TED2009
Tim Berners-Lee: The next web
Filmed:
Readability: 3.8
1,638,798 views
20 years ago, Tim Berners-Lee invented the World Wide Web. For his next project, he's building a web for open, linked data that could do for numbers what the Web did for words, pictures, video: unlock our data and reframe the way we use it together.
Tim Berners-Lee - Inventor
Tim Berners-Lee invented the World Wide Web. He leads the World Wide Web Consortium (W3C), overseeing the Web's standards and development. Full bio
Tim Berners-Lee invented the World Wide Web. He leads the World Wide Web Consortium (W3C), overseeing the Web's standards and development. Full bio
Double-click the English transcript below to play the video.
00:18
Time flies.
0
0
2000
00:20
It's actually almost 20 years ago
1
2000
2000
00:22
when I wanted to reframe the way we use information,
2
4000
4000
00:26
the way we work together: I invented the World Wide Web.
3
8000
3000
00:29
Now, 20 years on, at TED,
4
11000
3000
00:32
I want to ask your help in a new reframing.
5
14000
4000
00:37
So going back to 1989,
6
19000
4000
00:41
I wrote a memo suggesting the global hypertext system.
7
23000
3000
00:44
Nobody really did anything with it, pretty much.
8
26000
3000
00:47
But 18 months later -- this is how innovation happens --
9
29000
4000
00:51
18 months later, my boss said I could do it on the side,
10
33000
4000
00:55
as a sort of a play project,
11
37000
2000
00:57
kick the tires of a new computer we'd got.
12
39000
2000
00:59
And so he gave me the time to code it up.
13
41000
3000
01:02
So I basically roughed out what HTML should look like:
14
44000
5000
01:07
hypertext protocol, HTTP;
15
49000
3000
01:10
the idea of URLs, these names for things
16
52000
3000
01:13
which started with HTTP.
17
55000
2000
01:15
I wrote the code and put it out there.
18
57000
2000
01:17
Why did I do it?
19
59000
2000
01:19
Well, it was basically frustration.
20
61000
2000
01:21
I was frustrated -- I was working as a software engineer
21
63000
4000
01:25
in this huge, very exciting lab,
22
67000
2000
01:27
lots of people coming from all over the world.
23
69000
2000
01:29
They brought all sorts of different computers with them.
24
71000
3000
01:32
They had all sorts of different data formats,
25
74000
3000
01:35
all sorts, all kinds of documentation systems.
26
77000
2000
01:37
So that, in all that diversity,
27
79000
3000
01:40
if I wanted to figure out how to build something
28
82000
2000
01:42
out of a bit of this and a bit of this,
29
84000
2000
01:44
everything I looked into, I had to connect to some new machine,
30
86000
4000
01:48
I had to learn to run some new program,
31
90000
2000
01:50
I would find the information I wanted in some new data format.
32
92000
5000
01:55
And these were all incompatible.
33
97000
2000
01:57
It was just very frustrating.
34
99000
2000
01:59
The frustration was all this unlocked potential.
35
101000
2000
02:01
In fact, on all these discs there were documents.
36
103000
3000
02:04
So if you just imagined them all
37
106000
3000
02:07
being part of some big, virtual documentation system in the sky,
38
109000
5000
02:12
say on the Internet,
39
114000
2000
02:14
then life would be so much easier.
40
116000
2000
02:16
Well, once you've had an idea like that it kind of gets under your skin
41
118000
4000
02:20
and even if people don't read your memo --
42
122000
2000
02:22
actually he did, it was found after he died, his copy.
43
124000
3000
02:25
He had written, "Vague, but exciting," in pencil, in the corner.
44
127000
3000
02:28
(Laughter)
45
130000
2000
02:30
But in general it was difficult -- it was really difficult to explain
46
132000
4000
02:34
what the web was like.
47
136000
2000
02:36
It's difficult to explain to people now that it was difficult then.
48
138000
2000
02:38
But then -- OK, when TED started, there was no web
49
140000
3000
02:41
so things like "click" didn't have the same meaning.
50
143000
3000
02:44
I can show somebody a piece of hypertext,
51
146000
2000
02:46
a page which has got links,
52
148000
2000
02:48
and we click on the link and bing -- there'll be another hypertext page.
