A few days ago, Wolfram Research released a new "knowledge engine" called Wolfram|Alpha. You might be familiar with Wolfram if you used Mathematica in any of your math classes. Notice how they don’t call it a search engine – the knowledge engine is designed for giving you back information, rather than a website that might have that data. Google sort of does this already, as you can see here, here, here, and here.

Wolfram|Alpha tries to take this one step further. Interestingly, you "compute" your result rather than search for it. Wolfram|Alpha aims to use what’s called natural language processing to figure out what you’re trying to ask, and then feed that to its algorithms in a meaningful way. When you compute your search, the first thing you see is how WA interpreted your input, and any assumptions it made. For example, if you search for "1 dollar", you can see that it assumed you’re talking about the US dollar, not the Australian dollar. If you search for "US population", you can see its interptretation of your query, and you can see tons of information. You can get information on a lot of neat topics, and can try inputs like "weather Minneapolis 2008", "earthquakes near San Francisco", or "2 slices of swiss cheese."

Now, why do you care about WA? Try a search like "caffeine vs. aspirin", or "iron". Now we’re starting to get a little bit more information that might actually be useful. You can do any sort of calculation you want with it, like this one, this one, this one, or even this one, if you know how you use Mathematica syntax. If you’re a lit major, queries like "Isaac Asimov" or "The Grapes of Wrath" might not be too helpful, but it does seem to work nicely as a dictionary / thesaurus. This is sort of to be expected, since literature and other liberal arts don’t rely on computation and mathematics all that much.

So you can get all of this data. So what? You can combine all of that data and the mathematical capabilities of WA, and get results like this, or you can compare snack choices, this. It’s sort of odd at first, but you’re trying to compare two (or more!) things quantitatively, WA can help you.

All in all, it seems like a neat tool. It still has some kinks to work out – try asking it a question in plain English, chances are it’ll have a hard time figuring out exactly what you want – but it’s got some potential. Right now, it’s more fun to talk about what it can do than what it can’t do, and we barely scratched the surface here! They’ve got a list of topics, and are continually expanding. You might even find it to be a valuable tool next semester, making your research, homework, and studying a bit easier. For a more thorough introduction, check out their screencast!