Saturday, May 12, 2007

Just finished Probability Space by Nancy Kress, the third book in a trilogy, following Probability Moon and Probability Sun.

Aside from the silly titles, these are all excellent books. One nice thing about them is that they are all stand alone books - each one is a complete story with no cliff-hanger endings. The third one is a little less stand alone and is my least favourite. The first two are centered around the characters dealing with a strange new culture and trying to figure out what is causing it's unusual features. Probability Space deals with the consequences of what happens in the first two books and wraps up the over-arching plot line. It feels more like a standard "space opera" without the cultural exploration angle and suffers because of it.

Amazon Link: Probability Space

Thursday, May 10, 2007

I just finished The Long Tail by Chris Anderson. This is really an extension of an article Anderson wrote in Wired about this subject back in 2004.

The basic idea of the long tail is simple - in a market where the cost of carrying additional items has dropped to near zero, a significant aggregate amount of demand exists for items with a small amount of individual demand. The initial examples came from Rhapsody and Amazon. For a standard store like Tower Records, there is a physical limit to how much merchandise can be carried and therefore there is competition between products based on which one can produce the most revenue. The result is that records stores carry product heavily weighted towards hits and that affects what we can buy.

In contrast, there is almost no cost for Rhapsody or Amazon to add another product and therefore they can carry products that do not contribute a lot of revenue. The first result is that Rhapsody and Amazon have a much larger selection. The second result is that if you sum up all the small contributions of the marginal revenue products, you end up with a substantial amount - sometimes 20-40% of total revenues.

If you plot the revenues for each individual product, you get a power law like in the picture at the top of this post. In the past, retailers focused on the left hand side of the line because there was a limit on how much product they could carry. But now, more and more companies are finding that there is a lot of money to be made in the right hand side of that line.

I read Anderson's original article years ago and liked it. But I avoided reading this book because I had heard it was mostly redundant and Anderson over-stretched himself by trying to use the long tail concept more than is appropriate. Both of those criticisms are partially true, but the book is still worth reading. There is repetition but Anderson also adds enough new details about where the long tail economics apply and where they don't and the history of how companies learned to exploit it. Near the end, he does apply it a little more widely than I was comfortable with that is just a small section of the book.

If you are interested in why companies like Amazon, eBay and Google are having such a profound effect on the modern economy, this book is highly recommended.

Amazon Link: The Long Tail