-- 作者：admin
-- 发布时间：9/2/2006 10:33:00 PM
--
developerWorks Interviews: Tim Berners-Lee
Originator of the Web and director of the World Wide Web Consortium talks about where we've come, and about the
challenges and opportunities ahead
recorded 7-28-2006 LANINGHAM: Welcome to the developerWorks Interviews podcast series. I'm Scott Laningham, developerWorks podcast
editor, and we're IBM's technical resource for developers, offering a wide range of tools, code, and education on IBM
products and open standards technology.
Today, our guest is Tim Berners-Lee. I'm reading from the cover flap of the Harper edition of his book,
"Weaving the Web, here. It says, "Tim Berners-Lee, the inventor of the World Wide Web, has been hailed by Time magazine
as one of the 100 greatest minds of this century. His creation has already changed the way people do business, entertain
themselves, exchange ideas, and socialize with one another. With new online businesses and communities forming every
day, the full impact of Berners-Lee's grand scheme has yet to be fully known."
Tim Berners Lee is currently the director of the World Wide Web Consortium, the coordinating body for
Web development, and he occupies the 3COM founder's chair at the M.I.T. laboratory for computer science and artificial
intelligence. He's a recipient of numerous awards, more than can be mentioned here, including being named a member of
the Royal Society in 2001. For some perspective, Sir Isaac Newton was the first member of the Royal Society. Tim is
currently based in Cambridge, Massachusetts. Tim, thanks so much for coming on the podcast today. It's a real honor to
talk with you.
BERNERS-LEE: My pleasure. LANINGHAM: I wanted to ask you at the top here, with the amazing perspective you have on the last 25 years of
technology, especially that which has grown up around and on top of the Web, do you have a sense of how you would place
it all historically? BERNERS-LEE: I think, Scott, I'll answer that one in a thousand years' time, if that's okay. LANINGHAM: [LAUGHTER] sure. BERNERS-LEE: That's very difficult... I think trying to write the history, at the time I think it's very difficult. Certainly I know during the early days of
the Web, for example, we were watching its exponential growth and during a phase when the load on our server was
actually going up by more or less exactly a factor of 10 every year. And we didn't know whether we were going to be looking at history or not, because when you look at a curve like that, an
exponential curve going up can often just tilt over and crash back down again. And there were a lot of other projects
we've all had which have done that. So to try to say what the impact on the next few centuries is going to be at this stage I think is really foolish. It's
fun to try, but I think there's also so many new things coming up, a lot of them based on the Web, which I think are
going to change things again completely. LANINGHAM: Right. I hear what you're saying. I was just thinking that, without your creative and catalyzing work
that spawned the growth of the World Wide Web, we might not be talking about the flattening of the world right now, you
kow that phrase everybody is using. BERNERS-LEE: Yes, certainly the fact that we're all connected, the fact that we've got this information space, does
change the parameters. It changes the way people live and work. It changes things for good and for bad. But I think in general it's clear that most bad things come from misunderstanding, and communication is generally the
way to resolve misunderstandings, and the Web's a form of communications, so it generally should be good, but I think
also we have to watch whether we preserve the stability of the world like we don't want to watch this phenomena like the
stock market becoming unstable when it became computerized, for example. We need to look at the whole society and think, are we actually thinking about what we're doing as we go forward, and
are we preserving the really important values that we have in society? Are we keeping it democratic, and open, and so
on? LANINGHAM: That was another question I was going to ask, is are there some negative things that have been unleashed
that really demand an attention level from us that's different from things in the past, and I think you've started to
speak to that right there. Are there other issues like that surrounding this technology that we need to really be alert
to? BERNERS-LEE: There have been all kinds of things which have...I suppose which have worried us and some of it still
do. I think if you go to a person in the street, in fact, the consortium staff once went out into the streets of
Cambridge and just interviewed people to see what they thought about the Internet randomly, and I suppose you had to
pick one thing, they said, spam, get rid of spam.
