Grinding Noises

Don't Drown the Spirit of Generosity

A response to the W3C Patent Policy Framework Working Draft.

Thank you for the opportunity for public comment on your 16 August 2001 Working Draft. I write as an academic with an interest in the intellectual property issues surrounding the web, and as a web-designer who has confronted some of those issues directly in site development over recent years. (The comments below represent, of course, my personal opinions, and not necessarily those of my employers past or present.)

All of us know that the Web was built on a gift: on work that could have been patented, but was instead given to the world. Without that act of generosity it's doubtful whether the Web would have come as far as it has. I'm concerned that countless small acts of self-interest, incorporated into future W3C Recommendations in the form of patents, could drown out the original spirit of generosity that is so important to so many of us who use and build the Web.

After reading your draft and various third-party responses to it in recent weeks, I recognise that your development of this policy is an attempt to address existing problems in a way that aligns as closely as possible to 'the Web community's longstanding preference for Recommendations that can be implemented on a royalty-free basis'. I also acknowledge that it must be frustrating to receive hundreds of objections at the eleventh hour to a policy that has been arduously developed over many months. Nevertheless, I must add my voice to the list of critics.

My first concern stems from your statement of the 'Larger Role of Patents on the Web Landscape' (2.1). Your list of patent-related trends—convergence, rise in patent issuance, etc.—will be familiar to anyone who has followed these issues in recent years. But the conclusions you have drawn about where these trends will lead—for example, 'the tradition of patenting technology from [the telecommunications, broadcast media and consumer electronics] industries will likely be carried over to the Web'—are debatable, and not at all proven. Patent offices may be issuing patents at record rates, but there is no certainty that they will continue to do so; the very recognition of this fact in the software community has already prompted challenges to their unseemly haste. Non-computing industries, most notably the entertainment industry, have certainly sought to extend their interpretations of intellectual property into the Web, but despite some successes have met opposition at every turn, and could well continue to. The battle is not over yet.

My fear is that by inviting patents into the W3C Recommendation process, even in a heavily circumscribed form intended to prevent last-minute surprise claims by Members acting in bad faith, the W3C could make these prophecies of a litigious, patent-filled future self-fulfilling. The struggles over patents may simply move to the start of the Recommendation drafting process rather than the end; and in the meantime, the W3C will have sent the message that patenting key Web technologies is acceptable. The question of what exactly constitutes 'bad faith' immediately becomes much more vague and contestable.

The alternative is for the W3C to accept that some parties will occasionally act in bad faith and cause headaches in the drafting process, but to send a clear message about such behaviour by rejecting patents as part of W3C Recommendations. The very existence of your 16 August 2001 Working Draft suggests that such an outcome is unlikely, and that patents in some form will become part of future Recommendations. I can't help feeling that if they do, the moral authority of the W3C to insist on royalty-free Recommendations, or even a 'preference' for them, will be weakened—perhaps not immediately, but increasingly over coming years as more and more patents become incorporated into Web standards.

The proposed RAND licensing of patents within Recommendations attempts to address some of these concerns, but raises others. The concept of 'reasonable, non-discriminatory royalties or fees' is a weak link: what appears reasonable to a large corporation can appear wholly unreasonable to a small web-design company or private individual. Given that even the most modestly-priced shareware never attracts payment from all of its users, it is hard to imagine any definition of 'reasonable' that will satisfy the entire web-development community; as soon as royalties become an inescapable part of the web-design landscape, part of that community will be shut out. The threat is perhaps not so much from any single royalty payment as from the prospect of dozens or hundreds that add up to a considerable financial burden, whether paid directly one-by-one or indirectly through higher-priced web-development software or hosting fees.

The Working Draft, again, attempts to limit this potential impact by allowing RAND licences only when it is 'not possible' for Members to adhere to the 'preference' for royalty-free terms. This leaves another door open to those Members who would wish to advance their own interests, as it would appear that they will be the ones who determine whether or not it is 'possible' to contribute royalty-free elements to Recommendations. Is it too suspicious to imagine that many of them will say, more often than not, that it isn't?

My final concern is about our all-too-human inability to predict the future. I have already suggested that this Draft may be an over-reaction to a patent-filled future that may never come to pass. But I'm also concerned that we have no way of predicting with any certainty which of today's Recommendations will become the essential foundations of tomorrow's Web (the 'lower-layer infrastructure [that should] be implementable on an RF basis'), and which will be the 'higher-level services' that 'may have a higher tolerance for RAND terms'. It is quite possible that a Recommendation implemented on a RAND basis in 2001 might be an inescapable part of the Web of 2011, condemning every designer and developer to paying royalties in some form; we have no reliable way of knowing what the next 'killer app' web standard will be.

Today's luxury is tomorrow's necessity. Having just moved to a country that persists with an outdated 1950s policy of charging substantial licence fees for television sets, I would hate to see the Web start heading in the same direction. Our aim should be to remove any barriers to its use, including its design and development—not to encourage those who would erect them.

First published in the www-patentpolicy-comment Mail Archives, 11 October 2001.

11 October 2001
©2001 Rory Ewins