W3C founder says steer clear of vendor-controlled groups that masquerade as standards bodies.
The World Wide Web Consortium is a driving force behind the Webs interoperability and evolution. Founded in 1994 by Web inventor Tim Berners-Lee, the group has 450 member organizations worldwide. Since its inception, the W3C has developed standards such as XML and P3P. eWEEK Labs Senior Writer Anne Chen recently spoke with Berners-Lee, in Cambridge, Mass., about the changing role of standards bodies and how enterprise IT organizations should participate in the process.
What are the biggest issues facing standards bodies today?
Intellectual property rights are a much bigger issue [today]. Theres a fear of patents. ... Companies are realizing they have to formally tackle the issues of making standards royalty-free. A lot of [people who] participated in standards activities did so separately from the part of [their company] that has traditionally made money charging for things. These two sections are now meeting for the first time, and, of course, its been difficult. They are dealing with the crown jewels and deciding whether the long-term or the short-term future is what the company values the most.
In which direction are most companies moving on royalties?
Theres a general global shift toward the realization that royalty-free standards are the only standards that can support Internet technology. There is a lot of fussing around, and companies are dealing with it in different ways. Some are dragging their heels; some are rushing on, carrying the banner for royalty-free. In general, though, the shift over the last two years is definitely toward royalty-free standards.
What advice can you offer enterprise IT organizations that want to get involved in standards work?
Id say beware of organizations that might look like a standards body but are controlled by a vendor. Very often, theyre more or less set up as a marketing and branding exercise produced by a group of companies. Do not be misled. You do not want to be railroaded by someone elses company. That tends to slow things down.
Look at how a [standards] organization manages the idea of being open, of being fair, and look for speed but also coordination. One of the things about joining a standards body is that youre not alone. The W3C involves the coordination of various working groups and the mutual review of specifications. If youre doing a technology, you can review one anothers specs and make sure things are going to work. Something weve had to agree on as a consortium group is that were not just working with our own groups but also with those from other standards bodies. You want to make sure people are collaborating, that there are people working together. Are members coming to the table and wishing to share and to build a new market? Are they excited about whats happening, or are they trying to exclude other people? Find an inclusive group where companies get along with the open-source community.
Lastly, when youre looking at standard bodies, make sure theres quality. At the W3C, theres a candidate recommendation, where we spend time making sure people can implement a spec. We have this time where the working group says, "Work on the spec is tabled, and now you should implement it." If youre a company in the business looking to make a product ... you should be interested in a spec ... when it gets to that candidate recommendation point.
In the W3C, whats to keep vendors from pushing their own technologies through as standards?
We have a review before we start an activity so that a company cant start a standard around their own product. You cant even have three members decide theres going to be a W3C activity. Three of five self-selected people might be able to do work together, but all  members have to have a look at proposed work. ... Theres no secret in how that work gets done. All of the members have an equal opportunity to speak up when it comes time to say the W3C may assign resources to this new work.
Are there standards that were particularly problematic in getting done?
P3P [Platform for Privacy Preferences Project] is a classic example because it was the first to be ambushed by a patent problem. At that point, someone had the audacity to charge a patent infringement, and there was a two-year hiatus on work. For 18 months ... there was no work done on the standard. Two years taken out of anybodys business model was really terrible. But a lot of patent policy work grew from that experience. So it was a painful working experience but a learning experience. Now, P3P is a W3C recommendation with implementations worldwide, from government agencies to the European Commission. Its truly a technology that ended up serving the needs of international communities.
There are organizations that work on XML, but the XML Consortium is
the W3C. We did the work on building that infrastructure and the family of technologies. That really has a lot to do with being in the right place at the right time, which cannot be understated.
IT enterprise managers can trust results from the W3C because of the way weve done our work and because of the mistakes weve made. Its not that we were perfect from Day One. Weve had a lot of lessons, but we have a history of eight years of results. We have almost 50 recommendations, most of which are enjoying solid implementations. Allowing everyone to play puts us in an environment that faces challenges within the industry but provides all with tremendous opportunity.