Monday, September 22, 2008

The User-Agent String: A History of Biblical Proportions

One of the last projects I did for Room 214 was build a tool for tracking what kind of browsers were hitting a site. With a database of several hundred thousand hits spread over multiple content types, I had a lot to work with. Now, every web developer groans when asked about browser detection of the "big boys" (IE, Mozilla, Safari, Opera). But in this case I had hundreds, perhaps even thousands of distinct user agent strings. It was awful. And it made me wonder what the history of these awful nearly-unparseable things was.

On the WebAIM blog, there is an explanation in the form of a Genesis account. And it was good.