Ever found yourself painstakingly trying many combinations of dates to find the cheapest flight ? TechCrunch reports on FlySpy, a tool that will make that process painless :

The way it works is that I give it a departure city and a destination city and optionally a departure date and length of stay. The search result, which returns very quickly, will present me with a graph of flight prices over the next 30 days so that I can quickly look at which days are the cheapest to fly. To book a flight I just click on the point in the graph. Simple.

Jason Kottke envisions taking the idea even further by using historical data to predict the optimal date of purchase. He also suggest applying it to all industries where yield management is relevant.

Is transparent pricing coming to the air travel market ? Probably not even in your wildest dreams as Keith Devlin explained in 2002 :

Faced with all this confusion, with computers constantly monitoring sales and adjusting fares as often as ten times a day, the only real option for the fare conscious air traveler is to use a Web service to try to locate the best deal. [..] But just how well do those search engines do ? Not very, is the answer. And with good reason. Airline pricing has grown so complex that it is now practically impossible to design an algorithm that will find the cheapest fare. In mathematical terms, the (idealized) problem of finding [..] the lowest fare is NP hard [..] This is the perhaps surprising result obtained recently by mathematician Carl de Marcken.

Although depressing, this piece of research highlights an interesting aspect of FlySpy : it does not try to find the solution to the problem. Instead it follows the patch of scientific visualisation : when confronted with overwhelming amounts of data, the best way to understand it is to draw a nice picture.