I think time spent might be a better measure though - e.g. I can fire off 5 tweets in a couple of minutes, then never think about Twitter from the rest of the day, or I can spend an hour crafting a single tweet, reading replies and other feeds, and repeatedly check for new responses. I'd say the latter is much more akin to obsessive behaviour, but has a much lower tweets/day average.
I don't think you can measure it in absolute numbers, though; as far as I know, "obsessive behavior" is defined by factors such as the negative effects on the rest of your life.
And that's the problem with this comparison. SMS are not used like tweets, they're used like IM messages. So 250 SMS per day should be compared to e.g. 250 Telegram messages per day, not to 250 tweets per day.
I can easily imagine reddit users or even HN users making those kinds of daily contributions.
> During the first two weeks of February, he tweeted, on average, more than 15 times a day. He refrained completely from tweeting on only five days—all on weekends. That’s far from obsessive, but it’s even farther from “unplugged.” It is, in fact the opposite of “unplugged.”
More than 15 tweets per day is far from obsessive? I'm not much of a Twitter user, but that does seem like a lot.