On December 31, 2003, Google filed a patent application for “Information Retrieval Based on Historical Data.”

On March 31, 2005, it was put online and by April Fool’s Day (a coincidence?) it was discovered, and the SEO forums are ablaze with discussion about this Google patent.

Not just theory, the patent application seems to explain things that are currently going on with Google, including the “sandbox” filter, which is looking less and less like a theory.

Greg Boser calls it “The most important SEO related document in the last 5 years.” I’d agree. Not since Page and Brin’s PageRank paper has there been a document with more details about the factors and philosophies behind Google’s algorithm.

In a nutshell, you’ve always heard that there were “more than 100 factors” in the Google algorithm. Now, they’ve laid out for the world what some of those might be.

And there’s something for everyone. For example, the Google cynics love the part about tracking surfers’ bookmarks (through the Google Toolbar? From the desktop search application??)

I’m currently working on an article with my specific “what to do” suggestions. In the meantime, you should read the Google patent. Not easy reading, but the pictures are sure simple. Here’s one sample:

Also Rand Fishkin has already written a great article that covers all 63 claims in the actual Google patent application. Enjoy.