W3C home > Mailing lists > Public > www-patentpolicy-comment@w3.org > October 2001

Electronic Frontier Foundation comments

From: Seth David Schoen <schoen@eff.org>
Date: Thu, 11 Oct 2001 20:55:54 -0700
To: www-patentpolicy-comment@w3.org
Message-ID: <20011011205554.H21625@zork.net>
Dear W3C Patent Policy Working Group Members:

The Electronic Frontier Foundation (EFF), the leading civil liberties
organization working to protect rights in the digital world, submits the
following comments on the PPWG's draft patent policy.

In general, the draft policy of August 16 makes progress in addressing
the thorny patent issues standards groups may encounter.  We join
other commentators, for example, in supporting the proposed Disclosure
Obligations in Section 7 of the draft.

We focus our attention on the most controversial provision, Section 5.2,
which creates a RAND ("reasonable-and-non-discriminatory") licensing
mode for W3C Working Groups.  Adopting this policy would mean that, for
the first time, W3C would have a formal mechanism for promoting some
patent-encumbered web standards -- with the knowledge that these
standards could not be implemented by everybody.

As WWW inventor Tim Berners-Lee observes in _Weaving the Web_,
"patents ... are a great stumbling block for Web development. ...
Small companies may be terrified to enter the business [in the face
of patent claims]."  Because of its harmful effects on smaller
organizations, and the resulting risks to openness and interoperability
on the web, we urge W3C to reconsider its support for a RAND licensing
mode.

The draft policy notes that

   participants in a standards body will be unwilling and unable to work
   collaboratively if, at the end of the process, the jointly-developed
   standard can only be implemented by meeting licensing terms that are
   unduly burdensome, unknown at the beginning or even the end of the
   design process, or considered unreasonable.

This uncertainty is a significant risk to standards development, but
participants are not the only beneficiaries of the process (nor the
only parties whose support is called for).  Where a standards body
undertakes to develop public standards for general use -- clearly
the aim of W3C standards work -- the larger community of prospective
users and implementors also has a deep interest in standards' licensing
terms.  As the policy continues, this community has a "longstanding
preference for Recommendations that can be implemented on a royalty-free
(RF) basis".

This "preference" must not be treated lightly; it has been essential
to the success of the World Wide Web and the Internet as a whole, and
one of the key features setting the Web apart from closed, proprietary
content-delivery systems.  Royalty-free web standards have provided the
raw material for an explosion of creativity and the development of
diverse but interoperable implementations.

For many members of the web community, the RF licensing tradition is
not merely a "preference", but a requirement.  Royalty-based technology
licensing, whether "discriminatory" or "non-discriminatory", grew up
amidst large commercial players, who could typically afford a sizable
licensing fee, accepting it as a cost of doing business.  As you know,
the World Wide Web community is much more diversified.  It includes
tiny startups, multinational corporations, individuals, non-profit
organizations, consortia, libraries and archives, among other kinds of
entities.  Many of these participants are ill-equipped to cope with the
one-size-fits-all world of RAND licensing, and have very different
notions of what is "reasonable" or even "non-discriminatory".

Much of the software which runs today's web is open source, like the
W3C's own reference implementations.  The world's most popular HTTP
server package, Apache, is a leading example; W3C's own web site is
using it, as is EFF's.  But although a flat royalty structure might
seem perfectly "reasonable" to a large corporation, the Apache
Software Foundation -- and Apache licensees -- might well see things
otherwise.  Prospective implementors are all different, but when any
implementor is left behind by a patent licensing system, everyone suffers.

The draft policy attempts to distinguish between high-level and low-level
web standards, in a largely informal way.  Section 2.2, reporting on
consensus within the Patent Policy Working Group, draws this distinction:

   [I]t is especially important that the Recommendations covering
   lower-layer infrastructure be implementable on an RF basis.
   Recommendations addressing higher-level services toward the
   application layer may have a higher tolerance for RAND terms.

We agree that openness of infrastructure is particularly important.
However, the distinction between infrastructure and higher-level
services does not seem to be clearly drawn (nor does the policy appear
to implement this consensus view in any specific way, e.g. by
categorically forbidding the RAND licensing mode for certain Working
Groups deemed "architectural").  Experience has shown that this
distinction can be unstable; services once optional often become
indispensable.  We cannot stress enough that services originally
conceived of as applications may eventually -- even rapidly -- come
to be seen as infrastructural.  For example, HTTP is often used as an
example of an extremely high-level network protocol, yet it serves an
infrastructural role, in turn, for other protocols like SOAP.

We recognize that W3C cannot guarantee that none of its Recommendations
will ever be encumbered by patent claims.  W3C has no control over
third party patent holders who are not W3C members, and there is no
way to be absolutely certain that an encumbrance will not appear
after a Recommendation has been issued or even implemented.  (A
troubling example is BT's hyperlink patent, which was definitely not
foreseen as a risk to implementors of WWW user agents.)  However,
this does not mean that W3C should allow its members to use the W3C
Recommendation process to knowingly promote encumbered technologies
as public standards!

W3C does have the ability to decline to endorse a standard where it
is already aware of licensing problems (e.g. through the proposed
disclosure requirements).  It seems that the community strongly
expects W3C to use that ability, and to preserve the existing RF
tradition in the eventual W3C Patent Policy.

EFF thanks W3C for extending the comment period and for the opportunity
to comment on this draft.  Please do not hesitate to contact us for
any further information or clarification.


Sincerely,

Seth Schoen
EFF Staff Technologist

-- 
Seth Schoen
Staff Technologist                                schoen@eff.org
Electronic Frontier Foundation                    http://www.eff.org/
454 Shotwell Street, San Francisco, CA  94110     1 415 436 9333 x107
Received on Thursday, 11 October 2001 23:55:56 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 April 2010 00:13:42 GMT