53
150000
4000
02:52
Not impressive.
54
154000
2000
02:54
You know, we've seen that -- we've got things on hypertext on CD-ROMs.
55
156000
3000
02:57
What was difficult was to get them to imagine:
56
159000
3000
03:00
so, imagine that that link could have gone
57
162000
4000
03:04
to virtually any document you could imagine.
58
166000
2000
03:07
Alright, that is the leap that was very difficult for people to make.
59
169000
4000
03:11
Well, some people did.
60
173000
2000
03:13
So yeah, it was difficult to explain, but there was a grassroots movement.
61
175000
3000
03:17
And that is what has made it most fun.
62
179000
4000
03:21
That has been the most exciting thing,
63
183000
2000
03:23
not the technology, not the things people have done with it,
64
185000
2000
03:25
but actually the community, the spirit of all these people
65
187000
2000
03:27
getting together, sending the emails.
66
189000
2000
03:29
That's what it was like then.
67
191000
2000
03:31
Do you know what? It's funny, but right now it's kind of like that again.
68
193000
3000
03:34
I asked everybody, more or less, to put their documents --
69
196000
2000
03:36
I said, "Could you put your documents on this web thing?"
70
198000
3000
03:39
And you did.
71
201000
3000
03:42
Thanks.
72
204000
1000
03:43
It's been a blast, hasn't it?
73
205000
2000
03:45
I mean, it has been quite interesting
74
207000
2000
03:47
because we've found out that the things that happen with the web
75
209000
2000
03:49
really sort of blow us away.
76
211000
2000
03:51
They're much more than we'd originally imagined
77
213000
2000
03:53
when we put together the little, initial website
78
215000
2000
03:55
that we started off with.
79
217000
2000
03:57
Now, I want you to put your data on the web.
80
219000
3000
04:00
Turns out that there is still huge unlocked potential.
81
222000
4000
04:04
There is still a huge frustration
82
226000
2000
04:06
that people have because we haven't got data on the web as data.
83
228000
4000
04:10
What do you mean, "data"? What's the difference -- documents, data?
84
232000
2000
04:12
Well, documents you read, OK?
85
234000
3000
04:15
More or less, you read them, you can follow links from them, and that's it.
86
237000
3000
04:18
Data -- you can do all kinds of stuff with a computer.
87
240000
2000
04:20
Who was here or has otherwise seen Hans Rosling's talk?
88
242000
6000
04:26
One of the great -- yes a lot of people have seen it --
89
248000
4000
04:30
one of the great TED Talks.
90
252000
2000
04:32
Hans put up this presentation
91
254000
2000
04:34
in which he showed, for various different countries, in various different colors --
92
256000
5000
04:39
he showed income levels on one axis
93
261000
3000
04:42
and he showed infant mortality, and he shot this thing animated through time.
94
264000
3000
04:45
So, he'd taken this data and made a presentation
95
267000
4000
04:49
which just shattered a lot of myths that people had
96
271000
3000
04:52
about the economics in the developing world.
97
274000
4000
04:56
He put up a slide a little bit like this.
98
278000
2000
04:58
It had underground all the data
99
280000
2000
05:00
OK, data is brown and boxy and boring,
100
282000
3000
05:03
and that's how we think of it, isn't it?
101
285000
2000
05:05
Because data you can't naturally use by itself
102
287000
3000
05:08
But in fact, data drives a huge amount of what happens in our lives
103
290000
4000
05:12
and it happens because somebody takes that data and does something with it.
104
294000
3000
05:15
In this case, Hans had put the data together
105
297000
2000
05:17
he had found from all kinds of United Nations websites and things.
106
299000
5000
05:22
He had put it together,
107
304000
2000
05:24
combined it into something more interesting than the original pieces
108
306000
3000
05:27
and then he'd put it into this software,
109
309000
5000
05:32
which I think his son developed, originally,
110
314000
2000
05:34
and produces this wonderful presentation.
111
316000
3000
05:37
And Hans made a point
112
319000
2000
05:39
of saying, "Look, it's really important to have a lot of data."
113
321000
4000
05:43
And I was happy to see that at the party last night
114
325000
3000
05:46
that he was still saying, very forcibly, "It's really important to have a lot of data."
