Concerns regarding RAND licensing.

        Having read the W3C Patent Policy Framework working draft of 16 
August, 2001, I am concerned by the addition of RAND licensing terms to the 
W3C patent policy.  The RAND license gives technologies which require royalty 
payments the same status as royalty-free technologies.  In return, the RAND 
license requires that these technologies be available for licensing to all 
implementers, at least to the degree required to implement the W3C 

        The problem with such a licensing agreement is that it may prevent 
implementers with limited resources from using those technologies recommended 
by the W3C.  By allowing "standards" to have a price tag attached, the W3C 
will limit the ability of developers to follow their standards.  Simply put, 
I am concerned that small developers (individuals, small groups, startup 
companies) may be unable to pay the licensing fees, thus stifling development 
by these groups.  How ironic that a license called "non-discriminatory" 
should effectively discriminate against developers based on financial 

        Furthermore, I am concerned that academic developers may run into 
difficulty complying with the terms of the RAND license.  A private developer 
or corporation can sign a contract with a licensor, but an academic developer 
usually must go through their employer, leading to bureaucratic overhead 
which will, again, stifle development.  Please remember that one of the W3C's 
goals is to encourage the evolution of the Web, and hindering small 
developers and academic developers cannot possibly help in achieving this 

        I am more concerned, though, that standards licensed under the RAND 
terms will effectively prevent free software from complying with W3C 
standards.  By free software I mean software which is available for 
redistribution and use without restriction and without charge.  The terms of 
the "RAND" license are vague enough to require licensing fees on a "per use" 
basis, such that a licensing fee will need to be paid for each copy of the 
software which implements these standards, as with Unisys and their patent on 
LZW compression used in GIF and TIFF images.  This requirement would 
certainly eliminate the possibility of releasing W3C standard-compliant 
software without charge.  Even worse, it would prevent the copying of 
software without restriction, even when the developer licenses the software 
in this manner.  If a developer can be charged per copy of their software, 
then a mechanism for tracking the distribution of the software will be 
necessary, further increasing the financial overhead for the developer and 
thus (probably) the cost of the software.  Remember that one of the W3C's 
long-term goals is Universal Access, which will not be helped by increasing 
the cost of the software used to access a W3C-compliant Web.

        In summary, I feel that RAND licensing is not compatible with the 
W3C's long-term goal of Universal Access or with the evolutionary nature of 
the web.  The former particularly concerns me.  A Web which is accessible by 
only some people due to cost is simply not worth having.  The whole point of 
having a worldwide network is to give the maximum number of people access to 
as much information as possible.  This goal should not be sacrificed so that 
someone can make money by turning their patented technology into a W3C 
standard.  Any technology which cannot be used for the benefit of all 
(especially for the sake of money!) has no place being a Web standard.

Marcus Lauer

The brain immediately confronts us with its great complexity. The human 
brain weighs only three to four pounds but contains about 100 billion 
neurons. Although that extraordinary number is of the same order of 
magnitude as the number of stars in the Milky Way, it cannot account 
for the complexity of the brain. The liver probably contains 100 million 
cells, but 1,000 livers do not add up to a rich inner life.
-- Science Magazine, 1992

Received on Sunday, 30 September 2001 14:17:10 UTC