Computational knowledge engineWolfram Alpha has just become more accessible from a variety of applications. Wolfram Alpha has thrown open access to its API to all developers with version 2.0 of its Wolfram Alpha API, allowing you to integrate its results into a variety of sources of web, desktop, enterprise and mobile applications.
The new release, in addition to opening up free access to its API, also brings features like asynchronous operations, that allows certain results to be delivered earlier compared to the ones that are time consuming. A Java client library has also been added in this release.
To get started with the API, you need to sign up for a free development account. This gets you an Application ID that you need to provide in each of your calls. The documentation is comprehensive and includes an overview of how the Wolfram Alpha engine delivers its results. The API Is REST style and the result data format defaults to XML. However you can request the results in the response XML to be one of plain text, HTML, image and much more depending on a concept called “Pod," which is a rectangular portion of the output and corresponds to one category of the product.
This is a very interesting way of returning the results and you can determine which Pod content to then display in your application as needed. As an example, visit the standard Wolfram Alpha results for Pi and you will see several representations which can be referred to as Pods. An example call of the API where we query for “pi” and want the format to be returned in image and plaintext, is as follows:
It also has an API Explorer, where you can get a better understanding of the results that are returned. The rate limits are pretty stringent at this point in time and this could be an issue for applications that might drive large traffic to its API. The free development account is allowed a maximum of 2,000 calls per month. Helper libraries are available for several languages.
The Wolfram Alpha computation engine is unique in many ways and its API too requires you to understand how it returns results before you can integrate but it can surely help create much richer search experience. The API page highlights featured applications like Microsoft Bing, Touch Press and Mathematica and how they have used the API. Not only that, but the biggest user of the API seems to be Wolfram Alpha itself: Wolfram|Alpha Mobile and Tablet Apps for iPhone, iPod touch, iPad, and Android devices; the newly released Wolfram Course Assistant Apps; and Mathematica 8.
What do you think of the new Wolfram Alpha API? In which cases might you use it for your search results?