Newsletter Archives

• The Making of a Fly: feedback loop gone wild

  Posted on April 27, 2011 at 14:06 CDT by woody • Comment in the Forums

  Anybody who’s heard a guitar screech at a rock concert knows about positive feedback loops.

  Now you can watch one in real-time. Just don’t reach for your pocketbook.

  UC Berkeley evolutionary biologist Michael Eisen has a fascinating blog post about Peter Lawrence’s The Making of a Fly, a paperback book published nearly 20 years ago, that discusses the molecular biology and genetics of Drosophila fly egg fertilization, differentiation, and ultimate emergence of a full-fledged fly. No, the feedback loop isn’t in the fly. It’s in the price of the book.

  Eisen followed the listing for The Making of a Fly on Amazon, keeping track of its price. The book originally sold for $32.95, but it’s out of print. When Eisen found it on Amazon, the cheapest “New” copy he could find cost $ 1,730,045.91. Yes, you read that correctly.

  At first I thought it was a joke – a graduate student with too much time on their hands. But there were TWO new copies for sale, each be[ing] offered for well over a million dollars. And the two sellers seemed not only legit, but fairly big time (over 8,000 and 125,000 ratings in the last year respectively). The prices looked random – suggesting they were set by a computer. But how did they get so out of whack?

  When Eisen looked at the prices the next day, both of the prices had gone up. As the day marched on, so did the prices. First, one of the sellers would raise the price, then the other. Some very astute observation and a little bit of Excel revealed that two booksellers, bordeebook and profnath, were locked in an apparently-automated feedback loop.

  Once a day profnath set their price to be 0.9983 times bordeebook’s price. The prices would remain close for several hours, until bordeebook “noticed” profnath’s change and elevated their price to 1.270589 times profnath’s higher price. The pattern continued perfectly for the next week.

  Eisen goes on to speculate that bordeebook has a copy of the book, and they always want to undersell profnath by a small percentage. They thus peg their price, algorithmically, at 99.83% of profnath’s price. He further speculates that profnath doesn’t have the book, but they have a high sales ranking, and they can always buy the book and re-sell it, so they put their price at 27.059% more than bordeebook’s.

  Eisen watched the price go up to $23,698,655.93.

  The booksellers finally broke the feedback loop.

  But, alas, somebody ultimately noticed. The price peaked on April 18th, but on April 19th profnath’s price dropped to $106.23, and bordeebook soon followed suit to the predictable $106.23 * 1.27059 = $134.97.

  Did they learn their lesson? Apparently not.

  As I write this, on April 26, the Amazon site only lists one new copy of The Making of a Fly. Bordeebooks has it available at a whopping $976.98. Since there’s only one new copy listed, bordeebooks must have a feedback loop based on some other criteria, and it’s still grinding away, raising the price of the book 820% in two weeks.

  In addition, Amazon lists eight used copies of the book. One of them, from bordeebooks, is shown as “Used – Good” condition. It sells for… wait for it… $976.98, the same price as the new copy. A second used copy, from seller quality7 (do any of these guys capitalize their names?) in “Used – Very Good” condition will set you back a paltry $999.

  The moral of the story: if your company adjusts prices dynamically, based on competitor’s prices, now would be a good time to make sure there’s some sense built into the feedback loop.

  Or maybe you should consider writing a book about fly molecular biology.

  Other positive feedback loop, The Making of a Fly