Re: comments on Wolfram Alpha

*To*: mathgroup at smc.vnet.net*Subject*: [mg100344] Re: comments on Wolfram Alpha*From*: Bill Rowe <readnews at sbcglobal.net>*Date*: Mon, 1 Jun 2009 07:10:38 -0400 (EDT)

On 5/31/09 at 6:37 AM, beliavsky at aol.com (Beliavsky) wrote: >input: arima >output: Arima,Trinidad and Tobago >Population: >city population | 34997 people >my comment: Nothing about ARIMA time series models, which comes up as >the 3rd link on Google. Wolfram|Alpha is clearly not Google and is clearly not intended as a Google replacement. Since Google is not doing anything to determine the meaning of what you type as a search term, there is no difficulty in returning links to quite different results. All that is needed for Google to do its thing is that others have used the same search term to link to the things you are interested in. But Wolfram|Alpha is trying to parse your input and extract meaning. Obviously, when you input a single term as you did above, there is essentially nothing to give context and provide meaning. And when it could be as either an acronym or a city name, Wolfram|Alpha clearly must make a choice in order to return any result. Additionally, not all possible meanings for a given term will be available to Wolfram|Alpha yet and may never be available. Wolfram|Alpha is dependent on what is in the databases it can access. It is likely impossible for it to associate a term with a meaning not currently in one of the databases it can access. >Wolfram Alpha appears much less useful than the combination of Google >and Mathematica, so I wonder if resources are better spent adding >functionality to Mathematica. The degree to which Wolfram|Alpha is less useful than the combination of Mathematica and Google depends on the type of information you are looking for and the type of questions you pose. And this will certainly change as Wolfram|Alpha evolves.