W3C home > Mailing lists > Public > www-patentpolicy-comment@w3.org > March 2002

Proposed Patent Policy of February 26, 2002

From: Daniel E. Maddux <dpaladin@hal-pc.org>
Date: Wed, 20 Mar 2002 13:20:55 -0800
Message-Id: <p05100307b8bead66cd08@[192.168.123.158]>
To: www-patentpolicy-comment@w3.org
World Wide Web Consortium
Patent Policy Working Group
www-patentpolicy-comment@w3.org

W3C Patent Policy Working Group:

The Patent Policy Working Group ("PPWG") released a Working Draft on
February 26, 2002.  For the same reasons as I stated in my comments of
October 10, 2001 (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Oct/1398.html)
and November 06, 2001 (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Nov/0014.html),
I object to Sections 2 on Exclusions from RF Commitment and 4 on
Disclosure.  I also think that Sections 3 on Royalty-Free Licensing
Requirements and 5 on Exception Handling should be modified to remove
the potential for abuse which these Sections create.

I.  SECTION 2: Exclusion From RF Commitment

I have 2 objections to Section 2:

   (i)  allowing Members who will not RF license their patents to
continue
        participating in a Working Group
   (ii) allowing Members to exclude patents when joining an already
existing
        RF Working Group

   A.   Allowing Members Who Will Not RF License Their Patents To
Continue
        Participating In A Working Group

I object to Section 2 allowing Members who will not RF License their
patents to continue participating in a Working Group.  In my opinion,
allowing Members who will not RF License their patents to continue
participating in a Working Group creates the potential for these Members
to disrupt and/or abuse the Standard-setting process.  As I stated in
Point 2(i) of my October 10, 2001 comments (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Oct/1398.html),
allowing W3C Members to exclude Essential Claims of a patent or patents
while continuing to participate in a Working Group permits these Members
to negotiate a better/more profitable license for their patents and/or
bias the proposed Recommendation in the Member's favor in exchange for
RF Licensing their patents.  For example, IBM may refuse to RF license
their patents on Speech Recognition while continuing to participate in
the Voice Browser Working Group.  After refusing to RF license its
patents, IBM may advocate drafting the Voice XML 2.0 or 3.0
specification such that the specification reads on IBM's patents.  IBM
may then refuse to change the specfication and thereby force the Voice
Browser Working Group to either abandon the Voice XML 2.0 or 3.0
specification or RAND license IBM's patents covering the Voice XML 2.0
or 3.0 specification.

Alternatively, IBM may advocate drafting the Voice XML specification
such that it does not read on other Members' patents to prevent rival
Member's patents and technology from becoming a web standard.  For
example, MICROSOFT also owns patents on Speech Recognition.  Since both
IBM and MICROSOFT own speech recognition patents, IBM may advocate
drafting the Voice XML specification such that it does not read on any
of MICROSOFT's patents.  By drafting the Voice XML specification such
that it does not read on any of MICROSOFT's patents, IBM prevents
MICROSOFT's speech recognition patents and technology from becoming a
web standard.

Although I am presenting examples of IBM abusing their patents, other
W3C Members are just as prone to abuse their patents in developing web
standards.  MICROSOFT has recently been requiring OEM licensees to
provide MICROSOFT with RF licenses to the OEM's patent portfolios in
exchange for Windows Operating System Licenses (see "Sony: MS already
using Seattlement terms to screw us" at
http://www.theregister.co.uk/content/4/24096.html and "HP, Gateway: MS
Seattlement terms screw us too" at
http://www.theregister.co.uk/content/54/24162.html).

By requiring W3C Members who will not RF license their patents to resign
immediately from the Working Group, the proposed Patent Policy will
prevent W3C Members from abusing the W3C standard-setting process for
their own selfish gain.  If W3C Members who will not RF License their
patents are required to resign from the Working Group, then they will no
longer be able to influence the drafting of the specification.  For
example, requiring IBM to resign from the Voice Browser Working Group
for not RF licensing their patents prevents IBM from influencing the
drafting of the Voice XML 2.0 or 3.0 specification.  Since requiring IBM
to resign from the Voice Browser Working Group for not RF licensing
their patents prevents IBM from influencing the drafting of the Voice
XML 2.0 or 3.0 specification, IBM will not be able to abuse the W3C
standard-setting process for its own selfish gain.

