So Sir Tim, thank you very much for speaking to us today. I interviewed you in 2006 and we talked about the semantic web. I expected to hear a lot more about it in the media, in the intervening time, but haven’t really. How are things going in that field?
Well, what has happened is that we’ve realised that the data layer of the semantic web is the one that we need to focus on, (for most people), so people in IT really need to use it for sharing the data. Semantic web is basically data integration technology where you don’t just integrate across the company, but you can integrate across the whole organisation and integrate across the planet and not just within a domain, but between different domains.
So we focused on the data and as that layer is really very simple, we’ve been calling it linked data and this is the Linked Open Data Movement and so really linked data is easier to understand. The semantic web has a lot of technologies and it’s got query languages, which of course you need; imagine doing databases without SQL. When you integrate that data you use Sparkle. Our ontology language is quite complicated, but, in fact, people use very lightweight ontologies when they’re surfacing their data.
Just to say, by the way, when we say latitude there and lat here we meant the same thing; that is amazing. What it does, the Linked Data Movement has been picking up on and if you look at the Linked Open Data cloud, and you watch how it’s been going over the years, it is very satisfying.
We noticed, when we were doing research, that you have a Engineering and Physical Sciences Research Council (EPSRC) grant. I think it is called ‘Enacting the Unbounded Data Web’. Is that linked in with that? Could you tell us more about that?
Yes, that’s research to look at situations where we’ve got a bunch of data linked together, to see what we can do with them. This research looks at the second level questions such as how can you make a really good user interface to it? How can you, when different people have resolved and surfaced different things separately, (which is the way to do it by the way)? Then afterwards obviously you think, oh well, where is the best place to sew these sets of terms together, what is it worth and what is the best way of figuring out how to convert data dispersed in that vocabulary into another vocabulary.
In managing a vocabulary socially there are all kinds of really interesting questions, which will arise in the future and will arrive when we have got masses of data. To be responsible you have got to look ahead and search out when all this will happen and when this work is done, look toward the secondary issues.
We posted some comments on Twitter, when we knew we were coming to see you, and got some interesting questions back, plus a couple of odd questions. HTML 5 came up. People want to know how you envisage that changing things as it seems like a sea change in the HTML sphere. What do you see?
In one way, it is not really a sea change at all. HTML has progressed over the years and every now and again it needs new things as a mark-up language, but really the big difference with HTML 5 is that the people who are most excited about new features in it are the people who are using it for whatever web platform.
So, in fact, it is a user interface platform, and the switch from looking at a web page, which is static and has gone through a content management system, to looking at a web page, which is actually a programme and which gives you access to a whole new world out there, that of course is a sea change. So that’s the change of attitude. It’s still HTML, in fact it’s the same language core, which is being used for both web applications; it’s not just HTML spec, it’s lot of web application specs around it that make it into a powerful programming environment.
It seems, when I was reading up about it myself that more of the media approach is being taken into account with the language, like embedding video and audio. Is that to respond to the huge growth in social networking and that kind of thing?
There are all kinds of places where you use video. However, it’s impossible to embed into HTML, but to have a video tag is a clean and easy option and makes it easier to be able to say what controls you want and people do that a lot. I think that really comes from the fact that bandwidth is getting to people’s homes; it’s straightforward because people do use it a lot for social networks and all kinds of sites. Public relations sites use it, when you are buying a car, all kinds of things. Video is ubiquitous in a sense it turns up on all kinds of sites, but remember, of course, there are people who don’t have broadband and just use their mobile devices.
One of the more strange questions that came from Twitter was this one: do you think that artificial life forms may already exist on the internet? A bit of a philosophical one really.
No, I don’t. I think a lot of people have been trying to make things, which have life-like properties. If you want a strict answer to this, you could say anything we produce is alive, so in that case a virus is an artificial life-form.
But, I think what people mean when they talk about whether something has an emergent intelligence or not, is ‘have we got so many computers “talking” together’. Like in the famous Arthur C. Clarke story ‘Dial F for Frankenstein’. He imagines all the traffic lights going red or going green, because the ‘baby’ has woken up and it’s crying and is learning to use its new sense of outward peripherals, it’s figuring out what they do.
I think there is nothing which you’d call a thinking thing, like a mind, out there. However, there are lots and lots of really interesting emergent properties, the whole blogosphere, and all kinds of social systems, which happen because they’re intermediated by machinery on the web. These social machines help people create new democratic systems. For example, they let them do peer review journals or let them do a peer review on each others comments.
I think it is not artificial intelligence really, but there are very powerful systems which do more than an individual person could do; if you like a collective intelligence. They’re always asking the question ‘why can’t we make a group of more than one person smarter than one person?’ I think we need to do a lot more in that area.
