Web 3.0 = Semantic Web?
Forward Links
Intelligent web ... technically doable!
Alright ... here goes the bad news
Yes, it may sound a little weird; but World Wide Web is getting marked with versions these days! Web 1.0 refers to the version of the web which had clear line of separation between producers and consumers of web content. Producers were the organizations which used to build websites and consumers were the people who used to “browse” those websites using a browser[1]. Web 1.0 created eCommerce opportunities which in turn gave birth to one of the greatest booms in US history – DotCom. DotCom bubble burst earlier this century has then been followed by another boom called Web 2.0. There are striking similarities and subtle differences between the two booms; which we have already discussed in the last post. Web 2.0 essentially removed the line of separation between producers and consumers of web content and fostered active participation of web users by transforming the web into a platform for collaborations[2]. And now there are winds of next generation web i.e. Web 3.0!
| Semantic web, where machines can read web pages and extract useful information using cognitive decision-making ability emulating human beings, has been conceived by many as the next generation web[3]. Will the vision of semantic web be realized into the next generation of the web i.e. Web 3.0? Or will it remain as just another great but inconceivable dream? That’s the precise question of the moment! |
Web inventor's dream
Tim Berners-Lee, the inventor of World Wide Web, has expressed his vision[4] of web as a universal medium for knowledge exchange as follows,
| “I have a dream for the Web [in which computers] become capable of analyzing all the data on the Web – the content, links, and transactions between people and computers. A ‘Semantic Web’, which should make this possible, has yet to emerge, but when it does, the day-to-day mechanisms of trade, bureaucracy and our daily lives will be handled by machines talking to machines. The ‘intelligent agents’ people have touted for ages will finally materialize.” [5] |
Exactly opposite approaches
There have been two exactly opposite approaches followed to empower machines to extract information from the web and relieve human beings from spending hours searching and browsing web pages. The top-down approach is to develop smart software agents which can extract information from existing web content format. There are already such (partially fulfilling) smart agents being provided by different companies e.g. BlueOrganizer by AdaptiveBlue, PersonalWeb by Claria etc. Second approach is the bottom-up approach where web content itself is made readable for machines. This approach aims to convert the web from “collection of documents” to “collection of data”; and this is what precisely called as Semantic web (or intelligent web).
Intelligent web ... technically doable!
Current standard to represent information over the web, HTML, is incapable of defining relationship among pieces of data; rather it provides presentation information and document level assertions. A layer of semantic information needs to be added over existing web content to make it accessible by machines. Web Ontology Language (OWL) is a language to classify web content using a standard hierarchical categorization scheme. Resource Description Framework (RDF) is a data modeling framework that can be considered as a knowledge representation format. It can be used to provide meta-data for web content i.e. to represent information about the information over web. SPARQL Protocol and RDF Query Language (SPARQL) is a RDF query language. It will facilitate extracting information from the web where content is represented using RDF. Together, these technologies will annotate web content with meta-data and then help to extract required information by identifying nature of the content. This meta-data has to be added by publishers of the content or alternatively, it could be a responsibility of users of the content or it could be provided by a third party like some rating boards.
So far so good … this is definitely a great vision which has the power to revolutionize the way World Wide Web perceived today; moreover it seems technically doable! Now let’s consider a hard, real world, million dollar question … does it seem a practically achievable dream?
Well, there is good and bad news!
Alright ... here goes the bad news
The bad news says that this approach certainly poses some important problems in terms of feasibility. It is hard to come up with a description (or ontology) language which can capture all human comprehensible semantics. Secondly providing semantics to content in a manner consistent across the web is harder to realize. If it becomes a responsibility of content publishers or third party rating boards then they could get over burdened. On the other hand, if it is passed into the hands of users (like tagging in Web 2.0), then consistency and reliability is at the stake. Then does it mean that the vision of Semantic Web is far from being realized?
Then comes good news!
