Semantic Web, Web2.0, Web3.0 …

The past few months I have been puzzled with the message of some publications regarding the Semantic Web (or better “Semantic Technologies”), Web2.0, Web3.0, and in particular with interpreting what this means for uptake of Semantic Technologies in the market.

The publications in question are:

• Entrepreneurs See a Web Guided by Common Sense, John Markoff, New York Times: November 12, 2006 (free registration required to access NYT)



• Hype Cycle for Web Technologies, 2006, publication date: 7 July 2006 (not freely available)



• Several publications related to the success of Web2.0 in 2006 (blogs, youtube, wikipedia, flickr, etc.).

At first sight, it seems that Web2.0 is receiving the media attention that previously was given to the Semantic Web. In the figure below, which is generated with Google Trends, the blue line represents the Semantic Web and the red line Web2.0. The top part refers to web pages mentioning the terms; the bottom part refers to news appearance.

In the 2006 edition of Gartner’s Hype Curve, the Public Semantic Web is at the through of disillusionment and is estimated to take between 5 and 10 years to reach the plateau of productivity. Web2.0 and the Corporate Semantic Web, on the other hand, are at the Peak of Inflated Expectations, the former reaching its plateau in an estimated 2-5 years, and the latter 5-10 years.



One reason for these phenomena is that Semantic Technologies build up an infrastructure on top of which the web can grow to its full potential, whereas Web2.0 is related to communities of final users (e.g. http://www.youtube.com/, http://www.wikipedia.org/). One consequence of this is that Web2.0 is much more visible than Semantic Technologies, and thus it is much easier to attract attention of a large audience. Semantic Technologies is supposed to create an (invisible) infrastructure, whereas Web2.0 creates highly visible applications for final users using the existing web, but with a significantly improved user interface (thanks to AJAX). This would also explain why there is disappointment with the Public Semantic Web (the large public hasn’t seen anything yet); whereas the Corporate Semantic Web is still peaking. Businesses see a huge benefit of, for example, better access to important corporate information in unstructured documents (Mike Lynch, CEO and Founder of Autonomy, estimates that 80% of corporate information is hidden in unstructured documents). Businesses are not interested in final user applications, but in promising technology that can help them to better manage their information.
For me, one of the most striking achievements of Web2.0 is its capacity to involve large active user communities. Indeed, this has been a (happy) surprise. At the beginning of the Semantic Web effort, many of us were convinced that it would never be possible to annotate web content with (semantic) tags through human effort. Therefore, much research effort was dedicated to automatically generating annotations using advanced Natural Language Processing techniques. Web2.0 initiatives have shown this assumption to be false; it is possible to tag large amounts of multimedia and text documents by communities of people.

Recently the term Web3.0 is getting popular, which refers –freely interpreted- to the combination of Web2.0 and the Semantic Web.

My take on this is that we will hear much from Semantic Technologies in the context of corporations. The Public Semantic Web will regain strength in combination with Web2.0 aspects, strengthening the notion of Web3.0.

Happy 2007!