What Will Replace Web Search Engines?

Currently the primary way most people use the web to answer questions or do online research starts with a search engine such as Google or Yahoo!. However since before the web was invented people have dreamed of a machine which could compute answers to questions posed in natural language such as, 'How many weather stations in North America are reporting above average temperatures today?' With current search engine technology, this requires doing keyword searches and individually sifting through large lists of results filtering out the irrelevant, or suspect matches. One is left hoping serendipity and Google's proprietary algorithm quickly lead to a document that holds the answer you need embedded in it someplace. In the example above, actually computing the final answer might require collating readings found from each weather station in North America! If the answer to a question requires synthesizing information from more than one source that can rarely be done automatically and requires human intellect. What we want is to just ask the question and get a straight answer instead a list of documents that might have the answer.

The Semantic Web, a proposal originating from World Wide Web inventor Tim Berners Lee, is one attempt currently underway to move in this direction by augmenting content and services on the Web with semantic information that describes what kind of information that content actually contains in a format that can be understood and processed by machines themselves. A simple example would be tagging a line of text as one ingredient in a recipe for chocolate cake. The eventual goal is the elusive 'intelligent agent' which is a piece of software that goes out and uses these semantic tags to intelligently find just what you were looking for. Many are skeptical that this approach will work in the end but it has gained the most traction as a way to evolve the web into something with more possibilities for entirely new or at least more efficient uses of the Internet.

However, Steven Wolfram, inventor of Mathematica and author of 'A New Kind of Science', may be leap frogging ahead of the Semantic Web efforts with a project he is about to unveil known as Wolfram Alpha (no hubris in that name!). Here is a detailed blog post describing it titled, "Wolfram Alpha is Coming -- and It Could be as Important as Google". Hopefully I have included enough background in this post to help understand that article.

Cheers,
Barry Tolnas