Something doesn't seem right with those numbers. Or maybe I am reading it wrong. The total internet audience (in the US) has only grown by ~30% since 1996? That seems awfully low. My guess is that the 115M number for 1996 is just too high. That's almost half of the US population. No way was half of the United States on the internet in 1996.

Also, according to those stats, more than 10 million people accessed well.com in 1996. That seems incredibly high.

Anyone else agree?

Yes, seems skewed. The other part that seems skewed is the MSN sites. You know that ESPN takes the majority of that ranking.

Please Disney, break the ties to Microsoft.

I'd be willing to bet that ESPN's contribution to MSN's ranking is actually smaller than you might think. Remember, MSN is the default homepage for the other 95% of computer users, most of whom don't bother/know how to change it.

I thought it was interesting to see how high the .edu domains were in '96. Now, not a single .edu domain in the 2004 survey. I'd be interested to see when that transition (within a period of, say, six months) happened -- the transition from the original Internet of geeks and tinkerers to the circa 1999 bizdeved version we have now.

This will seem a terribly naive question, but how are the visitors counted? If I visit ESPN.com at least once every day (and perhaps twice a day, spaced by more than a couple of hours), am I counted in MSN's total 400 to 800 times? Seems more logical than the "Internet population" being 115 million people eight years ago.