The data that describes books comes from many, many places: warehouses, publishers, librarians, book jobbers, consumers. Each of those constituencies uses that data for completely different reasons. But it all converges on the open Web, and can look very, very messy. In an effort to create some alignment in all this chaos, Google, Bing, Yahoo and Yandex have gathered together to create Schema.org, an effort to establish some linked data best practices for the open Web.
Laura Dawson, R.R. Bowker’s product manager of identifiers, in her guest post on the metadata that makes collection development hell.
Read it if you want a crystal-clear understanding of why it's so hard to find what you want to find online, be it Amazon or a library vendor catalog. Yesterday at her Tools of Change panel, Dawson made a point of mentioning Schema.org. I don't quite get it yet, but I am trying.