Staff suggestion: Wolfram Alpha “Knowledge Engine”
July 14, 2009
Wolfram Alpha – a new “computational knowledge engine” – had been out for a few weeks now and it has taken every bit of that time for it to make even a little sense. When I first heard about it, I immediately pulled it up and tried a few searches. And I got nothing. “What kind of search engine is this?!” I cried.
Lesson: Don’t approach Wolfram Alpha like you do Google or Yahoo.
I think my first search was “woodworking” or something vague like that. I got a definition, pronunciation, and some synonyms. Not quite what I expected. Where were my search results?
But, Wolfram Alpha isn’t a search engine. It’s a “computational knowledge engine” – meaning we’re not going to get google-like results page. We’re going to get data. Wolfram Alpha seeks to pull you from the mire of millions of search results. Because who needs a search results page when you can get actual data?
Wolfram Alpha gives you data, not for a list of websites.
If you want to search for websites to plan your upcoming trip to Canada, you might be barking up the wrong tree.
A search for canada travel gives zero results.
If you’re looking for real data about a city, state, or country, Wolfram Alpha will knock your socks off.
A search for canada gives us the population, flag, satellite image, demographic info, cultural properties (with charts), related links, and a link to source information (so you can see where Wolfram Alpha got all this data).
Example Searches:
Looking for a historical stock price? How about the price of IBM stock in May of 1982. Click on the chart where the red line intersects the blue line and you get a price.
Looking for an overview of the exchange rate for Dollar to Yen? You can view the history of the exchange rate for the last month, last year, last few years…
Would you like to know weather conditions for Nashville on March 28, 2008? You can see results in metric and non-metric. You can get the temperature, rainfall, humidity, and historical temperatures.
Browsing Wolfram Alpha:
Maybe you thought your search string was a good one but Wolfram Alpha gave you zip. What do you do now?
When Wolfram Alpha doesn’t know how to interpret your search, you’ll get a list of category links to explore. Each category gives example searches. This will give you a good idea of the type of information you can pull out of Wolfram Alpha.
Want even more info?
Read this article from PC World.
Subscribe by
July 16, 2009 at 6:34 am
My inner nerd is jumping up and down and giggling right now. This site is awesome!
July 16, 2009 at 11:40 am
The times I’ve tried it I got it to me give the results I wanted, stats and things, and then just for fun I tried Wikipedia and got the same info. Not wikipedia’s will always give the same results, but like many people I feel I don’t know the proper syntax to really get powerful results from Wolfram Alpha. Or maybe the Wolfram Alpha people just want to make us feel that way.
Also, the Wolfram folks claim to own the copyright to any results “knowledge engine” gives you.
July 22, 2009 at 8:09 am
I agree with Melissa. Put in a zip code and got all kinds of cool information all in one place. Although I’d love to know their sources…
July 29, 2009 at 4:36 pm
There is a source link at the bottom of the page, but some of the sources are somewhat vague. I like to choose a variety of sources and know what I am getting as best I can so I can evaluate the data. One of the problems I have had with Wikis is the amount of unsourced material out there. Of course, then it would be a “scholarly text,” not a wiki. And nowadays even academics are using bibliographic essays to avoid directly sourceing their material so what the hey. It just sounds like all information; you have to evaluate it by whatever standards you use and decide how much trust to place in it. You also have to figure out how to extract what you want in the first place.
The copyright thing is baloney (bologna.) Unless, of course, they can demonstrate that their work did not come from other copyrighted work and could not be reasonably expected to be naturally derived from that work. What they do seems to disprove that in and of itself. Good question for our copyright session presenter at the STA meeting this year.