Hacker News Reader

Comment by jyothi | Hacker News Reader

You might want to consider youtube traffic there, which is #4 on alexa rank. This will push the cross over by few more years for even statistical predictions. http://www.alexa.com/siteinfo/google.com+facebook.com+youtub...

Few other interesting things to note of the alexa graph is:

- Facebook crossed youtube traffic just about an year ago.

- Google crossed yahoo traffic only late 2008, that too mostly for yahoo traffic went down hill.

- What caused the sudden rise in internet usage in oct 2009 (for all top sites sites except yahoo show a spike in traffic). Can't be early christmas traffic, guess facebook triggered this high adaption of users who spilled onto other sites too.

View on HN

It's not so easy to undouble between youtube and google, my guess is that since youtube is a google property that it actually is a subset, so it would not make a whole lot of difference. Only those people in the youtube set that are not yet in the google set would skew that graph and hence the conclusion.

After all, this is about 'reach' and 'uniques' which translates in to people not pageviews.

True the overlap might be very high and it is extremely hard to find the common userbase with the available open tools. Will have to wait for hitwise, comscore or someone to publish a report.

I happened to look more into the global top sites http://www.alexa.com/topsites/global. It is lot more confusing - Google's India, German, HK, UK sites are considered separately. A clear example is http://www.alexa.com/topsites/countries/SA and http://www.alexa.com/topsites/countries/Italy where the top site is not google.com, so these users are not accounted in .com but use other google properties or variant country TLDs.

The reason why I looked in to this (this time) was because I was wondering if the whole privacy backlash had put a dent in facebook's growth.

It seems that's not the case, they're still growing at more or less the same rate (to my surprise).

I wouldn't mind analyzing a days worth of logs of either company, but I don't think I have a machine large enough to store the files on (nor enough CPU to process the data).

It would be quite the goldmine of information.

Youtube vids are embedded everywhere. I'd bet a load of people view youtube vids, that may not use anything else 'google'.

I don't know if counting 'embeds' makes much sense though.

Reply on news.ycombinator.com