WolframAlpha searching goes public

The new WolframAlpha "computational knowledge engine" went public over the weekend with its British-born creator, Stephen Wolfram, hailing it as a "a new paradigm for using the computer and the web."

However, he stressed that it is designed to complement rather than rival search engines such as Google. And in an introductory video he says it is "an ambitious long-range project that in many ways is just getting started."

WolframAlpha answers questions from its own store of knowledge, curated by Wolfram staff, rather than trawling and indexing web pages like Google. The knowledge base is therefore relatively limited and it has been criticised as too US-oriented.

But knowledge stores can be expanded and WolframAlpha certainly does things that Google-style search engines do not. It will, for instance, solve equations and do calculations thanks to its use of Wolfram's highly-successful Mathematica maths engine.

This also allows it to make relationships between information. A query " temperature UK 1940 to 2007" brings up an average temperature (47 degrees Fahrenheit, 8 degrees Centigrade) and a graph of temperatures over the period.

Clicking for more information gives you humidity, pressure and wind-speed information over the same period. It also lists what it calls weather station comparisons, providing minimum, maximum and average temperature values.

But the weather stations are identified only by code numbers and their positions are given by distance, bearing and elevation without saying what the reference point is, making the information meaningless.

WolframAlpha said today that it had already processed 13.7 million queries, including some for instant information on a small earthquake in California over the weekend.

Wolfram said in a press statement: "Fifty years ago, when computers were young, people assumed that they'd be able to ask a computer any factual question, and have it compute the answer. I'm happy to say that we've successfully built a system that delivers knowledge from a simple input field, giving access to a huge system, with trillions of pieces of curated data and millions of lines of algorithms."