A mild-mannered British physicist is trying to render Google irrelevant. Stephen Wolfram, the creator of Mathematica, a grandiosely ambitious piece of software, has come up with Wolfram Alpha, a grandiosely ambitious engine of knowledge.

Grandiosely ambitious, and grandiosely inexplicable. Put simply, Wolfram Alpha, due to launch in May, will "compute" answers to questions, where Google and other search engines merely trawl the Web for pages which might hold the answer.

To do this, Wolfram has had a small army of researchers working on systematically analyzing and structuring the corpus of human knowledge so that a computer might be able to answer questions with concrete answers, such as, "How far will the Earth be from the Sun tomorrow?", a question Google completely fails to answer.


The blogosphere has exploded in a jargongasm, with normally sane reporters pretending that sentences like this are English:

Notably, the engine is not built using standard semantic web languages such as RDF, OWL and Sparql, in part because these ontologies are too difficult to build and curate for such a wide field of knowledge.

Whatever. What's important to note here: Where Google's founders have long assumed that computers were the best tools to "organize the world's information and make it universally accessible," Wolfram thinks that humans can actually do the job.


In the tradition of the great French encyclopédistes of the 18th century, his Wolfram Research has employed in stealth dozens of brainiacs translating specialized databases into machine-computable form. His approach is a riposte to both Google's idolization of algorithms and the fetish for crowdsourcing that swept Silicon Valley in the middle of this decade. Sometimes the best way to get an answer is to ask someone really smart. Like the Wizard of Oz, Wolfram's researchers lie behind the curtain of the answers Wolfram Alpha will provide. How comforting: What the babbling geeks are calling the next great leap in human knowledge actually has a human side.