Thus, by allowing W3C Members to exclude their patents from RF Licensing
and continue participating in a Working Group, Section 2 allows these
Members to abuse the W3C standard-setting process for their own selfish
gain.  Since Section 2 allows these Members to abuse the W3C
standard-setting process for their own selfish gain, Section 2 should be
rewritten to *require* W3C Members who will not RF license their patents
to resign immediately from the Working Group.  In particular,
Sub-Sections 2.2.1 and 2.2.2 should be deleted and a new Sub-Section
2.2.1 should be inserted requiring W3C Members who will not RF license
their patents to resign immediately from the Working Group.  By
requiring W3C Members who will not RF license their patents to resign
immediately from the Working Group, the proposed Patent Policy will
prevent W3C Members from abusing the W3C standard-setting process for
their own selfish gain.  By preventing W3C Members from abusing the W3C
standard-setting process for their own selfish gain, my proposed
Sub-Section 2.2.1 will help ensure that W3C standards remain RF and
non-proprietary.

   B.  Allowing W3C Members To Exclude Patents When Joining An Already
Existing
       RF Working Group

For the same reasons that I stated in I.A. above, I think that W3C
Members should not be allowed to join an already existing Working Group
and exclude any of the Member's patents from RF licensing.  In
particular, allowing a W3C Member to join an already existing Working
Group and then choose to exclude its patents from RF licensing provides
the Member with the same opportunities for abusing their patents as
stated above in I.A.; namely, to draft the specification such that it
reads on the Member's patents or else to draft the specification such
that it does not read on a competitor's patents.  Since allowing a W3C
Member to join an already existing Working Group and then choose to
exclude its patents from RF licensing provides the Member with the same
opportunities for abusing their patents as stated above in I.A., Members
should not be allowed to join an already existing Working Group and then
choose to exclude their patents from RF licensing.

Sub-Section 2.2.3 is superfluous and should be deleted from the proposed
patent Policy.  Since a Member knows the patent policy of an already
existing Working Group before joining, the Member who wants to exclude
its patents from RF licensing can exercise the option of not joining the
Working Group.  Since the Member who wants to exclude its patents from
RF licensing can exercise the option of not joining the Working Group,
the Member already has a solution for not RF Licensing its patents.
Since the Member already has a solution for not RF Licensing its
patents, Sub-Section 2.2.3 is superfluous and should be deleted from the
proposed patent policy.

II.  SECTION 3: Royalty-Free Licensing Requirements

I object to the last clause of Sub-Section 7 as allowing W3C Members to
abuse their patents.  The relevant part of Sub-Section 7 states:

   ...but may include reasonable, customary terms relating to operation
or
   maintenance of the license relationship such as the following: choice
   of law and dispute resolution;

The phrase "reasonable, customary terms" is vague.  Since the phrase
"reasonable, customary terms" is vague, it allows W3C Members to
interpret this phrase to further their own selfish interests.
Specifically, this phrase allows W3C Members to add additional terms to
an RF license which hamper open standards.  For example, a consortium
called the Liberty Alliance has proposed a standard for individuals
purchasing items over the internet, while MICROSOFT has proposed a
competing standard called Passport.  As I stated above in I.A.,
MICROSOFT has required Windows OS Licensees to RF License their patent
portfolios to MICROSOFT as part of the licensing contract.  Since
MICROSOFT has required Windows OS Licensees to RF License their patent
portfolios to MICROSOFT as part of the licensing contract, MICROSOFT may
(actually will) also require licensees of their patents that cover a W3C
specification to agree to use only Passport and not the Liberty Alliance
standard.  MICROSOFT will argue that requiring licensees of their
patents covering a W3C specification to use only Passport is a
"reasonable, customary" term and thus meets the requirements of
Sub-Section 7.  In particular, MICROSOFT might require licensees of its
patents covering the Voice XML 2.0 or 3.0 specification to use only
Passport.  Since the phrase "reasonable, customary terms" allows W3C
Members like MICROSOFT to require licensees of their patents covering a
W3C specification to consent to additional onerous terms to the RF
license, Sub-Section 7 should be rewritten to delete the last clause.
Specifically, Sub-Section 7 should be rewritten as follows:

     7.    may not impose any further conditions or restrictions on the
     use of any technology, intellectual property rights, or other
     restrictions on behavior of the licensee;

By removing the last clause of Sub-Section 7, Sub-Section 7 now prevents
W3C Members from abusing their patents by adding onerous terms to RF
licenses of these patents.