[LAUGHTER] And that in fact for a lot of people makes the e-mail unusable, and it's a ridiculous amount of humanity's time which is
being spent just leafing through it and throwing it out. So there are certainly some things which need really urgent attention in the Internet space. Another thing that
happened I suppose during the consortiums, this was the evolution of our process for creating the standards, it started
off fairly ad hoc, much more in the start of the Internet engineering task force, people get together, they know each
other, and all with a passion for the same thing, get together and produce some technology and roll it out and then sort
of create a new market. Then we found with one particular standard that came through the consortium, that one of the people in the working group
turned around and said, oh, right, you're all going to be paying me royalties, by the way, when we're finished. And
there was shock and horror. This was during the dot com boom, so this was when things were moving very fast, people
were talking about 2.6 Internet years being a year, or was it an Internet year being 2.6 months, things going very fast. And the whole process stopped for 18 months. It cost us I think $150,000 to find a lawyer who would investigate it and
write an opinion. And the opinion was that actually these people would not be paying anybody royalties, that a patent
did not apply in that case. But that sort of sent a shudder through the community, yes, we can all work together producing something. And then
somebody could come along and claim that actually they owned the idea. So that started the consortium on a path of producing a patent policy so that we could make sure that when people come
together to make standards they come together understanding that they're trying to produce something which is going to
be royalty-free. So it will be generally usable by everybody. It took several years to actually get that policy agreed, but now it is. So that's the sort of, that was something we
may not have anticipated at the beginning, just because the ethos was so much of...where the openness was just, the
default, it was the understood thought of the whole Internet community at the beginning. LANINGHAM: Do you feel like, I've read some of your blog entries about the Net neutrality issue. And do you feel
like that subject right now is the same kind of challenge or even more of a challenge than the one you just described in
the earlier days of the Web? BERNERS-LEE: I think the importance in that neutrality is well understood by a lot of people, and I think for example
outside the United States I haven't heard any concern about it. If you go to Europe and ask somebody about whether
they're concerned about Net neutrality, they say, what? You know, people didn't know the phrase. It's only that in the US there were some speeches by executives of large telecommunications companies in which they
suggested that maybe they should change the whole way they charge for the Internet and block people like Google, so that
Google didn't get connection to their customers, quote, for free, unquote. As though Google doesn't pay for its
Internet connectivity. So I think what happened was that put a scare through. And when you look at...when you look at American industry, you
know that companies are beholden to their shareholders to try to make whatever money they can. And so where they can
find a business plan which they think will produce larger profits in the short term than they may feel obliged to try to
pursue that, even though it might be very completely destructive, it would be destructive of the Internet community, the
Internet markets and the amazing phenomenon which is built on top of the Internet at the moment. So I think in fact Congress will understand the concern. There's a huge education process going on at the moment that a
lot of people are taking the trouble to explain to American congress what it means, this important concept that if I've
connected to the Internet, and if you've connected to the Internet we can then communicate. Nobody else can then start
suddenly charging us extra money to building up because they feel that you're making too much money selling me your
audio podcast or something. I think it's really important that the market for Internet connectivity and the market for content are independent
markets. It's thus why the Internet has expanded so quickly, because we've had one group of people working on
innovations to the Internet infrastructure, and it's more or less since we started, you know, it's gone up, we're
talking about, we started off by talking about thousands of bits per second and now we're talking about millions of bits
per second, so that's a factor of a thousand increase in speed, which has happened during the time I suppose that the
Web was being developed over the last 15 years. LANINGHAM: Considering that arc of evolution that you're alluding to there a little bit, if we go back, if I
understand this bit of history correctly, and around 1980 when you created Enquire utilizing concept of hypertext, and
