ALKHOBAR: Yesterday, Wolfram Alpha LLC announced the general availability of Wolfram Alpha, the world’s first computational knowledge engine, offered for free on the Web at www.wolframalpha.com. Wolfram Alpha draws on the innovative work of scientist Stephen Wolfram, particularly his technical computing software platform Mathematica. According to Wolfram Alpha over 200,000 people from throughout the world have contacted the company to learn more about the computational knowledge engine since news of the service first surfaced broadly in March.
The long-term goal of Wolfram Alpha is to make all systematic knowledge immediately computable and accessible to everyone. Wolfram Alpha draws on multiple terabytes of curated data and synthesizes it into entirely new combinations and presentations. The service answers questions, solves equations, cross-references data types, projects future behaviors, and more. Wolfram Alpha’s example pages and gallery show a few of the many uses of this new technology.
“Fifty years ago,” said Stephen Wolfram, the founder and CEO of Wolfram Research, “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. Wolfram Alpha signals a new paradigm for using computers and the Web.”
Arab News was invited to attend an hour-long webinar hosted by Stephen Wolfram to preview the new service and was given preview access to the computational knowledge engine. What was immediately clear is that the difference between Wolfram Alpha and search services such as Google, is that Wolfram Alpha computes new knowledge based on true facts rather than just presenting available information.
Wolfram Alpha is made up of four main “pillars” or components: Curated Data, Dynamic Computation, Intuitive Language Understanding and Computational Aesthetics. Wolfram Alpha contains terabytes of factual data covering a wide range of fields. Teams of subject-matter experts and researchers collect and curate data, transforming it into computable forms that can be understood and operated on by computer algorithms. When Wolfram Alpha receives a user query, it extracts the relevant facts from its stored computable data and then applies a collection of tens of thousands of algorithms, creating and synthesizing new relevant knowledge.
To allow Wolfram Alpha to understand inputs entered in everyday language, currently English only, its developers have examined the ways people express ideas within fields and subject matters and they are continually refining algorithms that automatically recognize these patterns. In presenting the results of its computations, Wolfram Alpha creates customized graphics, tables, charts, etc. – unique to the query.
Wolfram Alpha has been entirely developed and deployed using Wolfram Research Inc’s Mathematica technology. Wolfram Alpha contains nearly six million lines of Mathematica code, authored and maintained in Wolfram Workbench. In its launch configuration, Wolfram Alpha is running Mathematica on about 10,000 processor cores distributed among five colocation facilities, using gridMathematica-based parallelism. And every query that comes into the system is served with webMathematica.
“Wolfram Alpha is an extremely powerful way of harnessing the world’s knowledge. Now, anyone with web access can tap into that knowledge to find relevant information and discover new insights,” said Theodore Gray, co-founder of Wolfram Research.
It must be pointed out that Wolfram Alpha builds on the achievements of science and other systematizations of knowledge to provide a single source that can be relied on by everyone for definitive answers to factual queries. This means that it can compute the shape and slope of a cubic meter of dry sand but it can’t determine who the most beautiful woman in the world might be.
Wolfram Alpha relies on factual sources for its data and most of them seem to come from Western developed nations. This means that it can easily provide information on an input regarding a group of US presidents, even when the query contains only last names, but it can’t similarly identify Middle Eastern leaders. Sometimes its “facts” are actually incorrect, such as when it listed the full name of King Abdullah bin Abdulaziz Al-Saud as King Abdullah. Wolfram Alpha didn’t know the birthplace of Jalal Talabani or Hosni Mubarak — even though this information is widely available through online search.
The company is looking for experts and databases to expand its “facts” in many areas, so with time such deficiencies will likely be overcome. What isn’t clear however is how Wolfram Alpha will handle issues regarding disputed information such as certain national borders for instance. And even though Wolfram Alpha “knew” that Makkah, Saudi Arabia was the city required, probably based on my Internet IP address, when asked the distance between Dammam and Makkah, the computational knowledge engine didn’t offer up information on Association Football when the query “football team” was entered. Instead it only provided information concerning American football teams.
Stephen Wolfram believes that students and certain professions will be early avid users of Wolfram Alpha with some users turning to the service to do the same computation again and again. The basic query service provided by Wolfram Alpha will be free but there will be a professional offering available for a fee that will allow more computational time. On some complex queries the free service’s CPU processing time may not be enough. Companies will also be able to sign up for a Wolfram Alpha solution that will enable corporate users to do computations using that company’s proprietary data.
For now, advanced mathematics students will enjoy watching Wolfram Alpha solve their homework problems. Just input the equation and press enter. Engineers will find it a useful tool. The service’s ability to present data in graphical terms makes it extremely useful in handling comparisons. Too often though, the service responds to queries with the reply: “Wolfram Alpha isn’t sure what to do with your input.” This means that in areas outside the fields of math and science, we’ll have to use our own brains to provide life’s answers — at least for a little while longer.
