Web 3.0: The Semantic Web at Work
Ben Rothfeld, global marketing strategy director at Acxiom Digital, a division of the Acxiom integrated marketing consulting firm, replied to the first mention of the phrase "enterprise portal" by saying: "I have a sticky note on my desk that says Personal Dashboard, who will own it?" Back in 1998, when he was lead account planner for Answerthink, then a custom application developer, Rothfeld posed the question, "What do you prefer to do? Go through a portal like Yahoo!, or would you rather just have a single box on an empty page where you could simply type what you were looking for?" By essentially predicting the supremacy of Google back when it was scarcely a household name, Rothfeld has a track record for prescience.
Rothfeld introduces the notion of Web 3.0, saying, "If Web 2.0 was about making it easier to put up a webpage, then Web 3.0 makes everyone a data analyst. For instance, I want to look at what a particular stock does each Tuesday at a certain time. The web is not currently structured to do that."
The main problem with getting personalization to work is due to limitations in the way the data is stored and tagged. A frequent question Forde receives from customers is, "Why can’t your search work like Google?" The answer, Forde explains, lies often in a lack of semantics for the content. "The customer who asked this question most recently works at a firm that sells 401(k)s. So when he searches for his personal 401(k) material, he has first to look through 10 documents that come from the marketing department." Jarlath continues that the search works, but the underlying data needs to be tagged and structured.
A company that promises to help companies leapfrog from standard taxonomies to autotagging and the use of semantics is Thomson Reuters. The company has deployed a new system based on technology from the Open Calais Project, an open source project funded by Thomson Reuters that builds implementations to create semantic autotags for content. According to Forde, when faced with a great deal of legacy content Thomson Reuters wanted to access, the company "started to think of its product as an ingredient" and sought to blend the old content with the new in a searchable manner. Forde explained that any piece of data sent into the application interface will be automatically tagged and can then be loaded into the database. Reuters merged its robust taxonomy with the semantic approach of Open Calais and is among the beacons that may make the semantic web a reality. Cheung believes that portals can begin to take on a role of "operating systems for the web"; with this sort of technology, that promise is one step closer.
The notion of the enterprise portal that held sway in its infancy may appear to have outlived its usefulness—given the evolution of the web. However, that very evolution promises a personalized way for workers to access what they want and need to know to get their jobs done. The difference is that today, a top-line view (the view visible through a portal, if you will) does not suffice. Workers need a webwide view in order to get the job done.
Companies Features in This Article:
Goodman & Co.
Open Calais Project