I am also wary of allowing patentees to use alternative dispute
resolution ("ADR") to resolve conflicts with licensees.  I think
allowing patentees to specify ADR for resolving conflicts with licensees
also has the potential for abuse.  Specifically, I think the allowing
patentees to specify ADR may allow patentees to choose forums that are
partial to the patentees for resolving these conflicts.  By allowing
patentees to specify ADR may allow patentees to choose forums that are
partial to the patentees for resolving these conflicts, this proposed
patent policy could be unfairly discriminating against licensees who do
not have the financial resources to contest unfavorable outcomes in ADR.

III.  SECTION 4: Disclosure

I reiterate my objections stated in Point 2 of my October 10, 2001
comments (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Oct/1398.html)
and "SUMMARY OF MAIN ISSUES DISCUSSED: 2. RF OR RAND OR MIXTURE" in my
November 06, 2001 comments (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Nov/0014.html).
Specifically, I think that all W3C Members should either commit to RF
Licensing their essential patents when joining a Working Group or be
required to search for and disclose all essential patents relating to a
Working Group's proposed specification.  I think that requiring Members
of a Working Group to search for and disclose all essential patents
relating to a Working Group's proposed specification is necessary to
prevent Members from seeking "submarine" patents like RAMBUS did with
the SDRAM standard for JEDEC.

Searching for patents is not a difficult or onerous task.  At the very
least, Members can search the United States Patent and Trademark
Office's Patent Database at http://www.uspto.gov.  Since the United
States is the primary issuer of software patents, most patents which may
affect a W3C Working Group will be United States patents.  Furthermore,
the World Intellectual Property Organization ("WIPO") maintains a
database on the patent laws of most industrialized nations (see *).
Since the United States Patent and Trademark Office and WIPO maintain a
database of issued patents, requiring W3C Members to search the patent
databases of the United States Patent and Trademark Office and WIPO is
not a difficult or onerous task.  Since requiring W3C Members to search
the patent databases of the United States Patent and Trademark Office
and WIPO is not a difficult or onerous task, W3C Members should be
required to search for patents affecting a Working Group's
specification.

I also think that attorneys, both in-house counsel and retained law
firms, should be subject to the proposed patent policy's disclosure
requirements.  Since W3C Members claim that they have difficulty
searching their patent portfolios, the W3C should require the Members'
attorneys to search and disclose relevant patents.  Furthermore, I
notice the MICROSOFT retains an attorney for participating in the Patent
Policy Working Group (Mark DeLucca or Woodcock Washburn).  Since
MICROSOFT retains an attorney for participating in the Patent Policy
Working Group, I fail to see why other Members cannot also retain
attorneys for participating in the Patent Policy Working Group.  And if
MICROSOFT and other Members can retain attorneys for participating in
the Patent Policy Working Group, then they can also retain attorneys for
searching their patent portfolios for patents affecting a Working
Group's specification.  Since the W3C Members can also retain attorneys
for searching their patent portfolios for patents affecting a Working
Group's specification, this proposed patent policy should require these
Members to subject their attorneys to these patent disclosure
requirements.

Requiring Members' attorneys, both in-house counsel and retained law
firms, will promote the full disclosure of Members' patent portfolios.
Members' patent attorneys prosecute (i.e., draft, apply for, and obtain)
patent applications for the Member.  Since Members' patent attorneys
prosecute (i.e., draft, apply for, and obtain) patent applications for
the Member, they will be most familiar with a Member's patent portfolio.
Since Members' patent attorneys will be most familiar with a Member's
patent portfolio, they should also be subject to the W3C's patent policy
on disclosures.  By making Members' attorneys subject to the W3C's
patent policy on disclosures, this patent policy will promote the full
disclosure of Members' patent portfolios.  By promoting the full
disclosure of Members' patent portfolios, this proposed patent policy
will prevent Members from using their patents to obstruct a Working
Group's ability to draft specifications for new web standards.  By
preventing Members from using their patents to obstruct a Working
Group's ability to draft specifications for new web standards, this
proposed patent policy will promote Working Group's ability to develop
new and open standards for the world wide web.

