After reading Ericks’ article and comments on how Google will crush Wolfram|Alpha I thought it would be good to summmarise the discussion as there seem to be a lot of confusion what Wolfram|Alpha is/isn’t.
See the differences in the table below and feel free to adda comment if you think I missed anything.
At the same time I have to say I’m very excited about the Wolfram|Alpha project and I wish it had been available few years ago when I was still at the uni.
|Search Engine||Yes – google is indexing the web and returning list of pages||No – it generates output by doing computations from its own internal knowledge base|
|Free||Yes||Yes, free for personal noncommercial use|
|Audience||Mainstream audience, thanks to its simplicity||Everyone, but given its complexity and it will attract mainly academic audience|
|Link with information source||connects users with information source||connects users with information|
|Database updates||real-time dependent on the indexing frequency||potentially real-time subject to manual data input and approval|
|Typical queries||general interest, shopping||scientific, academic|
|Business model||Mainly Pay Per Click contextual ads||Paid subscriptions with enhanced features for large-scale and commercial use and business partnerships|
|Fundations||1995: Based on a search engine called BackRub, BackRub operated on Stanford servers from 1995||1988: Wolfram|Alpha is built on the Wolfram Mathematica system, which has been in continual development since 1988|
|Limitations||authority based on PageRank, lack of common data structure||Wolfram|Alpha can only know things that are known, and are somehow public. It only deals with facts, not opinions|
|Key strenghts||Simplicity and a wide range of knowledge/data||computation and mathematics|
|Data checks||Mainly automated and managed by Google spam team. But, given its mainstream audience highly affected by spam||Automated and manual, including exper review. Less likely to be targeted by spammers|
|Data upload||No – data needs to be published on the web||Yes – submission process sumilar to wikipedia|
|Private personal information||Yes||No. It contains only information that is considered public.|
|Adult content||Yes, particularly when SafeSearch Filtering is off||No. It has no adult images or narrative. It gives only factual answers to factual questions.|
Many similarities but they are two completely different tools – targeting different audience.
I think Wolfram|Alpha has its right to exist as a niche and reliable tool for research and I’m sure there are businesses who will pay for additional features. What do you think?