Backbone is about business, technology, lifestyle, innovation, bold ideas, trends and events
Defining Web 3.0 | March 31, 2009
Ask five industry luminaries what Web 3.0 means and you'll get at least as many answers, but wherever we're going one thing is certain: we're moving beyond the dumb Internet of yesterday
By Danny Bradbury
Telecommunications firms have moved from 2G to 3G. Windows moved beyond version 3.0 years ago. And yet, if the marketing community is to be believed, the Web is still stuck at version 2.0. Web 3.0 is being floated as a concept, but does anyone know what it really means?
Publishing mogul Tim O’Reilly, who first sparked the Web 2.0 label with a conference of the same name, has argued that the concept of Web 3.0 is nonsense. He defined the term Web 2.0 not as a milestone for the development of new technology, but rather as a means of pulling the Web industry out of its post-crash economic mire. In short, by his own definition, it was more of a marketing concept than anything else.
“For ‘Web 3.0’ to be meaningful, we’ll need to see a serious discontinuity from the previous generation of technology. That might be another bust and resurgence, or more likely, it will be something qualitatively different,” he said in a 2007 blog post on the subject.
Gartner vice-president David Smith describes what such a departure could look like, defining it across three contexts: technology, community interaction and business. “On the technology side, the biggest thing bar none is the continued emergence of the mobile Web,” he said. “On the community side, the biggest thing is the bringing together of location and community.” With the introduction of GPS-based mobile devices, these two things have gained more synergy. Finally, “on the business side, that’s where a lot of the cloud computing stuff comes in, because a lot of the subscription-based services funded through advertising fit that very well.”
Then there are other definitions. Some think 3.0 will be a more advanced form of the group intelligence inspired by Web 2.0 sites like Wikipedia. Others think that it will be a more ubiquitous and powerful form of application integration than what we saw with the publication of open APIs by Web 2.0 companies like Flickr and Google.
They’re all right, said Adriana Lukas, a social media consultant and founder of the UK-based Big Blog Company. “If you look at each of those technologies, they spell out layers of the Web,” she said. Cloud computing represents the bottom, most technical layer of the Web 3.0 stack, where the infrastructural stuff—the hardware and networks—live. Intelligent applications map onto the higher layer, where software lives.
Web ownership
For Lukas, though, the user is perhaps the defining characteristic of Web 3.0. Web 1.0 was designed around systems. Users who wanted to interact with it had to learn HTML and overcome the limitations of non-dynamic Web sites. Web 2.0 is more user centric in the sense that it enabled users to contribute their own content and document their own experiences. But it was still built by developers, who defined the rules.
“User centric was a step forward, in the sense that you built a system around the user, but the user is still not at the core of the system. They’re not driving it,” she said.
The key to a true user-driven Web is what Lukas and Dan Grigorovici, who chairs a conference on Web 3.0 in New York this May, both call “vendor relationship management” or VRM. VRM inverts the relationship between the user and the service provider typified by traditional customer relationship management (CRM). Today, companies collect data on user activities, often without the users knowing. In a VRM scenario, the users own their data and can make it available to a variety of different service providers to suit their own needs. This will revolutionize the monetary model for the Web, he said.
“One key element is monetizing the data in a different way than just slapping a banner on a Web site,” Grigorovici said, adding that the Web 2.0 concept of mashing up data only takes you so far. “When you monetize that information you won’t just mash it up. You add more value than simply slapping one data set on top of another.
“Let’s say you’re in the mood to buy a sweater but you couldn’t find anything on the Amazon site,” he said, explaining how user-owned data would work. “If you could make available the fact that you’re checking this particular sweater on Amazon or Overstock, then this is information that Wal-Mart would not otherwise have access to. And they might want to give you a discount.”
For users to communicate that, they must own their own data, rather than having it hidden and locked down by companies in their own CRM systems. Lukas goes as far as suggesting that users maintain data in a separate repository of their own, using a system that her team is currently coding, called MINE!
Leaving the dumb Internet behind
Ownership in Web 3.0 isn’t the only thing about the data that will change. According to the inventor of Web 1.0, its structure and meaning will change, too. Tim Berners-Lee developed the protocols underpinning the Web in 1990 when working at the European Organisation for Nuclear Research (CERN). Today, Berners-Lee laments the Web’s one crucial flaw: it doesn’t understand what its own data means. For example, looking up a movie online turns up a variety of predictable results, but it’s an inherently dumb search.
Think about using the Web as if you were having conversations with clerks in two video stores. Bob knows nothing about movies, and could just as easily be working at a grocery store. Alice is a film student earning some spare cash while she finishes her master’s degree in German Expressionism. You ask both of them for a copy of 8 1/2.