   A.  Sub-Section 4.3 Good Faith Disclosure Standards

Sub-Section 4.3 "Good Faith Disclosure Standards" should require a
patent search if a Member does not agree to RF License all patents
essential to a Working Group specification.  The first sentence of
Sub-Section 4.3 states in relevant part:

     Satisfaction of the disclosure requirement does *not* require a
patent
     search, ...

For the reasons stated above, I disagree.  I think that satisfaction of
the disclosure requirement *does* require a patent search to prevent W3C
Members from pursuing submarine patents on W3C specifications.
Furthermore, I think that W3C Members should be prohibited from filing
patent applications on proposed or adopted W3C specifications to also
prevent Members from obtaining submarine patents on W3C specifications.
As I discuss in "2. RF OR RAND OR MIXTURE" of my November 06, 2001,
comments (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Nov/0014.html),
not requiring Members to perform patent searches encourages Members to
obtain and use submarine patents covering W3C specifications and
standards.  Futhermore, not prohibiting Members from filing patent
applications on W3C specifications and standards encourages them to do
just that.  Since not requiring Members to perform patent searches
encourages Members to obtain and use submarine patents covering W3C
specifications and standards, this proposed patent policy should require
Members to search their patent portfolios for essential patents
affecting a Working Group's purpose or specification.  Since not
prohibiting Members from filing patent applications on W3C
specifications and standards encourages them to do just that, this
proposed patent policy should also prohibit Members from filing patent
applications on W3C specifications and standards.

Invited Experts should also be required to search their patent
portfolios for essential patents affecting a Working Group's
specifications or standards.  Invited Experts will usually own, or by
employed by the owner of, patents that may cover a Working Group's
specification or standard.  Since Invited Experts will usually own, or
by employed by the owner of, patents that may cover a Working Group's
specification or standard, they will have an economic interest in
drafting the Working Group's specification or standard such that the
Invited Expert's patents cover the Working Group's specification or
standard.  Since an Invited Expert will have an economic interest in
drafting the Working Group's specification or standard such that the
Invited Expert's patents cover the Working Group's specification or
standard, they should be subject to the same disclosure requirements as
Members.

IV. SECTION 5: Exception Handling

I object to Section 5 for not including a procedure for allowing a
Patent Advisory Group to solicit prior art from the public and use this
publicly solicited prior art to invalidate patents which read on a W3C
specification and which are not available on RF licensing terms.  As I
discuss in "A. The Patent Policy Working Group is pro-patent" of my
November 06, 2001, comments (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Nov/0014.html),
a Working Group has the option of soliciting prior art from the public
and using this prior art to invalidate a patent which reads on a W3C
specification.  Since a Working Group has the option of soliciting prior
art from the public and using this prior art to invalidate a patent
which reads on a W3C specification, this proposed patent policy should
recognize and include this option in Section 5.

   A. Sub-Section 5.2: PAG Procedures

As I stated in the above paragraph, PAGs should have the option of
soliciting prior art from the public and filing a reexamination request
or declaratory judgment action to invalidate a patent covering a W3C
specification or standard.  Since PAGs should have the option of
soliciting prior art from the public and filing a reexamination request
or declaratory judgment action to invalidate a patent covering a W3C
specification or standard, Sub-Section 5.2 should explicitly include
this option in its charter.  Since Sub-Section 5.2 should explicitly
include this option in its charter, Sub-Section 5.2 should be rewritten
as follows:
    ...  The charter should include:

    * clear goals for the PAG, especially a statement of the question(s)
      the PAG is to answer *and whether or not the PAG should solicit
      prior art from the public to invalidate the blocking patent*;

    * duration (maximum 90 days); ...

Filing a reeamination request or declaratory judgment action to
invalidate a blocking patent will have a deterrent effect on
recalcitrant patent owners.  Patent owners spend at least US$10,000 to
prosecute a national patent application to issue.  Patent owners spend
many times this amount to obtain patents in other nations.  After
obtaining these patents, most patent owners license these patents to
generate additional revenues.  Since most patent owners license these
patents to generate additional revenues they do not want these patents
invalidated.  Since a patent can be invalidated or significantly
narrowed in scope in a reexamination request or declaratory judgment
action, most patent owners prefer to avoid reexamination requests and
declaratory judgment actions.  Since most patent owners prefer to avoid
reexamination requests and declaratory judgment actions, filing a
reeamination request or declaratory judgment action to invalidate a
blocking patent will have a deterrent effect on recalcitrant patent
owners.

