An interesting piece of technology made it onto the web this week. I’ve thought for a while that Google employees must have a fair amount of fun watching what the world is searching for, and they are probably not the only ones. Being levelling types, though, as well as masters of the slick and simple web interface, the nice chaps at Google have decided that we should all be able to have a look. Put a search term into Google’s latest web page, and it shows you how many people have been conducting searches similar (identical?) to that one. You can compare two searches by putting a comma between them, which is a nice feature. It’s clearly still in a preliminary stage, and I’m not quite sure what those links on the front page are about - they seem quite out of date.
What kinds of things can you find out from it? Well, for example, popularity of this attempt at humour (enter the term French Military Victories into Google and hit “I’m feeling Lucky” or select the top “hit” for a fake results page) seems to have peaked in late 2005. As for why that should have been, I’m at a loss, since I’m sure one of the earlier blips marks the time when it first hit the web, and it was hardly very, very funny. Interestingly, it also seems to have coincided with renewed interest in “Freedom Fries”, so anti-French feeling in the English- and (who knew?) Swedish- and Finnish-speaking internet seems have resulted in more than one statistical blip around then. Material for a Master’s Thesis anyone?
If the data being used comes from searches, I can’t know what people are searching for and how many people are searching for it with accuracy both at the same time, can I? A sort of digital equivalent of the Heisenberg uncertainty principle!
But Google, Google, Google! Did your maths teacher never tell you not to draw graphs without scales on them? Mine certainly told me never to trust them.