Tim Berners-Lee, raw data now! - Video talks
Time flies has been almost 20 years ago that I decided to rethink the way we use the information. the way we work together - I invented the World Wide Web now, two decades later, at TED, I want to ask your help for a new change of perspective.
So, going back to 1989, wrote a memorandum in which I suggested the global hypertext system. No one did much the case right now. But 18 months later - so come with innovation - 18 months later, my boss said I could take care of that project but as a side project, to break in a new computer that we had. It gave me time to write the program. So I threw down a draft as it should be HTML, the hypertext protocol - HTTP - the idea of \u200b\u200bthe URL - these names for things - that begin with http. I wrote the program and put the code available to all.
Why did I do? Well, it was basically frustration. I was frustrated - I was working as a software engineer in the laboratory huge, really exciting full of people who came from all over the world. Working with all sorts of computers, different from each other they use every kind of format for the data every kind, any kind of system documentation. So with all these differences, if I wanted to find a way to do something about taking something from one side and one from the other, anything I wanted to investigate, I had to connect to a new machine, learning how to run a new program, and eventually found the information I wanted some new data format. And they were all incompatible with each other. It was very frustrating. The frustration was all this untapped potential.
In fact, there were records on all disks. So if you could not imagine as part of a large virtual system documentation somewhere, perhaps on the Internet, life would be easier for everyone. Well, when you have an idea like that, something that gets under the skin and even if people do not read your memorandum - in fact my boss read it after her death, her copy was found. He wrote, "vague but exciting" in pencil, in a corner.
(Laughter)
But overall, it was hard - very hard to explain how the Web would be and it's hard to explain to people today as it was difficult to explain. But .. - OK, when Ted was born, the web did not exist then things like "click" did not have the same meaning. I could show someone a piece of hypertext, a page with links, click on the link and bing, here is a new page of hypertext. Nothing special. We had already seen - we hypertext documents on CD-ROM. The difficult thing was to get others to imagine then, imagine that that link could get you virtually any imaginable document. Ok, this is the gap that was really the most difficult to do. Well, someone he could not. And although it was difficult to explain, however, it created a spontaneous movement from below. And that was what made it more fun. This was the most exciting thing, not the technology, not what people have done, but the community spirit of all those people who met and exchanged emails. Here's how things then.
And you know what? It 's weird, but right now, things are returning to be back as then. I asked more or less at all, to make available their documents - I said, "Could you upload your documents about this, anywhere? "And you did. Thank you. It was spectacular, right? I mean, it was a very interesting thing because we have discovered that things that are outside of the Web in some way displace us. They go well beyond what we imagined at the beginning when we put together the website from which we started. Now, I want you to load your data anywhere. You will see that there is still a huge untapped potential. There is still a huge frustration because the data are not currently on the Web in the form of data.
What do you mean data? data, documents - that's the difference? Documents read them, OK? More or less, the documents are read, you can Follow the links between one and and the other is everything. With the data - you can do a lot of things if you have a computer. Those of you who were present, or at least seen the talk by Hans Rosling? One of the most beautiful - are many of you have seen - one of the best TED Talks. Hans showed a presentation in which he showed, for various countries and in different colors - the levels of income on one axis and levels of infant mortality, animating everything on a time scale. So, Hans took these data and gave a presentation that has broken many of the myths that people have on the economies of developing countries.
showed a slide like this. It reports to a variety of data OK, the data are boring brown box, and that's how we imagine them, no? Because the data itself does not have immediate effect but in reality, the data determine many things in our lives and this is because there is someone who takes this data and does something. In this case, Hans had put together the data that was harvested from each kind of United Nations website. He had collected the data, was combined into something more interesting than the individual parts and then he had entered into a software, which I think has developed her son, originally, and did this wonderful presentation. And Hans insisted, "Look, it's really important to have lots of data." And I was happy to see that the party the other night, he repeated it with emphasis, "It 's really important to have lots of data."