I’m glad I asked that question from Twitter. How do you feel about those kinds of tools? You do have your name reserved on Twitter at the moment - Tim Berners-Lee - and the chap’s got your name posted as ‘waiting for the great man to come along.’ What do you feel about that thing?
Twitter, in general, is interesting. Both from the architecture point of view and from the computing point of view of course. Twitter is one of the centralised sites. I actually have timbl on identi.ca, another twittering, microblogging site; it’s actually open for software.
There are 30 odd clones of it around and anybody can start a Twitter site. They’re looking at the question of ‘how, if I’m tweeting on one site do you follow me from a different site?’ The really important thing about the web is that it is de-centralised, that anybody can start a site server. The question is ‘how can we make protocols so that you can follow me on other sites?’ I will probably get hold of, thankyou, whoever is holding ‘timbl’ for me on Twitter. I don’t have ‘timbl’ and that’s what I use and I miss that on Twitter.
What are you currently working on yourself?
There are lots of things happening; I’m director of the World Wide Web Consortium, which is doing more and more stuff everyday. We’ve talked about HTML 5. There’s the push for mobile internet which is still very important of course. There are all kinds of technology and all kinds of semantic web technologies. If you go to the W3C site at www.w3.org you can catch up with what I’m doing.
I do encourage people, whether they’re from light companies or from a background of programming in their garage, to get involved with the W3C, if they’re interested in the standards. I’m involved with that. I’m involved in web science. In the case of the Web Science Research Initiative, we’re now making it into a registered charity, the Website Trust, and that is trying to get people to look at the web as some big system we really need to study.
We’ve talked about ‘what are the emergent phenomena?’, ‘Are there good things?’ ‘Are there bad things?’, ‘Is it stable?’, ‘What properties does it have?’, ‘What properties do we need from it?’, ‘What is happening out there or what could happen?’, ‘What could we do?’, ‘How should we change web protocols or invent new web protocols, so as to make new really interesting and valuable things happen with all the people connected together?’
The web connects to humanity and humanity is connected by the web. The Web Science Trust is trying to promote and coordinate and make websites happen.
The goal of all these things, I suppose, is that the web should serve humanity. In fact that is a very broad goal to serve humanity. And thinking about that goal we realised we need something broader than just the web science.
We now have the Web Foundation, which I’m a founder of; I’m still on the board. It’s now run by Steve Brunt, who used to be CEO. W3C, however, has very much broader scope and way of doing things. In realising that 80 per cent of the world doesn’t use the web at all we had a look at a bit of Africa, had a look at the way people use the internet and are looking at projects to tweak the technology so that it is more appropriate for use in developing countries.
So I’m involved with all those things, plus I’m a full time professor here at MIT and I’m a professor at Southampton. I’ve got quite interesting projects going on at both sides of the Atlantic, collaborating on a good day, and we’ve got students doing all kinds of things; my particular interest, for the past few years, has been the user interface to the semantic web, looking at how there’s a lot of stuff out there and how you can make best use of it all.
Let me just pick up on something you said there about 80 per cent of humanity not having access to the web at the moment. One of the goals of BCS, as the Chartered Institute for IT, is to get society web-enabled internationally. Is the biggest obstacle to this goal the fact that there is still a huge chunk of humanity that still don’t have access to the web or are there other challenges?
The fact that there is a large amount of humanity which does not use the web is a drawback; it’s something that is changing and we’ve got to be aware of what is going to happen, as a very much larger number of people get on board and start producing content in various different languages and for various cultures.
I think there are a lot of other things that we have to think about as well. I think some of the existing things that people are experimenting with across the web - with different types of self-governments, or different types of governance in different organisations - I think it’s exciting, because maybe they will find ways of building organisations which are somehow fairer and more effective.
Maybe they’ll find democracies or meritocracies that work well? We hope to see people working together more respectfully, more internationally, crossing borders and so on. There are a lot of things which are social and technical about the web, which are very important.
You seem to have a very positive view of the way the web enables people, seeing it as generally a strong social tool.
It’s designed to be a neutral social tool. It is designed not to try to constrain what you do with it. What you do with it is up to you and so what humanity does with it is up to humanity. There are of course good and bad uses for any powerful tool, and out there you will find stuff that enthrals you and stuff that you abhor. But, in the end it’s all about humanity and it’s humanity I am optimistic about.
By the way there is one thing I am working on which we didn’t mention, which is particular to this year. 2009 has been the year not just for putting data on the web, but particularly in pushing government data on to the web. So we’ve got this project where the Cabinet Office has put data onto the web in the UK, which is really exciting and there are lots of people in UK government and around the government who get it and they know and are excited about linked data.
As we talk, we are accumulating data sets linking them together. I think that’s very exciting: it’s very nice that there is a mirror initiative, different but also very strong, in the US. The Obama administration has made it mandatory to put government data online and there’s data.gov and recovery.gov, which are places where the government has been transparent about what it’s doing by putting data out there.
