RealTime IT News

Tim Berners-Lee, Director, W3C

Sir Tim Berners-LeePerhaps for some, being credited with inventing the World Wide Web would be more than enough as far as life accomplishments go. But for Sir Tim Berners-Lee, it's merely the beginning of something even more fulfilling.

The man who was named by Time magazine as one of the 100 greatest minds of the 20th century is now busy channeling his energies into the work of the World Wide Web Consortium (W3C), a standards body he founded and now directs. He spends his time hailing the virtues of the Semantic Web and working to create standards that will drive the Web to its full potential.

Moments after delivering a keynote address at the SpeechTek Conference in New York earlier this week, Sir Tim sat down with internetnews.com to discuss the state of the Web browser market, the growth of the Semantic Web and some of the challenges facing the W3C.

Q: Are you pleased with the way the Web browser has evolved since you created the first Web client in 1990?

Oh yes. We've come a long way with Web browsers, but there are still things to do to get it right. When we talk about Web browsers, we have to talk about standards compliance. That's an important issue that we're working hard to address at the W3C.

The other big issue for all users is the question of security. In general, software companies should take more responsibility for security holes, especially in browsers and e-mail clients. There are some straightforward things the industry should be doing right now to fix things, and I don't know why they haven't been done yet.

For instance, we know that not many people are interested in downloading executable code, either from a browser or from e-mails. Generally, when you browse the Web, you don't need to be running executable scripts. People aren't sending each other software programs. You'll agree that it's very rare that you'll be e-mailing a program to a friend or a colleague.

But somehow the software client itself doesn't understand that obvious fact. You could build the client to display pictures, text, documents and not ever execute scripts. The fact that there's no concrete wall between browsing e-mail and installing software is amazing.

You can simply make an e-mail client that doesn't execute attachments and fix that problem right away. That's a straightforward way to protect users. What has happened now is that people have become scared of attachments in general. People don't send me photos because they say I don't like attachments. Wrong. Attachments are fine but I'm not just going to execute any of them.

Also, that operating systems aren't able to distinguish between things which can carry a virus and which don't is a mistake.

Q: Is that an underlying problem with the architecture of the browser?

No, not at all. It's a very straightforward fix for a serious problem, and I'm amazed it hasn't been taken care of. Another thing that makes viruses spread more rapidly is this illegal practice of forging 'from:' addresses in e-mails. I think SPF is something that should be more widely used to thwart e-mail address forgery.

The technology is there to deal with this. SPF can be used right now to distinguish legitimate mail from illegal mail containing viruses before any message data is transmitted. I encourage people to use SPF because it provides the technology that works.

Q: Let's talk about the Semantic Web. How far away is large-scale adoption of that "giant database" concept?

We're coming into phase two. It's an exciting phase but we still have a long way to go. We have the foundation in place with the approval of RDF [Resource Description Framework] and OWL [Web Ontology Language]. In this phase, we can build up and out from those foundations.

In practical terms, it has reached a certain level of maturity. At SpeechTek here, there are a few people discussing the connection of speech to the Semantic Web, and that's always exciting. There are some students independently at MIT doing some work and sparking a lot of discussion about the connection. There are a lot of programs coming out connecting a lot of data and a lot of ontologies.

I suppose it's a lot like where we were in 1992 and 1993. Back then, the Web wasn't stable, but we knew it was there and it held a lot of promise. We knew it would grow and mature, but there were a lot of things that we needed but didn't have. This was pre-Google. Around 1991, you would go on the Web to look for something that wasn't there. Today, that information is there and we can find it easily.

So, I think that's where we are with the Semantic Web. We know it will mature, but we're not quite there yet.

The excitement that it continues to generate is encouraging. The military needs it; the health sector needs it. There's already an academic field around it. We have RDF and OWL as W3C recommendations, which are big pluses. To that extent, the Semantic Web has already reached a certain level of maturity.

Q: At this conference, IBM announced plans to donate software for speech-enabled applications to the open source community. Can the open source community play a key role in driving innovation?

I'm not exactly sure what IBM will be releasing, but I think that that's a very positive thing. The industry is at a stage where everyone is counting on improvements. Technology innovation is starting to explode, and having open-source material out there really helps this explosion. You get students and researchers involved, and you get people coming through and building start-ups based on open source products.

It means that people in government, for example, are experimenting with open source. That's exactly what happened with the Web, pre-servers and pre-browsers. People started using it without having to go through a single person or company, and it took off. So there are a lot of ways in which open source can be used for education, for prototyping. It just spurs the market in many ways.

Q: In your speech here, you discussed the problem of big companies taking control of standards. What can the industry do to avoid that?

The industry needs to start thinking long-term. Start thinking about what happens five years out, not six months. That means coming to a conference and thinking about where your company is going, and what will be important down the road.

The big buyers can play a significant role here. A lot of people are now realizing that their data will live longer than their software. A lot of companies are also subject to pressure from big buyers, and that does help the cause. Large buyers have a lot of clout, and they don't like to be beholden to a particular company.

Q: What's the biggest challenge facing the W3C today as it works to develop these interoperable technologies?

I'll give you three things that we always have to deal with. One is the tension over patents and proprietary standards. We're trying to get the standards adopted despite the fact that big companies are doing their own thing.

Another big problem is the question of patents and royalties. We've made some huge strides, but it's still something that may loom over us going forward. It's encouraging that a lot of large companies are realizing that they have to allow royalty-free use of patents to take the industry forward.

Q: But companies are still boasting about being granted thousands of patents a year.

Yes, that's true. But when it comes to joining a standards group, they know they'll be expected to allow the use of those patents, royalty free. They've made that commitment, even without investigating what patents they have.

We're at a stage where the companies are getting the patents but saying "we're not going to charge you for implementing this particular patent in a standard." Some of these are huge companies that we never expected would make that commitment. They were always staunch in their advocacy of patents and royalties, but they are now realizing that going the royalty-free route will create even bigger markets.

That realization, I think, has rippled through the industry and, with that, we've been able to create the W3C Patent Policy.

Web Services also present a challenge for us. This is an area that's quite big, and we have a working group at the W3C to develop the technologies that will lead Web services to their full potential.

Q: When you look at emerging technologies, is there something that stands out and excites you?

Yes, one area to watch is the mobile Web. I'm seeing a lot of initiatives and energy in this space. The same energy you see here at this [SpeechTek] conference with the voice sector, you'll see that energy replicated in the mobile space. This will be a big area. Over the next six months, you should watch this space for some exciting things.

But, we're seeing the same problems with the big companies pushing for something that will eventually fragment the sector. For instance, this push for a .MOBI domain. Why did these people go out and push for a new domain name? That makes no sense. They say it's because they want a strong mobile Web, but what does that mean? A strong mobile Web is about device independence. It's about letting users browse the Web from any phone and from any device without limitations. How does .MOBI offer that?

If you were to isolate all the mobile stuff under one domain, it would die. It won't be the real Web. In fact, what we suggest is to take all the energy, and this desire to make the mobile Web accessible, and pour it into the work at the W3C. We have already done a lot of work around Device Independence.

What we really need to do is show how to use the existing technology of the mobile Web to allow access to a unified Web from any device, in any context, by anyone. I'd like to see us work towards some really strong interoperability to say, "look, let's make phones that meet these standards. Let's make Web sites that meets these standards."

You don't do that by creating a .MOBI domain. You do it by working towards standards compliance.