the modern Internet was still unfolding with the first TCP/IP Wide Area Network, was still a few years away, did you
have a sort of epiphany with Enquire and that experience along the lines of, wow, I've stumbled on to something here
that's far bigger than what can yet be seen or what I think might become of it, but I can tell that this is a sea change
moment? BERNERS-LEE: I don't believe in the sort of eureka moment idea. I think it's a myth. I'm very suspicious that
actually Archimedes had been thinking about that problem for a long time. [LAUGHTER] And it wasn't that suddenly it came to him. Maybe at some point he felt the pieces had fallen into place. You know,
there must have been a point, yes, okay, when he thought, okay, we can lower the crown of the water, and maybe it
happened when he was in his bath. But in general, he'd probably, that was backed up probably by an awful lot of
thinking about density and volumes and physical structures, which he'd been doing for a long time. And so Enquire, for example, Enquire was a program which allowed you to make random associations between different
things. They were more like note cards. And it was all using much older technology of course, it was using sort of 24
by 80 green terminals. And it was a step that I'd had a long term interest in the idea that the brain can store random associations between
things. You can store a random association between a place, between what you're doing at the moment, for example, if
somebody walks in with a particularly flavored cup of coffee, then later on when you remember that flavor you remember
what you're doing just now, it will make association between flavors, and running your equipment and conducting a
interview, totally otherwise unconnected things. So computer systems couldn't do that. And so that was a general interest, which is, and that's related to the ability
to make, link anything to anything, which is a fundamental part of I suppose of the Web architecture, and then now of
the Semantic Web architecture. But it was no one time when it was just struck with, aha! This will revolutionize the world. You know? I don't think
that happened. LANINGHAM: I've read some places where Enquire has been described in the words sort of like a wiki. Is that
accurate, and what do you think about wikis, and is that just another way of doing something conceptually that was
already being done? BERNERS-LEE: Well, there had always been systems where a lot of...systems I'd say you could roughly divide into two
categories: those which were, where people created information systems and created...[in one level] people created it
and the other people browsed it, like sort of CD ROM hypertext based systems. And systems where you are interacting with
the data, things like a calendar. Where when you look at your calendar, at any point you're browsing through the
calendar, you can put any event in there. Your address book, you can create a new address, or if you see that
somebody's telephone number is wrong, you can go fix it. And Enquire was [several nodes], and any point you could be reading something about something which was documented in
the system, you could press a single key to say, and say link, i want to link this to something else, or I want to add a
new link, add a new data. And wherever you added new data to it you had to make a link to some previous data. You could
only add new data by linking it on to something old. So you were actually creating this linked data structure. And the original World Wide Web browser of course was also an editor. I never imagined that anybody would want to write
in anchor brackets. We'd had WYSIWYG editors for a long time. So my function was that everybody would be able to edit
in this space, or different people would have access rights to different spaces. But I really wanted it to be a
collaborative authoring tool. And for some reason it didn't really take off that way. And we could discuss for ages why it didn't. You know, there
were browser editors, maybe the HTML got too complicated for a browser just to be easy. But I've always felt frustrated that most people don't...didn't have write access. And wikis and blogs are two areas
where suddenly two sort of genres of online information suddenly allow people to edit, and they're very widely picked
up, and people are very excited about them. And I think that really for me reinforces the idea that people need to be creative. They want to be able to record what
they think. They want to be able to, if they see something wrong go and fix it. LANINGHAM: Our core audience on developerWorks is obviously software developers. And I wonder if you have thoughts
for that audience regarding the challenges and opportunities around the Web going forward. BERNERS-LEE: Well, my passion now is the Semantic Web, of course. And we just, I've got a bunch of undergraduate
students at M.I.T. who have been hacking away on a little tabulator program I put together, which is an Ajax program.