I think the value of having government data out there in the end will be huge and people will use it, not just to hold the government accountable, but because it’s valuable data. It’s useful data for industry, so I’m hoping that once government has put data on there industry will realise ‘actually, if we put our stuff on there as data, the whole market will run much better, the whole industry will run more efficiently.’
I want to ask you about the mobile web a little bit more, because obviously there’s a huge demand for web on the move and that’s only going to increase. Are we getting our approach right?
Well, we have got a mobile web initiative at W3C, which has been looking at the question of when you have this huge diversity in devices of people looking at the web on their wrist watches, more or less, and looking at them on screens, which are becoming bigger and bigger. It is very different from the days when people were mislead into putting ‘please adjust, you should be using a 800 x 600 screen to look at this website’ messages on their devices.
Life is not like that; not everyone has the same sort of computer. That’s been the general situation with the web, but now it’s very acute and it’s very important in developing countries where often a mobile is the only device somebody has. If they can’t figure it out on their mobile device there’s no laptop to go to.
The art of how you take the same information and make it available to people on all these different devices, or even to people who are illiterate for example, who need it to be read to them or need to watch a video that is something that we are only starting to understand. To a certain extent the semantic web helps you think at the logical level or conceptual level that this is not a web page about a flight, this is flight BA124.
This is a particular flight and when I bookmark it then I’m bookmarking a flight. Depending on when I de-reference the bookmark I get the details on my laptop; I might get a map as well: when I do it on a huge screen (which are getting cheaper and cheaper) then I may not just get the map, I might get a picture of the plane, I might get a picture of the people on the flight, a list of friends on the flight and so on.
The concept that is important is the idea of the flight itself: that’s where the bookmark helps with that, but it’s early days and there’s lots more work to be done. The mobile initiative is in its second phase, which now continues.
How do you access the mobile web? I imagine you must do. Are you an iPhone man or Blackberry?
I’m still Director of a Consortium, which is neutral, so I don’t generally compare the various devices, but I do enjoy being able to access the web from a little device that I’ve got in my pocket at the moment.
That’s very neutral, very good! I asked about the iPhone because all the apps are so interesting. This new development of having apps on all the different platforms is really quite cool, isn’t it; it’s fun to play with. Do you have time to play?
I do play with things, but remember that what is really important from the point of view of the long-term growth and a healthy market is that you can buy your connectivity from one place then you can buy your phone from another place and buy an app independently from a third place. It is really important that whatever you use as a platform (this is very important to me for my computer) is that I can install, I can buy an app from anywhere, rather than have to go to the phone manufacturer or rather than have to go to the operator who is giving me my connectivity.
I want it to be an open market for connectivity, where I get the best plan for me at the best price. I’m prepared to pay a lot of money for a phone, if it’s a nice one, but I want, on that phone, to be able to write my own app without having to ask any central authority and also have to pay tax.
Interesting, thanks for that. One question I got asked by Twitter is whether bandwidth will keep up with the demand because the more things we use, the bandwidth seems to be eaten up more and more doesn’t it. Will it keep up with demand?
The person who sent you that message used a bit over 140 bits of information to send you that so it’s worth remembering that actually, apart from video, everything else, once you’ve got video and got fibre around the country for video, then everything else is lost in the noise. The idea of actually paying serious money to send an SMS from a phone is ridiculous. If you work out how much it costs to get video and divide the price you’re paying for a month by the number of bits you can get that month, you’ll find the price of bits is very, very small.
The idea that you have to pay to send an SMS internationally is a little bit weird.
There are many, many things which are very exciting and don’t actually use a lot of bandwidth. Those will not use up the bandwidth. Video is always going to be the thing to push it up. You got video, people will want HD, then they’ll want stereoscopic, then they will want 3D, then they will want total immersion. I can see that will go on just pushing and pushing.
It’s about 18 years or so now, isn’t it, since the development of the web as we know it. Are you amazed about where we’ve got; are we behind in some areas or ahead?
I had it all planned really! I’ve always been amazed by the amount of diversity of stuff out there, the things that people think up, the creativity that people show when they get a domain name and a website to play with; so that continues to be wonderful.
I think the way in which I’d hoped the web would go was that it would very much be a collaborative design space. I feel that the tools we have to play around together, doing a design together, working on a political problem together, or just planning a family holiday together, they’re really crude compared to what you can imagine. For example, I should be able to move my photograph about so you will see it moving around in real time. Any document I edit I should be able to share it with you so you can edit back in real time, changing it where it’s a document on the web.
There have been exciting things happening ever since the dawn of the web. There is still a huge amount of challenge out there.
Sir Tim, thank you for speaking to us.