Delving into the data
We’ve always spent a lot of time looking at the universe of books (new, used, out of print, international) and the way they’re categorized, distributed, and sold, so we can figure out how best to serve the community of book shoppers online. Here are some pointers to what we’ve been reading.
I’ve enjoyed reading tech columnist Glenn Fleishman’s analysis of commercial book-related metadata issues, and his two latest blog posts deal with issues that we’ve been addressing since we launched BookFinder.com: managing multiple editions, and brainstorming clustering strategies. He also points to an interesting piece on book-related metadata in the Boston Globe.
Wired editor and Long-Tail-Guy Chris Anderson has been trying to figure out the size of the “long tail” for books in the US. He took a stab at it on his blog, but after some further back of the envelope calculations, figures that the long tail constitutes 15% of book sales today. I think even his new numbers are a bit conservative, but his discussion and methodology are worth reading.