It's Javascript, it's got an audience library in it, that you can use it to pick up any Semantic Web data, and allow
people to browse around through that data, and then analyze it, and map it, and put it on to timelines, and calendars,
and maps, and things like that. And this form of programming where you're programming at the RDF level, I think is going to be the future, because at
the moment when you program systems, databases, when you write that sequel you have to be aware of the structure of the
database. And the structure of that database is really got a lot of arbitrary design constraints in it, arbitrary design
decisions. When somebody designed that database they didn't have tables like this, so they moved from a model of
abstract things they wanted to represent into some physical database structure. And when you program this in SQL you
have to be aware of that. And similarly, at an object oriented system you have to be aware of what slots are in each object, and in an XML system
you have to be aware of whether we've got an XML file which has got a set of books and in each book there's a shelf, or
it's a set of elements for shelves and each element, the shelf has got an element for each book. When you write your X
path queries you have to be aware of which way that was done, but that decision is kind of arbitrary. The nice thing about programming at the RDF level is that you can just say, I'll ask for all the books. You can ask for
all the shelves. You can ask for a given shelf whether a book was on it. And you're not worrying so much about the
underlying syntax. So I think that we'll be able to write code faster. We'll be able to write application code which
doesn't depend on exactly what sort of XML schema you're using, so that when you find you need to change the schemas you
don't have to rewrite everything. And the fact that the SPARQL query language has come out now means that all kinds of data can be surfaced as RDF. So I
think the exciting thing now is to take existing data sets, whatever they are, look at them and keep the existing
systems running but write little SPARQL servers for them just as in the early days with the Web, we kept the existing
documentation systems, ran little HTTP servers for them so that the data was served. You can write it. Just as then we
wrote little Perl scripts and things to convert the data into HTML, now we can just do the same thing, write these
little PHP scripts, or Perl scripts, or whatever is your favorite language, Python say, to convert the data when the
request comes in for [UI] convert the data, send out some data in RDF, giving data about that particular URI even if
it's a URI for an abstract, some abstract concept. LANINGHAM: Do you get excited at all about emerging Internet technologies as described, you know, Ajax and things
like that? Or are you thinking most of the time at such a conceptual level and a solution level that you don't really
feel like you have time to get too worked up over those things? BERNERS-LEE: Oh, this tabulator project, the RDF tabulator, I coded it up in Ajax. So yes, I'm very...I find that
Ajax is the.... It's the one computing platform where everybody who has a Web browser has got that platform. So the
nice thing about it is when you do code up an Ajax implementation then other people can take it and play with it. And I
think that's one of the things that's so exciting. Also it's actually got a really powerful, you've got a really powerful user interface toolkit with the HTML DOM and also
with the SVG DOM, so I think scalable back to graphics, Ajax applications are going to be really fun in the future. And then when you've got an RDF and SPARQL library which gives you access to the Web of data, then that gives you
extremely deep programmatic access both to the user and to the data underneath so that application becomes something
relevant. So yes, I get excited about things which allow people to have fun and make progress and prototype new ideas. LANINGHAM: You know, with Web 2.0, a common explanation out there is Web 1.0 was about connecting computers and
making information available; and Web 2 is about connecting people and facilitating new kinds of collaboration. Is that
how you see Web 2.0? BERNERS-LEE: Totally not. Web 1.0 was all about connecting people. It was an interactive space, and I think Web 2.0
is of course a piece of jargon, nobody even knows what it means. If Web 2.0 for you is blogs and wikis, then that is
people to people. But that was what the Web was supposed to be all along. And in fact, you know, this Web 2.0, quote, it means using the standards which have been produced by all these people
working on Web 1.0. It means using the document object model, it means for HTML and SVG and so on, it's using HTTP, so
it's building stuff using the Web standards, plus Java script of course. So Web 2.0 for some people it means moving some of the thinking client side so making it more immediate, but the idea of
the Web as interaction between people is really what the Web is. That was what it was designed to be as a collaborative
space where people can interact. Now, I really like the idea of people building things in hypertext, the sort of a common hypertext space to explain what
the common understanding is and thus capturing all the ideas which led to a given position. I think that's really
important. And I think that blogs and wikis are two things which are fun, I think they've taken off partly because they
do a lot of the management of the navigation for you and allow you to add content yourself. But I think there will be a whole lot more things like that to come, different sorts of ways in which people will be
able to work together. The semantic wikis are very interesting. These are wikis in which people can add data and then that data can then be
