Quote:
|
However, there may be other new calculations added since then, based on context, age of the link, etc.
|
According to patents that mention PageRank, the basic mathematical formula hasn't changed. But I expect it to have a few possible modifiers:
1. If a Google is in the middle of a PageRank iteration and discovers a new page, instead of restarting the iteration it takes an educated guess.
2. If Google doesn't trust the link (e.g. if its in a sitewide footer with keyword-loaded anchor text, exchanged link footprint is obvious), the PageRank may be devalued or nullified completely. Proof? When Big Daddy was released there was a surge of supplemental results complaints. That was due to Google modifying its crawling/indexing behavior (i.e. pickier indexing using minimum PageRank threshold) and becoming more aggressive with link devaluation.
2a. If Google thinks a page is selling links, it loses its ability to pass PageRank, or so I've heard.
3. I remember Bill talking about a patent about Future PageRank
BTW, internal PageRank is likely a float between 0 and 1. So people who take a linking page's TBPR and divide it by the number of links on the page to figure out how much PageRank a link might pass is really barking up the wrong tree.