115
328000
4000
05:50
So I want us now to think about
116
332000
2000
05:52
not just two pieces of data being connected, or six like he did,
117
334000
4000
05:56
but I want to think about a world where everybody has put data on the web
118
338000
5000
06:01
and so virtually everything you can imagine is on the web
119
343000
2000
06:03
and then calling that linked data.
120
345000
2000
06:05
The technology is linked data, and it's extremely simple.
121
347000
2000
06:07
If you want to put something on the web there are three rules:
122
349000
4000
06:11
first thing is that those HTTP names --
123
353000
3000
06:14
those things that start with "http:" --
124
356000
2000
06:16
we're using them not just for documents now,
125
358000
4000
06:20
we're using them for things that the documents are about.
126
362000
2000
06:22
We're using them for people, we're using them for places,
127
364000
2000
06:24
we're using them for your products, we're using them for events.
128
366000
4000
06:28
All kinds of conceptual things, they have names now that start with HTTP.
129
370000
4000
06:32
Second rule, if I take one of these HTTP names and I look it up
130
374000
5000
06:37
and I do the web thing with it and I fetch the data
131
379000
2000
06:39
using the HTTP protocol from the web,
132
381000
2000
06:41
I will get back some data in a standard format
133
383000
3000
06:44
which is kind of useful data that somebody might like to know
134
386000
5000
06:49
about that thing, about that event.
135
391000
2000
06:51
Who's at the event? Whatever it is about that person,
136
393000
2000
06:53
where they were born, things like that.
137
395000
2000
06:55
So the second rule is I get important information back.
138
397000
2000
06:57
Third rule is that when I get back that information
139
399000
4000
07:01
it's not just got somebody's height and weight and when they were born,
140
403000
3000
07:04
it's got relationships.
141
406000
2000
07:06
Data is relationships.
142
408000
2000
07:08
Interestingly, data is relationships.
143
410000
2000
07:10
This person was born in Berlin; Berlin is in Germany.
144
412000
4000
07:14
And when it has relationships, whenever it expresses a relationship
145
416000
3000
07:17
then the other thing that it's related to
146
419000
3000
07:20
is given one of those names that starts HTTP.
147
422000
4000
07:24
So, I can go ahead and look that thing up.
148
426000
2000
07:26
So I look up a person -- I can look up then the city where they were born; then
149
428000
3000
07:29
I can look up the region it's in, and the town it's in,
150
431000
3000
07:32
and the population of it, and so on.
151
434000
3000
07:35
So I can browse this stuff.
152
437000
2000
07:37
So that's it, really.
153
439000
2000
07:39
That is linked data.
154
441000
2000
07:41
I wrote an article entitled "Linked Data" a couple of years ago
155
443000
3000
07:44
and soon after that, things started to happen.
156
446000
4000
07:48
The idea of linked data is that we get lots and lots and lots
157
450000
4000
07:52
of these boxes that Hans had,
158
454000
2000
07:54
and we get lots and lots and lots of things sprouting.
159
456000
2000
07:56
It's not just a whole lot of other plants.
160
458000
3000
07:59
It's not just a root supplying a plant,
161
461000
2000
08:01
but for each of those plants, whatever it is --
162
463000
3000
08:04
a presentation, an analysis, somebody's looking for patterns in the data --
163
466000
3000
08:07
they get to look at all the data
164
469000
3000
08:10
and they get it connected together,
165
472000
2000
08:12
and the really important thing about data
166
474000
2000
08:14
is the more things you have to connect together, the more powerful it is.
167
476000
2000
08:16
So, linked data.
168
478000
2000
08:18
The meme went out there.
169
480000
2000
08:20
And, pretty soon Chris Bizer at the Freie Universitat in Berlin
170
482000
4000
08:24
who was one of the first people to put interesting things up,
171
486000
2000
08:26
he noticed that Wikipedia --
172
488000
2000
08:28
you know Wikipedia, the online encyclopedia
173
490000
3000
08:31
with lots and lots of interesting documents in it.
174
493000
2000
08:33
Well, in those documents, there are little squares, little boxes.
175
495000
4000
08:37
And in most information boxes, there's data.
176
499000
3000
08:40
So he wrote a program to take the data, extract it from Wikipedia,
177
502000
4000
08:44
and put it into a blob of linked data
178
506000
2000
08:46
on the web, which he called dbpedia.