surfaced and sliced and diced using all kinds of different semantic Web tools, so that's why it's exciting the way
people, things are going, but I think there are lots of new things in that vein that we have yet to invent. LANINGHAM: I have one that I really struggle with, and I talk to co workers about, one of the tools of the
Internet, and that's e-mail, and how the e- mail client is kind of like the communications desktop or "the" desktop in
the office for so many people. And I feel like it's archaic in a way that it's a long list of stuff sorted by date and time or sender, and we waste so
much time sorting, and it's almost like an addiction, you can't pull away from it. Do you see some kind of paradigm
shift coming that will really change the way we communicate, and that e-mail won't be so dominant in our lives? BERNERS-LEE: E-mail is interesting, we can't live with it and you can't live without it [LAUGHTER]. As I said, first
of all, spam is clearly something which needs to be fixed. But if you try to fix e mail you'll often end up trying to
fix life. In the Web consortium we use the Web, we live in the Web. So a lot of the things other people do, back to
changing e mail, we do by editing Web documents. And sometimes that can be a lot easier. You just have shared editing,
and so documents don't move around, and because they're linked together we don't change the URIs, that means you can
find things. You don't have to keep running back for e-mails. So I think one of the ways in which within the Consortium we work differently is that we have an editable Web. We also
use e-mail, and of course they connect, because every e-mail which is sent to a public list, or a central list in
general, is archived and becomes a Web page. What's funny is that e-mail isn't completely integrated into this hypertext link space. E-mail messages have message
IDs on them, but if you click on a message ID there is in fact a message ID, MID:URI space. If you click on one of
those not a lot of systems know how to, what to do with it. Actually the W3C archive system does have a way of looking
at message IDs, you can give it a message ID and it will give you back an e-mail if it's got it. But most other systems,
a lot of other systems don't. People aren't used to using them, so they've got these URIs but they're not actually
used, so you can't, often you can't follow through the threads very easily because a lot of mailers don't give the links
back which allow you to create the thread. There are some ways in which I think e-mail could be improved. But on the other hand, the fact that a lot of people
want you to do things is a fact of life. LANINGHAM: Right [LAUGHTER]. BERNERS-LEE: And organizing the things that you're.... I get a lot of e mail, and people get...and some of it is
from people that I promised to do things for, and some of it is from random people I don't know asking for things. And sorting that and deciding on your priorities, how you're going to spend the next 10 minutes is always a challenge,
and that's just part of the art of living your life, I suppose. LANINGHAM: Can I ask you a wrap-up question by just saying, what are your long term hopes for the Internet and for
the Web and this amazing invention that we're all using now? Where do you see its place in our lives in the coming
decades? BERNERS-LEE: Well, the Internet was a really nice clean platform on which I could develop the Web, and I hope the Web
will continue to be a nice clean platform on which people keep developing other things. I hope the Semantic Web will take off so that the data basically all the data which is out there which you have access
to, to the Web pages, will now be available as data so you can treat it as data. There will be lots of very exciting
applications built on that. And we're starting to see that now, but it really is, you know, we're seriously into the
exponential growth of the Semantic Web right now, and that's very exciting. I think it too will be again a platform on which lots of other things to do will happen. It will be basically used for
data interoperability and for data integration across companies and between companies and between applications. But it
will then provide a basis for all kinds of exciting new systems built on which will be able to make use of that data. In general, I hope that we as humanity can learn to use this information space to understand each other, that we can
form ourselves into groups in lots of interesting ways so that between us that sort of just tangled web of human groups
spans the world and makes that so that it's not...you aren't too many clicks across the social Web from any one person
to any other one person, so that start really pulling together in the world and solving the huge challenges which we've
got without being distracted by fighting each other. So I suppose scientific progress and world peace, and the things that we've all got in the back of our minds, have
always got to drive us. And I think anybody in fact who does Web development, when you ask them about what they're
doing for a few minutes they'll end up back at those fundamental human desires. LANINGHAM: A wonderful vision, a worthwhile vision. Tim, this has been a real pleasure and honor. Thanks so much
for taking time to talk with us today. BERNERS-LEE: Scott, you're very welcome. LANINGHAM: Our guest has been Tim Berners-Lee, inventor of the World Wide Web, director of the World Wide Web
Consortium, and the 3Com Founders chair at the MIT Laboratory for Computer Science and Artificial Intelligence. Find
out more about developerWorks podcasts at ibm.com/developerworks/podcast. For everyone at developerWorks, I'm Scott Laningham. Talk to you next time. [END OF SEGMENT]
| 
