Reading about what Siri will do is like reading Berners-Lee’s early descriptions of the Semantic Web, which is about the highest compliment you can give to new software. Siri, which will be launched in public beta in the first half of 2009, is described by its developers as virtual personal-assistant, a “do-engine” instead of a search engine. Leveraging semantic technologies developed for the military, this personal assistant will be take over online tasks, such as making restaurant reservations or handling travel plans, using learned user preferences to act in the user’s place.

If, for example, you're planning a trip to Atlanta, instead of searching the Web for entertainment options, just ask Siri what there is to do in the ATL.

"Siri pulls information relevant to the user's question from a variety of Web services and tools,” Siri co-founder Adam Cheyer said in an interview with Technology Review.

“In this case, it checks the weather, event listings, and directories of local attractions and uses machine learning to select certain options based on the user's past preferences" said Cheyer. "Siri can connect to various Web applications and then integrate the results into a single response."

Instead of searching for flights to Atlanta, just give Siri the dates and it will find them for you, get your car and hotel and find the best restaurants in the area you’re staying in.

The program is based on Cognitive Assistant that Learns and Organizes (CALO) architecture, a project funded by the Department of Defense’s Defense Advanced Research Projects Agency (DARPA). It is the biggest AI project in U.S. history.

CALO’s goal is “is to create cognitive software systems, that is, systems that can reason, learn from experience, be told what to do, explain what they are doing, reflect on their experience, and respond robustly to surprise.”

Named one of Technology Review’s 10 Emerging Technologies of 2009, Siri should be one of the first Semantic applications to hit the market.

By Mark Alvarez