The Long Tail has become a fundamental belief of a large core of the internet, mainly because of the way that it contrasts retail and online methods of selling media.
To sum up its most relevant message, as the cost of storing and distributing things like music and video on the internet approaches zero the more likely you are to make a profit from the 'long tail' of your distribution curve.
This varies from the retail model because the cost of manufacturing the disks is constant whether it be a week after launch or five years later. Shelf space and storage also retain a value and when these are all tied together its easy to see that at some point in time the product's sales will no longer cover its distribution costs.
For a digital only store its fair to say that as costs of distribution approach zero the value of product in the long tail is as likely to be as great or greater than the head.
Seems straightforward and quite logical?
Well this article in the Times would suggest not. But looking at their research I'd have to say that the authors are either wrong or at least making an irrelevant point.
For example, look at the iTunes Music Store. Here the top sellers are invariably recently released tracks which are in the head part of the distribution graph. But do those few hundred new tracks outsell the rest of the catalogue (several million songs) which make up the long tail? I doubt it very much.
I guess the point I'm making is that selling one copy of a track a month (in the long tail) will be very profitable if you have enough tracks, no costs and an extended time to sell them.