So I think we all now rather than two simple types of data correlate, or six, as he did, but I want to think about a world where everyone has uploaded data on the web and therefore practically anything you can imagine is anywhere. and call all linked data. The technology is the linked data, and is extremely simple. If you want to put something on the internet, there are three rules are the first thing those names HTTP - those things that start with http:-- now we're using them just for the documents, but use them to indicate things that speak documents. We use it for people, use them to places, use them to your products, we use it for events. Every kind of concept, now has a name that begins with HTTP.
Second rule, if I take one of these names HTTP and try and go anywhere, and retrieves the corresponding data using the HTTP protocol from the web, he will find the data in a standard format that could be useful data, which may be of interest to someone about this or that event. Who went to that event? Whatever concerns one of those people, where she was born, or the like. So the second rule is that I can obtain important information.
The third is that when I return this information is not only the height, weight and date of birth of someone, but I will get reports. The data are relationships. Interestingly, the data reports. The person that was born in Berlin, Berlin, Germany. And when there are reports every time there is a relationship that has given the other reports received also one of those names that begin with HTTP. Then I can go and see this new information. So [if] I want a person - I can see the city where she was born I can see the region in which they find themselves in that city, what is the population of this city, and so on. So I can scroll through all of this information.
's all, really. These are the linked data. I wrote an article entitled "linked data" a couple of years ago and soon after began to happen a little 'things. The idea of \u200b\u200blinked data is that we can have many, many of these boxes which used Hans, and therefore many, many other things that will sprout. And it is not only a population of new plants. It is not just a root that nourishes a plant, but each of those plants, of whatever type it is - a presentation, analysis, someone who looked for patterns in data - [who does] see all the data and connects them with each other, and the important thing is that the more data you have to connect things to each other, the data become more powerful.
So, linked data. The meme has spread out there. And soon enough, to Chris Bizer Freie Universitat Berlin has been one of the first to put together something interesting, he noted that Wikipedia - you know, Wikipedia, the online encyclopedia that contains a lot of interesting documents. Well, in those documents, there are little boxes, little boxes. And in most of these info boxes, there are data. So he wrote a program that extrapolates that data from Wikipedia, and puts them into a set of linked data on the web, which he called DBpedia. DBpedia is represented by blue in the middle of this slide and if you go to look for "Berlin", you will discover that there are other sets of data which in turn contain information about Berlin, and are interconnected. So if some of the data retrieved DBpedia of Berlin, you're going to get all this other information also. And what is exciting is that it is starting to grow again is the beginning of a spontaneous movement from below.
Let us pause to reflect on the data for a moment. The data are in a lot of different shapes Think about the variety of the web, it's really important: the web allows you to upload any kind of information. So it is with the data. I might refer to any type of data. Think of the government data, the data of the companies are very important, there are scientific data, personal data, there are meteorological data, information on events, there are no data on conferences, on the news and on every type of thing. Will name only some of these to give you an idea of \u200b\u200bthe variety, so that you can see what the potential behind it.
start with government data. Barack Obama in a speech said that the data the U.S. government would be made available online in accessible formats. And I just hope to make them available as linked data. This is important. Why is it important? Not only for reasons of transparency, of course, government transparency is important, but that data - are the data from all government departments Think about how many of those data are related to how we live in America. They are really helpful. They have value. Can I use in my company. If I was a kid, I could use them to my homework. We are talking about how to run the world better by making these data available.
In fact if you yourself are responsible - if you know of data in a particular government department, often discover that these people are very tempted to keep them - Hans calls him "hold on to database." We held tight to its database, do not let go until it's done a nice website to log on. Well, I would suggest instead - yes, you make a nice website, who am I to tell you not to make a nice website? Take a deep site, but first give us the unadulterated data, we want the data. We want unadulterated data. Ok, we require the raw data now. And I ask you to ask them to work out, OK? Say "naked and raw."
Audience: Nudes and raw
Tim Berners-Lee: Can you say "data"?