Bob: “It’s not in the computer. The next thing on the list is 9 1/2 Weeks. It’s in the erotic entertainment section.”
Alice: “We don’t have 8 1/2, but that’s a Fellini film and it’s a self-referential work about a film director trying to make a movie. I also remember that 8 1/2 won an Oscar for best costume design. If you’re into costumes, then Fellini’s La Dolce Vita might be your best bet, because that won the same award. Alternatively, you might try another movie by 8 1/2’s producer, Angelo Rizzoli. The next movie that he produced was Deserto rosso, Il. Or if you’re into films with a similar subject and don’t care about the director or the period, then here’s a list of films about filmmakers trying to make movies. I recommend Living in Oblivion.”
Today, the Internet resembles Bob: it doesn’t understand what it’s looking for. And to gather Alice’s insight, you’d have to search multiple sites and cross-reference it all yourself to generate your own recommendations.
If, however, metadata could be included on Web pages that codified that information and its relationships in a form understandable by machines, then searches could turn up more meaningful results. That is what the semantic Web would look like, and that’s where we get into Web 3.0.
Data about data
“When the information comes back from a query, it contains meaning: the context about where the information came from and how it can be used,” said Hassan Qureshi, senior manager at Ernst & Young Canada. The context is encoded into the information using a system of tags similar to the HTML tags that tell your browser what a Web page should look like. In the case of the semantic Web, however, they’re telling software what information means.
Qureshi said the semantic Web will first evolve inside niche corporate applications. For example, an aerospace company needing to keep track of its vast array of parts, processes and organizational roles might find it useful to encode the information about the parts semantically. This could quickly evolve beyond the capabilities of traditional database software, especially when the information stored is tacit, rather than explicit. A senior manager might write down the experience that she has gained during her career in prose, which is difficult to search. But semantically tagging text relating to key concepts as diverse as customers, regulations and national security could make her experience easier for others to harvest.
If Qureshi is right, it’ll be the first time since the PC that an IT trend has started in the enterprise and moved out into the wider consumer space, rather than the other way around. The blogs, wikis and social media technologies that punctuated Web 2.0 started with consumers, and then moved inside the enterprise as demand rose among employees. The same is true of instant messaging. The challenge comes in making that leap from niche corporate applications to semantic Web services with a larger audience, especially when the audience works in different sectors.
Nigel Shadbolt, professor of artificial intelligence and deputy head of the School of Electronics and Computer Science at the University of Southampton in the U.K., has been using semantic technology as CTO of Garlik. That company is gathering and tagging terabytes of information about individuals and companies in order to give them control over the online information about them, and to help prevent identity theft.
Real-world 3.0
The power of semantic technology will become apparent as those islands of data begin to connect, Shadbolt said. He cites the dbpedia project as an example. This project complements Wikipedia, the poster child of the Web 2.0 era. The organizers are extracting all of the structured data available within Wikipedia (typically the information within the infoboxes at the side of Wikipedia pages) and tagging them with metadata to semantically enable them.
dbpedia is just one of many similar projects that are linking together under the auspices of Linking Opened Data (LOD), a community project created by Berners-Lee’s World Wide Web Consortium (W3C), the governing body for Web standards. The project advocates the publishing of online resources that go beyond simple Web pages and files to encompass a specific entity that could be encoded semantically.
Dissent on the definition
These concepts all underpin an idea of Web 3.0 that is gradually forming in the minds of industry leaders. It is possible to see how they may fit together. For example, if user-owned data could be semantically encoded, it could become just as useful as the data in dbpedia. Users could reveal that data to vendors, and to each other.
Let’s say you’re planning a trip to New York, and you poll members of your online community to see where they’ve been. If I make semantically encoded data about my trip to New York available to you, and you combine it with data from others, you might be able to build your own writer’s walking tour of Manhattan, or a list of establishments that are child friendly, perhaps without those terms ever being explicitly mentioned. You could then expose your planned trip to New York to a selection of vendors, who may offer you discounts on their goods and services.
That’s one possible outcome. But perhaps the ultimate and most controversial characteristic of Web 3.0 is that it will only exist on the pages of magazines like the one you’re reading. The idea of flicking a switch and announcing that Web 3.0 has suddenly arrived is just as meaningful as announcing Web 2.5, or the Web 3.0 alpha version.
Better to speak of a broad evolution of Web technology taking place on multiple fronts which will, thanks to the distributed nature of the Web economy, happen at different speeds and in different ways. The more user driven the Web becomes, the more chaotic and fragmented the whole affair will get. That’s something to look forward to, but arguably not something suitable to label.
