselves in court
but soon ran out of money, settled with Pixar
and was swallowed by nVidia. A sad story indeed.
A story of how a big company squashes a small one
using patents. Read more at:
http://www.renderman.org/RMR/OtherLinks/blackSIGGRAPH.html

The point here is that IBM may force PostgreSQL Global
Development Group to remove offending version if
patent is granted.

But, lastly, as it was pointed out before it would
be a very bad publicity for IBM and, in my opinion,
very good publicity for PostgreSQL. IBM will admit
that PostgreSQL is a worthy competitor. Thus, in my
personal opinion IBM will never threat PostgreSQL.

We can remove offending code but host patches to
introduce the code in a country that does accept
software patents. It would be even better for
publicity.

IBM can NEVER sue customers for using infringing
code before first informing them of infringement and
giving reasonable time to upgrade to uninfringing
version.

So, in short my advise is:

1. File an objection with USPTO. And maybe an informative
letter to IBM legal department mentioning USENIX paper.
2. If patent is granted, contact IBM and request
an unlimited, perpetual license to use the technology
3. If IBM refuses, remove the offending code, clean up
CVS and shout from the rooftops about the hypocrisy of
IBM.

Hope it helps make up your mind,
Best regards,
Nicolai Tufar

P.S. But if filing date really is 2002 and there
is no prior art me may skip step 1.