The evolution of cooperation

I hope readers of this blog will indulge a bit more of my current library obsession. I don't really plan to make a career out of this, but it continues to raise interesting issues.

On the geek front, both Derek Robinson (who did the fascinating in-page-editing experiment formerly found here) and Art Rhyno wrote to point out that the 508-character limit on bookmarklets in IE can be circumvented. Art's demonstration:

javascript:void((function() {var element=document.createElement('script'); element.setAttribute('src',''); document.body.appendChild(element)})())


Several folks have noted that the kind of integration we're talking about here is a two-way street. My original idea was to grab an Amazon ISBN and look it up in a local library. Prompted by a librarian's request to move from CARLweb to Amazon, Matt Croydon demonstrated that the reverse lookup is also feasible. In fact, as Matt immediately then realized, you can in principle connect anything ISBN-related to anything else ISBN-related.

The ISBN is the key enabler here, and there are questions about its efficacy. They surfaced again when Randy Grow wrote to his local library in Palo Alto to ask about LibraryLookup. Martha Walters answered, and I quote her with permission:

The problem with ISBN searches and libraries is that the ISBN you are commonly searching is for one format of the title, in the case of the example in the article, for a paperback copy, which libraries are not likely to buy. If you do a title search for the same book, you will get mulitiple copies in different format/editions and the like. So, many folks will get no hits, but the library will, indeed, have the item. The ISBN in the article is a perfect example. There are no hits at most of the libraries, because they don't purchase trade paperbacks. In order to truly search your local library, it would be neccessary to use the "All Editions" link on Amazon and then try every edition in your local library's catalog. You may be surprised how many paperback editions libraries do not have.

I knew that an ISBN denotes a species, not a genus, but hadn't really thought through the implications. One is a search strategy which I've also added to the tips section of the newly-reorganized project homepage: try hardcover editions first. Another is that, book titles being non-unique, there is no universal identifier for the genus. I wonder if there's a need for a Web service to solve this. Given an ISBN (for, say, a mass market paperback), it would map from species to genus, collect all species in the genus, and return a list of ISBNS (paperback, hardcover, audiocassette, etc.)

Should Amazon implement such a service? We live in such interesting times! I can imagine good arguments both pro and con. A book that is helping me to think about them, which I read about on All Consuming and bought on Amazon, is Robert Axelrod's The Evolution of Cooperation. 1 The central theme is the Prisoner's Dilemma, and that's just what may now be shaping up. Consider this further comment from Martha Walters:

Another thing to be aware of is that the iPac product has the capacity to link directly to Amazon or Barnes & Noble holdings, but most libraries are not enabling that "feature" since bookstores are seen (sometimes) as our competitors. I think, personally, that we could benefit by being Amazon associates (or whatever they are called) and getting a percentage of the money our patrons spend on books, especially in the highly literate and educated geographical area we live in.

Viewing LibraryLookup through the lens of the Prisoner's Dilemma, we could say that a cooperator exposes a simple, low-coordination-cost API, and a defector withholds one. I think libraries would be nuts not to cooperate, but there's also a perceived payoff in defecting. The central question, for Axelrod, is: How do stable cooperative strategies evolve, over time, as the game iterates? The answer to that may determine the future of Web services more than any of the XML standards we love to debate.

1 Until the book arrived on my doorstep, I had no idea it was published in 1984. I love the way that All Consuming abolishes newness as a criterion of interest. Ideas like the ones in this book are timeless. It makes perfect sense that in a community of bloggers, a two-decade-old book about strategies of cooperation would, suddenly, become a hot seller.

Former URL: