Copyright 1999 Nikkei Business Publications,Inc. all rights reserved, permission is hearby
given for reproduction, as long as attribution is given and this notice is included.
Owning ideas
By: Scott Bradner
Patents can be good
things. People involved in standards organizations can find these words hard to
say. All too often patents are seen as inhibitors of innovation since they can
be used to severely restrict efforts to standardize new technology. There have
been horror stories of companies using "submarine patents" to
hoodwink a standards organization into supporting a particular technology only
to have the company announce the magical appearance of a patent that will need
to be licensed to implement the standard. In one particularly egregious case in
the US a company was forced by the Federal Trade Commission to surrender the
rights to enforce a patent that had not been disclosed during the process of
developing a standard.
In general patents are
becoming an increasingly difficult issue for standards organizations and the
IETF is no exception. Any standardization of patented technology is seen by
their competitors as giving an unfair advantage to
a patent holder. But sometimes there is no apparent work-around that can avoid
the patent so, if the technology described in the standard is important enough
to the standards organization, it must proceed with developing the standard in
spite of the patent.
Another problem occurs
when someone claims that a patent of theirs, or worse a patent they have
applied for, covers a standards work in progress. In almost all cases the
development work on the standard is delayed and in some cases derailed. This is
the case even if the claim proves to be false. False or exaggerated claims of
the applicability of pending patents can easily become a denial of service
attack on the standards process itself and could be used to manipulate a
standards process.
Most standards bodies do
have a requirement that the holder of a patent which legitimately impacts the ability of a developer to implement a
standard be required to agree to license use of the patent at "fair and
reasonable terms". But the reasonableness and fairness of licensing terms
can be very much in the eye of the beholder, this is a case where one size
definitely does not fit all. In these cases there is pressure on the standards
body to evaluate the licensing terms for that technology and the process of
which can become very contentious since the technical people who attend
standards meetings are generally not the best people to evaluate the fairness
of licensing terms. The IETF requests a statement that fair and reasonable
terms will be available but does not require one nor does the IETF require that
the terms be made public. The IETF used to require both of these but those
requirements proved to be unworkable and have been dropped from the current
version of the standards process document.
A worse problem occurs
when the owner of the patent has not been involved in the development of the
standard and announces sometime later the existence of the patent. This can
sometimes happen years later. When this does happen the standards organization
can quickly find itself in the untenable position of supporting a standard
which may now require the licensing of technology which has unfair or
exorbitant license requirements, or the owner of the patent may feel that they
are better off not offering licensing at all. Some standards bodies have been
discussing ways to withdraw standards where these issues come up but few have
reached any definite process procedures to deal with the situation.
In an effort to avoid
the patent issues some people involved in standards organizations have
suggested that no standards be approved which involve patented technology.
Unfortunately this strategy only deals with a subset of the problem since the
patent holder can easily have no knowledge of the standards activity or not be
involved in it. The fact of the patent may not be known until long after the
standard has been approved. Some particularly naive people have suggested that
companies involved in a standards organization agree that once a standard has
been approved all patent rights to any required technology would be required to
be relinquished, at least in regards to implementing the standard. There is
little chance that many companies would agree to such a blanket requirement,
and clearly companies uninvolved in the process would not be required to
relinquish their rights since they were not part of the standards organization
and had not agreed in advance.
The new version of the
IETF standards process addresses the problem of standards which may be impacted
by patents by employing the already existing IETF 3 stage standards track. All
IETF standards are first published as Proposed Standards, then where there are
multiple interoperable implementations of software to implement the Proposed
Standard it can be considered for advancement to Draft Standard status. A Draft
Standard can be considered for advancement to Internet Standard status when the
Draft Standard has shown significant marketplace acceptance. The IETF standards
process assumes that technologies which have patents that must be licensed will
have been shown to have reasonably fair licensing terms if they can demonstrate
the multiple implementations required for Draft Standard status or the
marketplace acceptance required for Internet Standard status. Any significant
unreasonableness would have likely prevented the technology from meeting these
requirements.
A related problem is one
of disclosure. The IETF's rules say that people involved in a standards process
should disclose any pending or actual patents which impact the standard under
consideration. Clearly there are cases where such disclosure can run counter to
company rules. In this case the individual must absent themselves from all
discussion on the proposed technology.
A problem that a number
of standards bodies have already experienced is people purposefully not disclosing
the IPR issues of some technology in the hope that a standard will be adopted
which will require the licensing of their technology thus ensuring a captive
audience. This is known as a "submarine patent". Some standards
organizations attempt to deal with the situation by requiring that participants
agree in writing to disclose IPR information or to withdraw from all
discussions on the proposal. This has proven to be an incomplete answer since
someone else at the same company can, with out the participant's knowledge,
apply for the patent. This is also a problem when the standards organization,
like the IETF, does not have a fixed membership.
On the whole patents can
confuse the standards process but the positive effects of ensuring that
inventors continue to have strong incentives to keep inventing generally
outweigh the problems they create.