Google has some new beta offering called “Google Web Accelerator”:http://webaccelerator.google.com/ which promises to speed your web browsing experience. Sounds, great, right?
Conceptually, the accelerator isn’t all that different from the web caches used by a lot of ISPs and large corporations. Google basically keeps up to date copies of popular web pages and then transfers them to you using their superior servers and connectivity.
If widely used, this could mitigate “The Slashdot Effect” which often brings small sites to their knees when they suddenly find themselves featured on that popular website.
Which actually brings up another point. Often times, people will add a link to a version of the page from the Google Cache to the early comments of a Slashdot thread so that people can read the article even if the site has been swamped by the sudden burst of traffic. The accelerator is in some ways like an upgrade to the google cache.
Google also uses some other tricks, like anticipating what users might want to see and downloading the page before they click on it. Also, to save bandwidth (their’s probably, more than your’) if a page changes that you already have on your PC, it will only send the part of the page that has changed.
When I started this entry, I’d planned to post about how it was a fools bargain to turn over so much information to Google about browsing habits, but then it occured to me, they’ve long been collecting pretty much the same information, at a lot lower cost to themselves, from any user with the Google toolbar installed with the PageRank indicator turned on.
Of course, I’ve not had PageRank turned on for years. The value of the little guage in the toolbar is far less than the intrusion of them tracking my browsing behavior. This would seem to be an attempt to try to strike a new bargain with users to get at that same data.