Why Patents in Internet Protocols are Disastrous
By: Con Zymaris

The Internet is a human achievement of epoch-making proportions. It had
been gestating and forming in the hands of technologists (researchers,
engineers and code-hackers) for over a quarter-century before it was
'discovered' by the greater mainstream of industrliased civilisation. In
our role as protective parents of our creation, we technologists
sometimes raise our heads above our monitors to issue warnings on
possible threats to the growth and hegemony of the Internet. 

The Internet is constituted of a myriad network technologies, drawn
along functional and social boundaries, with one of the most popular
being the World-Wide Web. Devised by a researcher in a physics
laboratory around 12 years ago, the Web has experienced phenomenal
growth to become the greatest storehouse of globally produced and
publicly accessible information, opinion and content ever assembled in the
history of mankind.

There are a number of reasons why the Web (and through it, the Internet)
has become the pre-eminent vehicle for the electronic expression and
delivery of human ideas. But in recent days, a proposal has surfaced
from a World Wide Web Corporation (W3C, the body which formulates the
standards on which the Web is founded upon) Working Group, consisting
primarily of functionaries of very large corporations and from legal
firms (in effect a small clique of the membership of the W3C,) which
threatens to undermine the general smooth flow of the Web's wonderful
success.

The Internet works and is a success for two primary reasons. One is
technical, the other sociological. The technical reason is due to the
nature of the process that produced the protocols and file-formats of
the Internet, namely open and publicly available standards, with freely
available reference implementations, often embodied in open source
technologies like the Apache web server. This friction-free approach has
meant that anyone and everyone is able to participate in deploying these
platforms and converse electronically with the rest of the world. It has
also meant that the Internet seeks truly efficient economies, both of
scale and implementation.

This universality leads us to the even more important reason that the Web
and Internet technologies have been the great success they are, the
network-effect, which is a conceptual corollary to Metcalf's Law (Bob
Metcalf was the inventor of Ethernet): the value of a network grows in
proportion to the square of its number of users. This implies that if we
double our user-base of people who can communicate on the Web, we
quadruple the overall usefulness of the system for all parties. Through
open and interoperable standards, the Web has driven Internet growth
from 4 million daily users in the early '90's to perhaps over 400
million now. We'll leave the multiplier-effect calculation to you as an
exercise. It is exactly this attribute which is threatened by the
proposal put forward by the W3C Working Group on patent policy.

In simplified terms, the Working Group has recommended that the W3C
adopt a stance which will see the introduction of patents ratified as
international standards. While on one hand, their approach lends itself
to the clarification by vendors proposing technology standards, and to
their full disclosure and explication, thus likely averting the whole
Rambus SDRAM fiasco, the likely negative impact of this complete
reversal of policy on patents is too great to be supported by the wider
Web community.

While the vendors have always been and are still free (some would argue
unnecessarily so,) to patent the technology they develop, this
technology should not be given official blessing and allowed to become
an industry standard. This approach will lead to a substantial
destabilisation of the international standards-body in question (the
W3C) and the likely balkanisation of the Web. This is due to the fact
that most of the many and varied pieces of the network puzzle which
constitute the Web experience, are made of technology which can never
embody these patents: the Web, as with the rest of the Internet, runs on
open source platforms and tools. By far, most of the web sites, mail
servers, domain name servers, ftp servers, IRC servers (the list is
endless) that you use on a daily basis, are all open source: Apache,
Linux, BSD Unix, Sendmail, MySQL and BIND dominate and enable the whole
Internet. Without these, the Web would not now or would ever have
existed. Likewise, without these being freely available to everyone,
there would be no friction-free, dynamic, cost-effective, innovative
Internet economy.

The response to the patent proposal has thus far been near-universally
negative. A sample of the thousands of submissions to the W3C can be
read at http://lists.w3.org/Archives/Public/www-patentpolicy-comment/
While the issues involved are obviously complex and can only be hinted
at within the context this article, the core requirement of open, freely
available and interoperable standards, un-encumbered by patents, is
crystal clear. If the W3C does adopt the Working Group's proposal and
introduces patent-based technology standards, the bulk of the Web's
platfroms will not be able to, nor will want to, participate in the
ensuing protocol and document electronic interchange. The developers and
maintainers of these open source platforms will band together to form an
alternative to the W3C, to continue the vision of Web access and
interoperability for all. They will be joined by industry heavy-weights
such as Hewlett-Packard, which has renounced the pro-patent move. For
the continued success and health of the Internet, this bifurcation must
be avoided at all costs.

As technologists who have helped, in whatever small way, to create this
most wonderful of human-scale technologies, we should vehemently oppose
the W3C Patent Policy Framework proposal. We want to maintain the shared
vision of a universal collaborative, publishing and communication
medium, available equally to all participants, and one hopes, with the
passage of time, to all mankind.


Con Zymaris, has been the CEO of IT Services firm Cybersource for over
10 years. He can be contacted on conz@cyber.com.au