Chris Anderson's book The Long Tail has been a phenomenal success, hitting the top spot on Amazon, so it's hardly surprising that there have also been negative reactions. The Wall Street Journal has just taken a potshot, and has been smart enough to make it one of "today's free features", thus generating a ton of comment around the blogosphere.
In the WSJ, Lee Gomes writes that It May Be a Long Time Before the Long Tail Is Wagging the Web, and the story provides a link to Anderson's rebuttal: The Backlash, Chapter 1 (or Factchecking my ass).
Gomes makes one basic point, which is roughly that Anderson has overstated the economic value of the tail. This may well be true. Anderson's response is, roughly, that Gomes has misunderstood the statistics, which also seems to be true. But the plain fact is that there is no definitive division between the head and the tail, there's just one long powerlaw curve (above), and you can try to prove whatever you like by making an arbitrary division and moving it around. How many angels on the head of a pin, again?
The problem with the argument is that it misses the point. The Long Tail is making an impact because it is a powerful idea that provides us with a new(ish) way of looking at the world. Copernicus did the same thing for many people when he pointed out that the earth went round the sun, not vice versa, though no planetary bodies were physically moved in the process. (Of course, The Long Tail isn't that significant: my point is that it involves a mental rather than a physical shift.)
Whether you can make a profit and/or change the world using the long tail idea is another matter. It's very clearly profitable for Google, because its revenues come from a vast long tail of small text ads not from selling big flashy expensive banners to Nike, Coke, BMW and other major league advertisers. It's clearly of some benefit to iTunes because it can sell more than a million songs that a physical record store would never have room to stock -- but the cost of digitising, storing and listing a song is not zero, so its long tail cannot extend to infinity. Amazon and Netflix are also long tail businesses, with an even greater limitation on the length of the tail, and so on. And, of course, there are plenty of businesses that are not long tail at all....
Now, clearly, there have been long tail business before -- the Sears Roebuck catalogue is an example -- but it's an idea whose time may have come because of the Internet. If you have low or zero origination costs and low or zero distribution costs then you can offer many more products, and that really does mean -- as Anderson claims -- that the cumulative value of lots of slow-selling items (the tail) can be significant, relative to the value of large numbers of sales of a small number of products (the head).
The hits have value -- they always did. But the misses now have a potential value that they didn't have before, because they were too expensive to stock, sell or distribute. Whether that value is 25% or 50% or 125% isn't really worth arguing about, because it's going to different for different products in different industries in different countries at different times, and so on.
The long tail isn't Boyle's Law, it's an observation. Its significance is that it helps people understand the value of companies such as Google, Amazon, eBay, Netflix and so on, and blogs and Wikipedia and other net-based developments. And it's important not to lose sight of the concept while quibbling about the details.
Update: Nicholas Carr is hosting an excellent reply: Lee Gomes responds to Chris Anderson
Update 2: There's also a cartoon...