Even soliciting prior art from the public will have a deterrent effect
on recalcitrant patent owners.  Prior Art searches to invalidate a
patent are expensive, even more expensive than obtaining a patent.
Since prior art searches are so expensive, most accused infringers
prefer to license a patent rather than litigate a patent infringement
action.  However, by soliciting prior art from the public and posting it
to the PAG's or Working Group's website, the PAG can provide accused
infringers with a comprehensive prior art search for free.  By providing
accused infringers with a comprehensive prior art search for free, the
PAG encourages accused infringers to litigate rather than license a
patent owner's patent.  Since the PAG encourages accused infringers to
litigate rather than license a patent owner's patent, more accused
infringers will litigate a patent owner's patent.  By causing more
accused infringers to litigate a patent owners patent, the PAG increases
the probability that the patent will be invalidated or significantly
narrowed.  By increasing the probability that the patent will be
invalidated or significantly narrowed, the PAG discourages the patent
owner from demanding RAND licensing for its patents.  By discouraging
the patent owner from demanding RAND licensing for its patents,
soliciting prior art from the public has a deterrent effect on patent
owners unwilling to RF license their patents.

Since filing a reexamination request or declaratory judgment action or
even simply soliciting prior art from the public has a deterrent effect
on patent owners unwilling to RF license their patents, this proposed
patent policy should grant PAGs the power to file execute these actions.

    B.  Sub-Section 5.3: PAG Conclusion

I object to parts 3 and 4. because they open the door to allowing patent
owners to RAND license their patents which cover W3C specifications or
standards.  In other words, parts 3. and 4. allow patent owners to slip
RAND licensing into this proposed patent policy.  Since these parts
allow patent owners to slip RAND licensing into this proposed patent
policy, I think the offending clauses in parts 3. and 4. should be
deleted.

The last clause of part 3. should be deleted.  The last clause of 3.
states:

     ... or the terms under which acceptable licensing may be available.

This proposed patent policy only discusses RF licensing.  Since this
proposed patent policy only discusses RF licensing, the only other
"terms under which acceptable licensing may be available" is RAND
licensing.  Since the only other "terms under which acceptable licensing
may be available" is RAND licensing, the last clause of part 3. allows
PAG to recommend and/or adopt RAND licensing for W3C specifications and
standards.  Since the last clause of part 3. allows PAG to recommend
and/or adopt RAND licensing for W3C specifications and standards, it
should be deleted.  In other words, part 3. should be rewritten as
follows:

    3.  The Team should seek further information and evaluation,
including
        but not limited to evaluation of the patents in question.

Part 4. explicitly allows the PAG to recommend and/or adopt RAND
licensing for W3C specifications and standards.  The first sentence of
part 4. states:

    4. The specification under development should be produced on RAND
  (reasonable and non-discriminatory) terms, either at W3C or some
  other body.

Since part 4. explicitly allows the PAG to recommend and/or adopt RAND
licensing for W3C specifications and standards, it should be deleted.

As I stated in 3.(i) of my October 10, 2001, comments (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Oct/1398.html)
and "B. The Patent Policy Working Group is pro-RAND Licensing" of my
November 06, 2001, comments (see
http://lists.w3.org/Archives/Public/www-patentpolicy-comment/2001Nov/0014.html),
RAND Licensing will destroy RF Licensing because Members will always
find an excuse why their patents should be RAND Licensed.  Since RAND
Licensing will destroy RF Licensing because Members will always find an
excuse why their patents should be RAND Licensed, the only way for the
W3C to ensure that Members continue RF Licensing their patents is to
prohibit Members from RAND Licensing their patents.  Since the only way
for the W3C to ensure that Members continue RF Licensing their patents
is to prohibit Members from RAND Licensing their patents, this proposed
patent policy must prohibit RAND Licensing.  Since this proposed patent
policy must prohibit RAND Licensing, part 4. should be deleted.

Sincerely yours,

Daniel E. Maddux
Received on Wednesday, 20 March 2002 16:20:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 April 2010 00:13:49 GMT