HotTech Archive
By Danny BradburyTelecommunications firms have moved from 2G to 3G. Windows moved beyond version 3.0 years ago. And yet, if the marketing community is to be believed, the Web is still stuck at version 2.0. Web 3.0 is being floated as a concept, but does anyone know what it really means?
Publishing mogul Tim O’Reilly, who first sparked the Web 2.0 label with a conference of the same name, has argued that the concept of Web 3.0 is nonsense. He defined the term Web 2.0 not as a milestone for the development of new technology, but rather as a means of pulling the Web industry out of its post-crash economic mire. In short, by his own definition, it was more of a marketing concept than anything else.
“For ‘Web 3.0’ to be meaningful, we’ll need to see a serious discontinuity from the previous generation of technology. That might be another bust and resurgence, or more likely, it will be something qualitatively different,” he said in a 2007 blog post on the subject.
Gartner vice-president David Smith describes what such a departure could look like, defining it across three contexts: technology, community interaction and business. “On the technology side, the biggest thing bar none is the continued emergence of the mobile Web,” he said. “On the community side, the biggest thing is the bringing together of location and community.” With the introduction of GPS-based mobile devices, these two things have gained more synergy. Finally, “on the business side, that’s where a lot of the cloud computing stuff comes in, because a lot of the subscription-based services funded through advertising fit that very well.”
Then there are other definitions. Some think 3.0 will be a more advanced form of the group intelligence inspired by Web 2.0 sites like Wikipedia. Others think that it will be a more ubiquitous and powerful form of application integration than what we saw with the publication of open APIs by Web 2.0 companies like Flickr and Google.
They’re all right, said Adriana Lukas, a social media consultant and founder of the UK-based Big Blog Company. “If you look at each of those technologies, they spell out layers of the Web,” she said. Cloud computing represents the bottom, most technical layer of the Web 3.0 stack, where the infrastructural stuff—the hardware and networks—live. Intelligent applications map onto the higher layer, where software lives.
Web ownership
For Lukas, though, the user is perhaps the defining characteristic of Web 3.0. Web 1.0 was designed around systems. Users who wanted to interact with it had to learn HTML and overcome the limitations of non-dynamic Web sites. Web 2.0 is more user centric in the sense that it enabled users to contribute their own content and document their own experiences. But it was still built by developers, who defined the rules.
“User centric was a step forward, in the sense that you built a system around the user, but the user is still not at the core of the system. They’re not driving it,” she said.
The key to a true user-driven Web is what Lukas and Dan Grigorovici, who chairs a conference on Web 3.0 in New York this May, both call “vendor relationship management” or VRM. VRM inverts the relationship between the user and the service provider typified by traditional customer relationship management (CRM). Today, companies collect data on user activities, often without the users knowing. In a VRM scenario, the users own their data and can make it available to a variety of different service providers to suit their own needs. This will revolutionize the monetary model for the Web, he said.
“One key element is monetizing the data in a different way than just slapping a banner on a Web site,” Grigorovici said, adding that the Web 2.0 concept of mashing up data only takes you so far. “When you monetize that information you won’t just mash it up. You add more value than simply slapping one data set on top of another.
“Let’s say you’re in the mood to buy a sweater but you couldn’t find anything on the Amazon site,” he said, explaining how user-owned data would work. “If you could make available the fact that you’re checking this particular sweater on Amazon or Overstock, then this is information that Wal-Mart would not otherwise have access to. And they might want to give you a discount.”
For users to communicate that, they must own their own data, rather than having it hidden and locked down by companies in their own CRM systems. Lukas goes as far as suggesting that users maintain data in a separate repository of their own, using a system that her team is currently coding, called MINE!
Leaving the dumb Internet behind
Ownership in Web 3.0 isn’t the only thing about the data that will change. According to the inventor of Web 1.0, its structure and meaning will change, too. Tim Berners-Lee developed the protocols underpinning the Web in 1990 when working at the European Organisation for Nuclear Research (CERN). Today, Berners-Lee laments the Web’s one crucial flaw: it doesn’t understand what its own data means. For example, looking up a movie online turns up a variety of predictable results, but it’s an inherently dumb search.
Think about using the Web as if you were having conversations with clerks in two video stores. Bob knows nothing about movies, and could just as easily be working at a grocery store. Alice is a film student earning some spare cash while she finishes her master’s degree in German Expressionism. You ask both of them for a copy of 8 1/2.