Audience: Data
TBL: You can say "now"?
the audience: "Now!"
TBL: Well, raw data now! The audience
: raw data now!
Repeat: It is important because you have no idea of \u200b\u200bthe amount of excuses that people hold on to invent their own data and do not give it to you, even if you have already paid your taxes. And not just in America. It's so all over the world. And not just governments, of course - [but] even companies.
I will tell you just a few more reflections on the data. We here at TED, and we have in mind the challenges that human society is facing right now - treat cancer, understanding how the brain to [treat] Alzheimer's disease, understand the economy to make it a bit more stable , understanding how the world works. The people who will solve these problems - scientists - have in mind ideas developed in part, and try to comunicarsele over the web. But much of human knowledge at the moment is in the database, which often reside on their computers, and in fact, at present, is not shared.
To explain, go into a theme - if you think Alzheimer's disease, for example, the discovery of drugs - there are a lot of data links that are coming out because the scientists in that field have come to realize that it is a good way to get out of these silos, because have their genome data in a database in a certain building, and have the data on proteins in another And now they are sticking to each other - data link - and may put the question type, you probably would not do, that I did not do - [but] they. What proteins are involved in signal transduction and are also connected to the pyramidal neurons? Well, if you take these four words, and certainly not included in Google found a page that answers this Asked why no one has ever asked this question. You get 223,000 results - but none that is of any use. If you do the same question to the linked data - now they have put together - you get 32 \u200b\u200bresults, each of which is a protein with the properties that you can go and see. Being able to ask questions of this kind, as a scientist - questions that involve different disciplines - is a real change with a capital. It is very, very important. Scientists have their hands tied at the moment - the potential of other scientists have collected data that is inaccessible and we are making this data available to address these enormous problems.
Now, if this continues, you will think All data come from large institutions and have nothing to do with you. But this is not true In fact, the data about our lives When you log in to your social networking site, in your favorite, and say, "This is my friend." Bing! Report. Given. Say, "This photograph shows this person." Bing! That's a given. Data, data, data. Each time you do something on your social network, the site is obtaining the data and is using them - they're reusing - and is using them to spice up the lives of other people on the site. But when you go to another site linked data - and say that this is a travel site, and say, "I want to send this picture to all persons that group, "can not get over the virtual walls. The Economist has described him in an article, and many people have written in their blogs on this subject - a tremendous frustration. The way to destroy the silos is to achieve interoperability among social networking sites. We must do so through the linked data.
will talk about one last type of data is perhaps the most exciting. Before arriving here, I did a search on OpenStreetMap OpenStreetMap is a map, but also a Wiki. zoomed and that square is a theater - one in which we are now - the Terrace Theater. He had a name on the map. So I went into edit mode, I selected the theater, I added the name at the bottom, and I saved. Now if you go back and find this place on OpenStreetMap.org find that the Terrace Theater has a name. I did it myself! I did this on the map. I just did! I put it there myself. Hey, guess what? If I - the map is made up entirely of individual contributions and creates an incredible resource because everyone does their part. And the data link is just that. We're talking about people doing their part to add a small portion, and connect everything. This is how the linked data. You do your part. All the others do theirs. Maybe you have not a lot of data contents to be included, but you know how to ask them. We have practiced before.
So, linked data - are a thing of enormous. I've told very few things I have data in every aspect of our existence, in every aspect of work and leisure, and we're not just talking about the number of places from which there's data, we are talking about connecting them to each other. And when you connect the data, obtain power in a way that simply does not happen with the web, with the documents. You can come out of this immense potential. So, we got to the point where it is necessary to act --- the people who think it's a great idea and all the people - and I think there are a lot of acting at TED for a reason - even if there is an immediate return on investment because the return will come when everyone will have made their part - there will be people who do it because I'm the kind of people who do things that make if everybody do. Ok, so called linked data. I want you to produce them. I want you to requires them. And I think it's an idea that deserves to be widespread. Thanks
(Applause)
0 comments:
Post a Comment