« The financial crisis, as if told by Leonard Bernstein | Main | I won't touch that except to say that I'd have paid a lot to see Wolfram and Jacques Derrida go one-on-one »

This looks very cool, but what do I know? The WolframAlpha rollout

WolframAlpha went online a few days ago.

“So what?,” you say — same as I would have said before watching this introduction. A “computable” knowledge source like WolframAlpha looks much more useful than a simple look-it-up like Wikipedia or listing of links like Google, though the functions of the three do differ to some degree.

If you’ve spent any time taking WolframAlpha for a test drive, please let the rest of us know what you think. Is the hype warranted?

Comments

I think the hype is warranted. If we’ve learned anything from the Wiki/Web 2.0 era, it’s that masses of people with deep knowledge about narrow interests are willing to put in serious time perfecting information systems related to their interests.

[In fact, it would be an interesting survey to find out some motivations behind this behavior, but I digress.]

Let’s be clear, W|A is not currently more useful than other info-finding tools, but the “computational” aspect hints at something that could be very, very big. I mean, given enough datasets (with their attendant narrow-interest wiki watchdog groups), there is no reason to suppose that W|A won’t be able to compute something like “What should I get John Doe for his birthday?”. If John Doe self-reports his preferences on his facebook, his blog, or whatever, there is no reason to suppose that W|A can’t aggregate and compute this data.

Of course, we’re not there yet, and there are lots of clumsy, bad responses coming up on my W|A searches, but I think it has potential.

Lee,

Took it for test drive yesterday, and here are my thoughts - http://www.drewconway.com/zia/?p=717

It doesn’t even recognize the string “nash equilibrium.” I thought it was supposed to be integrating all systematic knowledge, doing any math you damn well please, yadda yadda yadda etc.?