Columns
Paying for Protocols; RANDom Nonsense at the W3C
- By Mark McFadden
- 10/31/2001
One of the main reasons Web services has emerged as a powerful tool for the enterprise is that it is built on industry standards. Those industry standards, including XML, XSLT, and the family of technologies based on XML, are vendor-neutral and available to every developer. New tools emerge because the standards are available to all – unencumbered by copyright or intellectual property protection.
What if that were to change?
Imagine if the fundamental protocols that make Web services work were suddenly exposed to claims of intellectual property and, in extreme cases, royalty payments for use of certain technologies.
Unimaginable? In fact, it’s exactly what the World Wide Web Consortium has proposed.
The XML family of protocols provides the foundation for Web services. Those standards are developed and promulgated by the W3C. In a stunning and controversial development, the W3C suggested that it would be possible to create standards whose implementation would require royalty payments to patent holders. The new licensing model, called RAND ( for “reasonable and non-discriminatory” ), makes it possible for vendors of new technology to bring their protocols to the W3C for standardization while maintaining protection for their intellectual property – “patents” – and the ability to achieve some reward for licensing the new technology – “royalties.”
The W3C has argued that there is a distinction between fundamental and high-end layers of the Web "stack." The standards body would have us believe that there is a “lower-layer” of the Web infrastructure that will remain royalty free. It is only advanced, or “higher-level” services that would be “appropriate for licensing on reasonable and non-discriminatory (RAND) terms,” according to the W3C’s backgrounder on patent policy.
That’s pure twaddle.
In the first place, standards don’t separate neatly into compartments like “lower-layer” and “higher-layer.” What’s “lower-layer” to one protocol is a high-level application to others. HTTP seems like a “high-level” application to implementers of the Internet Protocol ( IP ). HTTP seems like a “low-level” application to users of tools like SOAP and the Java APIs for XML. Will Web services implementers be subject to royalties on the use of technologies in some arbitrarily decided way?
Even if the Web services arena arbitrarily divided itself into a set of core technologies, crucial for all and royalty free, and newer, niche technologies – it makes you wonder why the W3C would choose to be involved in standards outside the core in the first place.
Fortunately, the W3C has extended the deadline for group working on patents so that it can get further input from the Internet community. If you are deploying Web services you are part of that community. Whether you think patents are inevitable, or you think patented technology and the Internet is an oxymoron, you should be involved in the debate that may change the way standards are built for Web services.
For more information on the W3C's patent policy or a chance to express you thoughts on this issue, visit www.w3.org/2001/ppwg/.
About the Author
Mark McFadden is an Internet infrastructure consultant based in Madison, Wisc. He writes on Internet technologies and standards.