179
508000
3000
08:49
Dbpedia is represented by the blue blob in the middle of this slide
180
511000
4000
08:53
and if you actually go and look up Berlin,
181
515000
2000
08:55
you'll find that there are other blobs of data
182
517000
2000
08:57
which also have stuff about Berlin, and they're linked together.
183
519000
3000
09:00
So if you pull the data from dbpedia about Berlin,
184
522000
3000
09:03
you'll end up pulling up these other things as well.
185
525000
2000
09:05
And the exciting thing is it's starting to grow.
186
527000
3000
09:08
This is just the grassroots stuff again, OK?
187
530000
2000
09:10
Let's think about data for a bit.
188
532000
3000
09:13
Data comes in fact in lots and lots of different forms.
189
535000
3000
09:16
Think of the diversity of the web. It's a really important thing
190
538000
3000
09:19
that the web allows you to put all kinds of data up there.
191
541000
3000
09:22
So it is with data. I could talk about all kinds of data.
192
544000
2000
09:25
We could talk about government data, enterprise data is really important,
193
547000
4000
09:29
there's scientific data, there's personal data,
194
551000
3000
09:32
there's weather data, there's data about events,
195
554000
2000
09:34
there's data about talks, and there's news and there's all kinds of stuff.
196
556000
4000
09:38
I'm just going to mention a few of them
197
560000
3000
09:41
so that you get the idea of the diversity of it,
198
563000
2000
09:43
so that you also see how much unlocked potential.
199
565000
4000
09:47
Let's start with government data.
200
569000
2000
09:49
Barack Obama said in a speech,
201
571000
2000
09:51
that he -- American government data would be available on the Internet
202
573000
5000
09:56
in accessible formats.
203
578000
2000
09:58
And I hope that they will put it up as linked data.
204
580000
2000
10:00
That's important. Why is it important?
205
582000
2000
10:02
Not just for transparency, yeah transparency in government is important,
206
584000
3000
10:05
but that data -- this is the data from all the government departments
207
587000
3000
10:08
Think about how much of that data is about how life is lived in America.
208
590000
5000
10:13
It's actual useful. It's got value.
209
595000
2000
10:15
I can use it in my company.
210
597000
2000
10:17
I could use it as a kid to do my homework.
211
599000
2000
10:19
So we're talking about making the place, making the world run better
212
601000
3000
10:22
by making this data available.
213
604000
2000
10:24
In fact if you're responsible -- if you know about some data
214
606000
4000
10:28
in a government department, often you find that
215
610000
2000
10:30
these people, they're very tempted to keep it --
216
612000
3000
10:33
Hans calls it database hugging.
217
615000
3000
10:36
You hug your database, you don't want to let it go
218
618000
2000
10:38
until you've made a beautiful website for it.
219
620000
2000
10:40
Well, I'd like to suggest that rather --
220
622000
2000
10:42
yes, make a beautiful website,
221
624000
2000
10:44
who am I to say don't make a beautiful website?
222
626000
2000
10:46
Make a beautiful website, but first
223
628000
3000
10:49
give us the unadulterated data,
224
631000
3000
10:52
we want the data.
225
634000
2000
10:54
We want unadulterated data.
226
636000
2000
10:56
OK, we have to ask for raw data now.
227
638000
3000
10:59
And I'm going to ask you to practice that, OK?
228
641000
2000
11:01
Can you say "raw"?
229
643000
1000
11:02
Audience: Raw.
230
644000
1000
11:03
Tim Berners-Lee: Can you say "data"?
231
645000
1000
11:04
Audience: Data.
232
646000
1000
11:05
TBL: Can you say "now"?
233
647000
1000
11:06
Audience: Now!
234
648000
1000
11:07
TBL: Alright, "raw data now"!
235
649000
2000
11:09
Audience: Raw data now!
236
651000
2000
11:11
Practice that. It's important because you have no idea the number of excuses
237
653000
4000
11:15
people come up with to hang onto their data
238
657000
2000
11:17
and not give it to you, even though you've paid for it as a taxpayer.
239
659000
4000
11:21
And it's not just America. It's all over the world.
240
663000
2000
11:23
And it's not just governments, of course -- it's enterprises as well.
