My favorite line in the whole article? This one:
"We need mechanisms to ensure that search gets complemented with serendipity."Yes! That's what I loved about the old card catalogs that libraries used to have. You never found the book you were looking on the first try. You had to go past books that you weren't looking for but that might still catch your eye and show you something that you weren't looking for. But that seldom happens with search engines. When Google first came out, you would always end up with lots of unexpected results, but that doesn't seem to be the case much anymore [*]. You can't just browse with Google either, just to see what is out there on the web.
Zuckerman also talks about forcing physical structures to create serendipity, such as an effort in the Harvard libraries to reorganize the books by size instead of the standard subject classification. Nice idea, but with Google scanning every book in site, that won't do much good, will it?
Such physical structures already exist to some degree with the technical journals. Even very specialized titles have a breadth that no one reads an entire journal cover-to-cover. We all pick and choose what interests us, but there still is the chance to run across something unexpected. That's why I still look through Nature and Science whenever I can. As broad as their coverage is, there is something that is outside your scope.
Maybe we can hope that Google, with their extensive knowledge of what we already search for, can create a serendipity button, with an algorithm that directs you away from what you normally search. Or at least slips in the occasional mickey.
[*] Either their search algorithms have improved greatly, or maybe I've gotten better at using Google. Either way, I seldom get off on fun tangents anymore. And all those dumb commercials for Bing (where they have people ranting irrelevant responses to a simple question) leave me cold.
No comments:
Post a Comment