X-Sympa-To: patents
Return-Path: <mark@openpatents.org>
Delivered-To: patents@aful.org
Received: from localhost.localdomain (user-206-158-52-219.postridge0.noment.net [206.158.52.219])
	by stabilite.aful.org (Postfix) with ESMTP id E77F93E79C
	for <patents@aful.org>; Thu,  2 Jan 2003 14:22:57 +0100 (CET)
Received: from localhost.localdomain (k9 [127.0.0.1])
	by localhost.localdomain (8.12.5/8.12.5) with ESMTP id h02DOMZv027654;
	Thu, 2 Jan 2003 08:24:22 -0500
Received: (from mark@localhost)
	by localhost.localdomain (8.12.5/8.12.5/Submit) id h02DOC5w027652;
	Thu, 2 Jan 2003 08:24:12 -0500
X-Authentication-Warning: localhost.localdomain: mark set sender to mark@openpatents.org using -f
Subject: Re: [Patents] pooling patents, version 2
From: Mark Shewmaker <mark@openpatents.org>
To: riel@nl.linux.org
Cc: patents@aful.org
In-Reply-To: <Pine.LNX.4.50L.0301012145210.2429-100000@imladris.surriel.com>
References: <Pine.LNX.4.50L.0301012145210.2429-100000@imladris.surriel.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 02 Jan 2003 08:24:11 -0500
Message-Id: <1041513851.5875.213.camel@k9>
Mime-Version: 1.0
X-Validation-by: mark@primefactor.com

On Wed, 2003-01-01 at 18:46, riel@nl.linux.org wrote:
> Changed after feedback from many people
> 
> 
> 	Pooling patents for progress and protection
> 			version 2

Excellent!

> 3) If somebody starts an IP related lawsuit against any of the
>    members of the patent pool, the plaintiff forfeits the right
>    granted in (2) and will no longer be allowed to use the technology
>    described in any of the patents from the patent pool.  This means
>    that the whole "arsenal" of the patent pool is available to the
>    participants, for defensive use in patent lawsuits.

I've tried to stay away from the poison pill approach, somewhat because
I'm afraid of it opening up more risk for the honest participants than
any benefits it gains them in, but mostly because it seems to make the
idea lose focus.

I would want to concentrate on pulling people up, letting them use
patents cooperatively if they so desire, or not if they don't.  There
are already mechanisms (lawsuits) to address the case of people who
don't play by the rules.  I don't see a need to add more.

(And as much as it sounds like flamebait to say this:  To me it's
reminiscent of governments making laws making things that are already
illegal more illegal just for people to feel better about it.  Often
these laws have other unintended consequences.)

Here are some potential consequences I see for a poison pill as you
suggest:

Suppose Company A distributes a program they own the entire rights to
under the GPL, (which includes its own implied patent grant), and the
program includes patented methods for something-or-other for which
Company A owns the patents.

Suppose that Company A also owns unrelated patents, and also distributes
unrelated programs.

Further suppose that Company A licenses the set of patents used in this
one program in the way you describe.

Problem 1:  The poison pill in this patent license conflicts with the
implied patent grant of the GPL.  The license you describe has
conditions that seem to me to be more restrictive than the GPL would
allow.

Problem 2:  Company A is now in a position to be able to use any other
participant's IP, but the participants can't sue them for this without
forfeiting rights to Company A's contributed patents, no matter how
legitimate the lawsuit might be.

Problem 2A: Imagine, for instance, that Company A has incorporated code
from someone else without permission.  This other person would like to
sue, but that would mean they couldn't use Company A's products any
more.

Problem 2B:  You're implying that a participant can contribute just one
patent.  However, if part of the agreement is a poison pill against any
IP lawsuits, then effectively they can't contribute just one patent.  If
Company A has contributed just one patent, then they can't sue any other
participant who uses their uncontributed patent without permission, at
least without Company A losing rights to the rest of the patents in the
pool.

> 4) A participant can withdraw from the patent pool, losing the rights to
>    use the patent pool for protection in the future.  However, once a
>    patent has been contributed to the pool, that patent cannot be
>    withdrawn from the pool and will stay in the pool until the patent
>    expires; this to protect the other participants in the pool.

I would suggest a few additions:

1.  An initial change-your-mind "Oh s#%@#!" timeout, to help address the
case of larger companies contributing patents that other sections of the
company might object to.  Even just a week or so time period to change
their mind could help convince companies to join.  Sometimes in larger
companies no internal communication is as good as reading a published
press release.  One case of a large company inadvertently losing
millions in a planned product release because of a loss of a patent
monopoly could send shivers down the spines of future corporate decision
makers.  Letting them pull back a patent after the first week or so
could prevent that, and won't cost us nearly so much.  (And having this
in the license and rationale will help show that we are interested in
this being a win-win scenario, minimizing unexpected risks where we
can.)

2.  The patent technically stays in the pool forever.  This is to
counter problems with a patent expiring and then being resurrected
copyright-extension style.  Also it could counter problems about some
patents not being valid in some jurisdictions, then being ruled valid
again, etc.

3.  For any "I've submitted all patents" cases, especially if that gains
you a new level of participation or access to another pool, this should
include any invention made up through the withdrawal date.  No fair
saying you're submitting all your patents, have everyone use and
recommend you as an upstanding company, you get all this attention and
extra income to fund R&D to make an invention that you don't
contribute.  Patents for inventions made while under the agreement
should still be in this pool.

> 5) If a participant's last patent in the pool expires, the participant
>    needs to do an invention and contribute a new patent to the pool.
>    This clause is present because the patent pool should promote
>    technical progress and inventions, like patents are supposed to.

If you have the addition I suggest in #3 above, you obviously don't want
this clause.

As far as other things to add to the license:

You may also want to consider adding what I've so far called "Patent
Like IP"s.  (Got to think of a better name than PLIPs.)

The upshot is that in theory the following is (more or less) supposed to
be true:  "Patents protect ideas, copyrights protect specific
implementations, and trademarks protect company and brand identity."  
However, in actual fact there are non-patent forms of IP that restrict
people in ways similar to patents.  It does no good if this sort of
patent licensing agreement becomes very well accepted and used by many
companies if participants can still restrict each other from other
implementations of ideas in non-patent ways, even if they technically
can't restrict each other using patents per se.

For instance, user-interface copyrights could do this.  (They're not
looked upon favorably by the courts at the moment, but that could
change.)  Trade dress could do this.  Restrictions on reverse
engineering could do it too.

This kind of throws a wrench into the works, but, well, it's a potential
loophole in a patent-only license.

> This patent pool is primarily meant for small and medium size companies,
> who have few patents each and are a frequent target for bullying and
> predatory practices by large patent holders, who use patents as a way
> to prevent small companies from entering the market, thereby stifling
> technical progress itself.
> 
> This patent pool should:
> 1) protect inventors and companies making products
> 2) promote technical progress
> 3) be a win-win play for all participants

Excellent!  All very well said.

 -Mark Shewmaker
  mark@openpatents.org