241
665000
3000
11:26
So I'm just going to mention a few other thoughts on data.
242
668000
3000
11:29
Here we are at TED, and all the time we are very conscious
243
671000
5000
11:34
of the huge challenges that human society has right now --
244
676000
5000
11:39
curing cancer, understanding the brain for Alzheimer's,
245
681000
3000
11:42
understanding the economy to make it a little bit more stable,
246
684000
3000
11:45
understanding how the world works.
247
687000
2000
11:47
The people who are going to solve those -- the scientists --
248
689000
2000
11:49
they have half-formed ideas in their head,
249
691000
2000
11:51
they try to communicate those over the web.
250
693000
3000
11:54
But a lot of the state of knowledge of the human race at the moment
251
696000
3000
11:57
is on databases, often sitting in their computers,
252
699000
3000
12:00
and actually, currently not shared.
253
702000
3000
12:03
In fact, I'll just go into one area --
254
705000
3000
12:06
if you're looking at Alzheimer's, for example,
255
708000
2000
12:08
drug discovery -- there is a whole lot of linked data which is just coming out
256
710000
3000
12:11
because scientists in that field realize
257
713000
2000
12:13
this is a great way of getting out of those silos,
258
715000
3000
12:16
because they had their genomics data in one database
259
718000
4000
12:20
in one building, and they had their protein data in another.
260
722000
3000
12:23
Now, they are sticking it onto -- linked data --
261
725000
3000
12:26
and now they can ask the sort of question, that you probably wouldn't ask,
262
728000
3000
12:29
I wouldn't ask -- they would.
263
731000
2000
12:31
What proteins are involved in signal transduction
264
733000
2000
12:33
and also related to pyramidal neurons?
265
735000
2000
12:35
Well, you take that mouthful and you put it into Google.
266
737000
3000
12:38
Of course, there's no page on the web which has answered that question
267
740000
3000
12:41
because nobody has asked that question before.
268
743000
2000
12:43
You get 223,000 hits --
269
745000
2000
12:45
no results you can use.
270
747000
2000
12:47
You ask the linked data -- which they've now put together --
271
749000
3000
12:50
32 hits, each of which is a protein which has those properties
272
752000
4000
12:54
and you can look at.
273
756000
2000
12:56
The power of being able to ask those questions, as a scientist --
274
758000
3000
12:59
questions which actually bridge across different disciplines --
275
761000
2000
13:01
is really a complete sea change.
276
763000
3000
13:04
It's very very important.
277
766000
2000
13:06
Scientists are totally stymied at the moment --
278
768000
2000
13:08
the power of the data that other scientists have collected is locked up
279
770000
5000
13:13
and we need to get it unlocked so we can tackle those huge problems.
280
775000
3000
13:16
Now if I go on like this, you'll think that all the data comes from huge institutions
281
778000
4000
13:20
and has nothing to do with you.
282
782000
3000
13:23
But, that's not true.
283
785000
2000
13:25
In fact, data is about our lives.
284
787000
2000
13:27
You just -- you log on to your social networking site,
285
789000
3000
13:30
your favorite one, you say, "This is my friend."
286
792000
2000
13:32
Bing! Relationship. Data.
287
794000
3000
13:35
You say, "This photograph, it's about -- it depicts this person. "
288
797000
3000
13:38
Bing! That's data. Data, data, data.
289
800000
3000
13:41
Every time you do things on the social networking site,
290
803000
2000
13:43
the social networking site is taking data and using it -- re-purposing it --
291
805000
4000
13:47
and using it to make other people's lives more interesting on the site.
292
809000
4000
13:51
But, when you go to another linked data site --
293
813000
2000
13:53
and let's say this is one about travel,
294
815000
3000
13:56
and you say, "I want to send this photo to all the people in that group,"
295
818000
3000
13:59
you can't get over the walls.
296
821000
2000
14:01
The Economist wrote an article about it, and lots of people have blogged about it --
297
823000
2000
14:03
tremendous frustration.
298
825000
1000
14:04
The way to break down the silos is to get inter-operability
299
826000
2000
14:06
between social networking sites.
300
828000
2000
14:08
We need to do that with linked data.
301
830000
2000
14:10
One last type of data I'll talk about, maybe it's the most exciting.