Bob: “It’s not in the computer. The next thing on the list is 9 1/2 Weeks. It’s in the erotic entertainment section.”
Alice: “We don’t have 8 1/2, but that’s a Fellini film and it’s a self-referential work about a film director trying to make a movie. I also remember that 8 1/2 won an Oscar for best costume design. If you’re into costumes, then Fellini’s La Dolce Vita might be your best bet, because that won the same award. Alternatively, you might try another movie by 8 1/2’s producer, Angelo Rizzoli. The next movie that he produced was Deserto rosso, Il. Or if you’re into films with a similar subject and don’t care about the director or the period, then here’s a list of films about filmmakers trying to make movies. I recommend Living in Oblivion.”
Today, the Internet resembles Bob: it doesn’t understand what it’s looking for. And to gather Alice’s insight, you’d have to search multiple sites and cross-reference it all yourself to generate your own recommendations.
If, however, metadata could be included on Web pages that codified that information and its relationships in a form understandable by machines, then searches could turn up more meaningful results. That is what the semantic Web would look like, and that’s where we get into Web 3.0.
Data about data
“When the information comes back from a query, it contains meaning: the context about where the information came from and how it can be used,” said Hassan Qureshi, senior manager at Ernst & Young Canada. The context is encoded into the information using a system of tags similar to the HTML tags that tell your browser what a Web page should look like. In the case of the semantic Web, however, they’re telling software what information means.
Qureshi said the semantic Web will first evolve inside niche corporate applications. For example, an aerospace company needing to keep track of its vast array of parts, processes and organizational roles might find it useful to encode the information about the parts semantically. This could quickly evolve beyond the capabilities of traditional database software, especially when the information stored is tacit, rather than explicit. A senior manager might write down the experience that she has gained during her career in prose, which is difficult to search. But semantically tagging text relating to key concepts as diverse as customers, regulations and national security could make her experience easier for others to harvest.
If Qureshi is right, it’ll be the first time since the PC that an IT trend has started in the enterprise and moved out into the wider consumer space, rather than the other way around. The blogs, wikis and social media technologies that punctuated Web 2.0 started with consumers, and then moved inside the enterprise as demand rose among employees. The same is true of instant messaging. The challenge comes in making that leap from niche corporate applications to semantic Web services with a larger audience, especially when the audience works in different sectors.
Nigel Shadbolt, professor of artificial intelligence and deputy head of the School of Electronics and Computer Science at the University of Southampton in the U.K., has been using semantic technology as CTO of Garlik. That company is gathering and tagging terabytes of information about individuals and companies in order to give them control over the online information about them, and to help prevent identity theft.
Real-world 3.0
The power of semantic technology will become apparent as those islands of data begin to connect, Shadbolt said. He cites the dbpedia project as an example. This project complements Wikipedia, the poster child of the Web 2.0 era. The organizers are extracting all of the structured data available within Wikipedia (typically the information within the infoboxes at the side of Wikipedia pages) and tagging them with metadata to semantically enable them.
dbpedia is just one of many similar projects that are linking together under the auspices of Linking Opened Data (LOD), a community project created by Berners-Lee’s World Wide Web Consortium (W3C), the governing body for Web standards. The project advocates the publishing of online resources that go beyond simple Web pages and files to encompass a specific entity that could be encoded semantically.
Dissent on the definition
These concepts all underpin an idea of Web 3.0 that is gradually forming in the minds of industry leaders. It is possible to see how they may fit together. For example, if user-owned data could be semantically encoded, it could become just as useful as the data in dbpedia. Users could reveal that data to vendors, and to each other.
Let’s say you’re planning a trip to New York, and you poll members of your online community to see where they’ve been. If I make semantically encoded data about my trip to New York available to you, and you combine it with data from others, you might be able to build your own writer’s walking tour of Manhattan, or a list of establishments that are child friendly, perhaps without those terms ever being explicitly mentioned. You could then expose your planned trip to New York to a selection of vendors, who may offer you discounts on their goods and services.
That’s one possible outcome. But perhaps the ultimate and most controversial characteristic of Web 3.0 is that it will only exist on the pages of magazines like the one you’re reading. The idea of flicking a switch and announcing that Web 3.0 has suddenly arrived is just as meaningful as announcing Web 2.5, or the Web 3.0 alpha version.
Better to speak of a broad evolution of Web technology taking place on multiple fronts which will, thanks to the distributed nature of the Web economy, happen at different speeds and in different ways. The more user driven the Web becomes, the more chaotic and fragmented the whole affair will get. That’s something to look forward to, but arguably not something suitable to label.
HotTech Archive
