<img attachment="wolfram-alpha-home.png" align="left" class="frame">Wolfram Alpha aims to save you a bunch of clicks when searching information online. You should probably check out the <a href="http://www.wolframalpha.com/screencast/introducingwolframalpha.html" source="" author="Stephen Wolfram">screencast</a> (131/2 minutes) in order to get really charged up. The reality is that it <i>kind of</i> works like it does in the screencast. I started off by requesting some information about the town in which I grew up, namely its population. That worked just fine, but it couldn't find historical information, so no fancy graphs for me. So, I chose a larger nearby city and it did just fine. I chose my current city<fn> and that returned about the same overview I'd get from a stop at the Wikipedia page (right-hand sidebar). That was pretty cool, because that search took me a few Google hits and a Wikipedia trudge when I had to do it without Wolfram Alpha a few weeks ago. I thought some more about burning questions about data I needed correlated, so I tried <a href="http://www97.wolframalpha.com/input/?i=average+age+US+president" source="" author="">"average age US president"</a>, which was too specific; I was offered <a href="http://www97.wolframalpha.com/input/?i=average+age+US" source="" author="">"average age US"</a> or <a href="http://www97.wolframalpha.com/input/?i=US+president" source="" author="">"US President"</a> instead, neither one of which Google couldn't do for me, though admittedly without the fancy charts. Those weren't, however, the questions I wanted answered. Age, in fact, seems to be missing, as queries for "youngest president", "age US presidents" all failed to produce any information or flashy charts. They will undoubtedly address this as they amass more data about the types of questions people are likely to ask. The web is full of amateur genealogists, so correlation over DOB information is not likely to be missing for long. Wolfram Alpha is, at least to some degree, ready for geeks, as a query for <a href="http://www97.wolframalpha.com/input/?i=What+is+the+air-speed+velocity+of+an+unladen+swallow%3F" source="" author="">"What is the air-speed velocity of an unladen swallow?"</a> produced the usable response of <c>11m/s</c>. As presented in the screencasts, the hard sciences are well represented, but questions involving more squishy topics don't even elicit a half-hearted link to potentially relevant information in Wikipedia or default to a Google search (<a href="http://www97.wolframalpha.com/input/?i=When+will+the+world+end%3F" source="" author="">"When will the world end?"</a> linked neither the upcoming Emmerich movie nor any information about Ragnarok or Judgement Day). When you return to things numeric, however, things are much better and this is where Wolfram can really save you a lot of time. Trying to stay in the spirit of the screencast, I asked for <a href="http://www17.wolframalpha.com/input/?i=compound+interest+5%25+20+years+1000+capital" source="" author="">"compound interest 5% 20 years 1000 capital"</a> and Alpha asked if I meant <a href="http://www17.wolframalpha.com/input/?i=5%25+20+years+1000+capital" source="" author="">"5% 20 years 1000 capital"</a> which, in fact, I did. A query for <a href="http://www97.wolframalpha.com/input/?i=UBS+stock+value+2006-2008" source="" author="">"UBS stock value 2006-2008"</a> timed out but a query without date restrictions returned the expected data. As expected, there is a tremendous amount of caching required for this kind of search engine, and you're kind of expected to jump in at the highest level of abstraction and drill down to give the server half a chance. That's bad news for those of us who like to skip steps but the silver lining is that Alpha does give it the old college try, indicating that this is likely more a server capacity limitation rather than a logistical or data-based one. You do, however, have to include the right kind of numbers for now, as queries for "iphone sales 2008" or "teen pregnancy NYC" (then "pregnancy 18-24 NYC ") returned nothing at all. Trying to find out how much Apple gadgets weigh with "iPhone weight" also gave no results. At this point, I'd killed my second Wolfram server<fn>. Socioeconomic searches really don't work at all, even for things where you would imagine they should have data (e.g. "US incarceration rate" or "hispanic, caucasian, african-american" turn up absolutely nothing) so you're stuck going to Wikipedia for anything but the simplest of searches. It'll probably be a hit with kids looking for a way to spend even less time researching their boring homework assignments than they do now. For the rest of us, it looks like we should stick to the searches outlined in the screencast for best results ... like comparing your name to that of your partner to find out who's cooler.<fn> <hr> <ft>Wolfram does not know what country codes are for those of you living outside the US, so be prepared to type out the whole name. Steve may have a British accent in the screencast but, as with almost every other tool on the web, Alpha is going to work for Americans best and for the rest of us eventually.</ft> <ft>You can see how it shunts you to a server when you perform a search. If you kill the server you're on (like I did with poor little <c>www97</c> and <c>www22</c> in quick succession), just get rid of the number in the address and hit enter to get a new server.</ft> <ft>It turns out the wife's name had its heyday in the 50's and mine is an up-and-comer.</ft>