302
832000
3000
14:13
Before I came down here, I looked it up on OpenStreetMap
303
835000
3000
14:16
The OpenStreetMap's a map, but it's also a Wiki.
304
838000
2000
14:18
Zoom in and that square thing is a theater -- which we're in right now --
305
840000
3000
14:21
The Terrace Theater. It didn't have a name on it.
306
843000
2000
14:23
So I could go into edit mode, I could select the theater,
307
845000
2000
14:25
I could add down at the bottom the name, and I could save it back.
308
847000
5000
14:30
And now if you go back to the OpenStreetMap. org,
309
852000
3000
14:33
and you find this place, you will find that The Terrace Theater has got a name.
310
855000
3000
14:36
I did that. Me!
311
858000
2000
14:38
I did that to the map. I just did that!
312
860000
2000
14:40
I put that up on there. Hey, you know what?
313
862000
2000
14:42
If I -- that street map is all about everybody doing their bit
314
864000
3000
14:45
and it creates an incredible resource
315
867000
3000
14:48
because everybody else does theirs.
316
870000
3000
14:51
And that is what linked data is all about.
317
873000
3000
14:54
It's about people doing their bit
318
876000
3000
14:57
to produce a little bit, and it all connecting.
319
879000
3000
15:00
That's how linked data works.
320
882000
3000
15:03
You do your bit. Everybody else does theirs.
321
885000
4000
15:07
You may not have lots of data which you have yourself to put on there
322
889000
4000
15:11
but you know to demand it.
323
893000
3000
15:14
And we've practiced that.
324
896000
2000
15:16
So, linked data -- it's huge.
325
898000
4000
15:20
I've only told you a very small number of things
326
902000
3000
15:23
There are data in every aspect of our lives,
327
905000
2000
15:25
every aspect of work and pleasure,
328
907000
3000
15:28
and it's not just about the number of places where data comes,
329
910000
3000
15:31
it's about connecting it together.
330
913000
3000
15:34
And when you connect data together, you get power
331
916000
3000
15:37
in a way that doesn't happen just with the web, with documents.
332
919000
3000
15:40
You get this really huge power out of it.
333
922000
4000
15:44
So, we're at the stage now
334
926000
3000
15:47
where we have to do this -- the people who think it's a great idea.
335
929000
4000
15:51
And all the people -- and I think there's a lot of people at TED who do things because --
336
933000
3000
15:54
even though there's not an immediate return on the investment
337
936000
2000
15:56
because it will only really pay off when everybody else has done it --
338
938000
3000
15:59
they'll do it because they're the sort of person who just does things
339
941000
4000
16:03
which would be good if everybody else did them.
340
945000
3000
16:06
OK, so it's called linked data.
341
948000
2000
16:08
I want you to make it.
342
950000
2000
16:10
I want you to demand it.
343
952000
2000
16:12
And I think it's an idea worth spreading.
344
954000
2000
16:14
Thanks.
345
956000
1000
16:15
(Applause)
346
957000
3000
ABOUT THE SPEAKER
Tim Berners-Lee - InventorTim Berners-Lee invented the World Wide Web. He leads the World Wide Web Consortium (W3C), overseeing the Web's standards and development.
Why you should listen
In the 1980s, scientists at CERN were asking themselves how massive, complex, collaborative projects -- like the fledgling LHC -- could be orchestrated and tracked. Tim Berners-Lee, then a contractor, answered by inventing the World Wide Web. This global system of hypertext documents, linked through the Internet, brought about a massive cultural shift ushered in by the new tech and content it made possible: AOL, eBay, Wikipedia, TED.com...
Berners-Lee is now director of the World Wide Web Consortium (W3C), which maintains standards for the Web and continues to refine its design. Recently he has envisioned a "Semantic Web" -- an evolved version of the same system that recognizes the meaning of the information it carries. He's the 3Com Founders Professor of Engineering in the School of Engineering with a joint appointment in the Department of Electrical Engineering and Computer Science at the Laboratory for Computer Science and Artificial Intelligence (CSAIL) at the MIT, where he also heads the Decentralized Information Group (DIG). He is also a Professor in the Electronics and Computer Science Department at the University of Southampton, UK.
Tim Berners-Lee | Speaker | TED.com