not set up by Google.
The biography page was released on 9/9/05 and gets "crawled" for rank & relevance by a search engine spider.
Hate to disappoint y'all, but the word "Failure" is probably in a blog or webpage which links to the white house website. There are probably a gazillion links to
www.whitehouse.gov http://www.google.com/press/newyorker.html The system, which Page called PageRank, permitted Brin and Page to improve on the standard practice of counting how often a key word appears on a Web site. They realized that if a page is linked to many other pages it's like a vote-the collective voice of the Web has decided that the page has a certain value. If millions of people link to a page, it's a good endorsement. It doesn't mean that the link is accurate, but it's likely to be a more useful authority than a page nobody points to. Page and Brin realized that it was possible to map the Web and rate pages primarily by analyzing links instead of words. (In fact, they are so confident of Google's accuracy that they put an "I'm Feeling Lucky" button on their page. Click on it, and you go directly to the highest-ranked site for your search.)
Such searches can require millions of computations, but essentially the rating you get is based on who "voted" for you by establishing links to your site. (The engine also looks at how many votes were cast for the pages that were linked to those pages. If the home page of the Times links to your page, you will be ranked more highly than if, say, just your cousin Harvey links to your home page. That's because many other pages link to the Times, so it brings in lots of votes.) "Before this, people were just looking at the content,'' Motwani told me. "They were completely ignoring the fact that people were going to the effort of putting a link from one page to another and that there must be a meaning to that."
Google is not the first search engine to look at the links on the page; Excite and Lycos have also done it. But Page and Brin's Google has raised the bar. "Their system just works much better than anybody else's does,'' Danny Sullivan, the editor of Search Engine Watch, told me. "Now every major search engine will have to use it. Nobody can afford to do anything less."
Google can be fooled, of course. Anybody who takes the trouble to set up a group of pages with links to each other can force his way into the rankings, with some rather odd results. Brin told me to type in the phrase "more evil than Satan itself," and the first response was the Microsoft home page. (The response just shows that there are many people on the Web who seem to use the words "evil" and "Satan" when referring to Microsoft-and that they tend to link to each other.) "It's not a trick,'' Brin said one evening. "But if you want to just say it's possible to get bad information on Google, I'll understand. It's possible to get bad information anywhere."