Marshall Brain's Blog

---

Google user rankings

I started using the Google toolbar because of a friend of mine. He called it one of the most useful pieces of software he's ever tried. It has 3 features that I use constantly:

• The Popup ad blocker (which keeps count of the total number of popup ads it has blocked, and is currently indicating 540 blocked ads)
• The Blogger tool, which makes it easy to add articles to a blog
• The search tool, which lets you search Google as you normally would, but also allows you to search within the domain of the current page. So if you are looking at a page on CNN, you can search for something else just on CNN

The toolbar is always there, and I use it many times every day.



When you download the Google toolbar, there is an information dialog that pops up during the installation process. This dialog is named "Choose your configuration", and your choice is to enable or disable the "advanced features" of the toolbar. The dialog says, "In order to show you more information about a site, the Google toolbar has to tell us what site you're visiting, which it does by sending us the URL. This does NOT tell us who YOU are (your name or email address), but does tell us that a user has requested the PageRank for a given site."

Let's assume that several million people are using the Google toolbar. I would imagine that many of them have the advanced features turned on, because the PageRank information is useful. This would imply that Google has a sample set of millions of people sending in all the URLs they look at every single day.

It would be fascinating if Google would grind through all of this URL information and produce rankings. These rankings would be identical to the book rankings on Amazon. Each page, as well as each domain, could have a rank based on the number of people visiting it each day. Google could do several different things with this information:

• When Google displays the search results, it could also display each page's (and domain's) rank.
• Google could create lists of the top 1,000 pages and the top 1,000 domains.
• On any given day, there are pages or sites that would be drawing a large number of hits. Google could detect these "buzz pages" and make a list of the top 1,000 so that people could go see what's up.

In theory, Google would be able to provide other statistics. For example, where does a site's traffic come from? Since Google gathers the URL information per user and in chronological order, Google would know how much of Site X's traffic comes from Site Y and could publish that.

All of this information would be a valuable public service that is similar to the PageRank information that Google already provides. Counting the number of hits that a page or domain gets does not seem like it would be computationally expensive. It would give Google users another measure of the value of a page.