Not really because the good news is that, despite of the hurdles posed on the way of realizing this great vision, there is remarkable adaptation seen from many reputed names in both – industry and academia. Google has financed a project called KnowItAll which is undertaken by groups of faculty members and students at University of Washington. Opine is a sample system built over it for extraction and aggregation of information from reviews posted by users about a product. Yahoo’s food section search has been improved using semantic technology. Citigroup is heading an initiative to use semantic technology to organize and correlate content from various financial data feeds to help identify capital-market investment opportunities. Oracle incorporated Siderean's Seamark Navigator system to find more relevant information from database through semantic search. HP is providing open source toolkit to create semantic web applications. Metaweb Technologies is developing a searchable and editable structured database like Wikipedia called Freebase but using semantic technology. RadarNetworks is developing a semantic search engine. A trial project has been undertaken at Massachusetts General Hospital and Harvard University to annotate clinical data using semantic tools to make it more easily and accurately searchable and sharable among researchers. Doug Lenat of Cycorp is building an intelligent system using web technologies which he claims would answer questions asked in written or spoken natural language!
Web 3.0 = Semantic Web?
Development of technologies and tools to make the web a more powerful information resource along with encouraging adaptation of those technologies definitely emitting encouraging signs about realization of this great vision. Nevertheless, the question “when that would be realized?” has not yet been answered. Complete realization of Semantic Web is not in eye-sight at this moment; it will take some more years to promise an answer in terms of realization timeline. Web 3.0 would not be a completely realized Semantic Web; rather it will be a blend of Web 2.0 technologies with some power of semantic technologies … and we have to wait for few more versions before getting rid of searching and browsing ... i.e. to realization of Semantic Web dream!
| They Said It! |
| “We are going from a Web of connected documents to a Web of connected data”, Nova Spivack, CEO Radar Networks. “We're just at the start, what we can do with search today is very primitive", Daniel Waterhouse, Partner, 3i Venture Capital firm. “What we're trying to do with the Semantic Web is build a digital Aristotle. We want to take the Web and make it more like a database, make it a system that can answer questions, not just get a pile of documents that might hold an answer”, Mark Greaves, Former Information Exploitation Office program manager, US DARPA. “The thing being called Web 3.0 is an important subset of the Semantic Web vision, It's really a realization that a little bit of Semantic Web stuff with what's called Web 2.0 is a tremendously powerful technology”, Jim Hendler, Professor, Computer Science, RPI. |
[1] User interaction with the web was supported in Web 1.0; but it is different from facilitating users to provide web content that could be browsed by many others.
[2] Web 2.0 is the era of applications like social networking, online media sharing etc where content is provided by users of it.
[3] Along with semantic web, other possible candidates claiming the crown of Web 3.0 are Pervasive Web, 3D Web, and Media Centric Web.
[4] Tim Berners-Lee on the Semantic Web: www.technologyreview.com/video/semantic
[5] http://en.wikipedia.org/wiki/Semantic_Web
Technology Business Discussions
- Open source or proprietary?
- Is One Laptop Per Child really a solution?
- What’s business model, precisely?
- Blocking ads - have your cake and eat it too?
- What's more important - idea or execution?
- Would you suggest a fresher to join a startup?
- Do online social networks indeed connect people?
- Is working for a startup your cup of tea?
- What an entreprenur wants?
- Do you need anonymity over the internet?
- Are you really safe on the web?
- Why do you wanna be an entrepreneur?
- Are software patents evil?
- Can business schools build entrepreneurs?
- Free = open source?
- Who is entrepreneur, exactly?
- Why are we here?
Alex (Comment this)
I agree with the first part - "getting relevant information faster is important" ... no doubt about it. Semantic web will precisely realize the same.
Second part, that I am not completely agree with - "letting machines solve complex problem is less important". I think distributed computing is the main motivation behind computer networking and information sharing is a by-product. So according to me "letting machines solve complex problem together" is equally important with "getting relevant information faster" as far as computer networks are concerned.
Anyways, thanks for the appreciation! (Comment this)