The number of users is irrelevant when measuring user engagement, at least when it is defined as the article describes. (But comparing early app.net versus a m

The number of users is irrelevant when measuring user engagement, at least when it is defined as the article describes. (But comparing early app.net versus a more mature Twitter may be relevant.) To calculate user engagement they should take the number of daily posts and divide by the number of users (so with these numbers, for Twitter it is 350 million/150 million = 2.33 and for app.net it is 10,000/[# of app.net users...20,000 according to the study linked to by diego] = 0.5. This isn't necessarily the best metric since somebody could easily write a bot that makes several posts per second, which could skew things considerably. (I don't think anyone cares about how active peoples' bots are.) It is quite possible that Twitter has a similar situation where most tweets come from a smaller number of users.