People often believe that internet traffic from desktop applications isn’t included in statistics such as these, when in fact Hitwise’s system is capable of measuring all internet traffic (not just world wide web traffic – although I suspect it’s limited to HTTP).

That doesn’t mean that other twitter traffic is included – their report says they did not include all third party applications, although due to the way that many of the twitter APIs work, I suspect that a significant amount of this traffic was not counted.

It’s also worth mentioning that I use Facebook chat via my instant messaging client – so I am effectively on facebook almost constantly – although I only visit the site about once a month, so I disagree that “Facebook is very much a web destination” in comparison to twitter.

As mentioned in their most recent Audit from PricewaterhouseCoopers, Hitwise haven’t always been perfect at responding to queries about their measurement methodologies (although their methodologies got flying colours under Audit), and I haven’t contacted them – so I can’t be 100% certain in this situation.

what’s your twitter handle so we can have a discussion about this?

Seriously- to what extent do you think details of page views/time spent on twitter etc… are altered by people using desktop apps like tweetie/tweeetdeck etc… which are a necessary tool to make sense of the unfilitered randomness of Twitter (in my experience) Facebook is very much a web destination, whereas Twitter is something that has spread out onto numerous mobile platforms and as a result a large amount of that engagement must slip the notice of webtrends and tracking tools.