I discovered this wonderful statistics-gathering Perl script via the equally excellent Rands in Repose blog. It gathers data about your Twitter activity for general analysis which is incredibly geeky but, to one such as myself, also fantastically cool.
Since Twitter is rapidly becoming my preferred method of remote communication, I simply had to try this out. Unfortunately the default script package was designed to use Apple’s Numbers spreadsheet software which I don’t own so I used the variant script found in the comments for the original script post to use Google instead and came up with the following:
Tweets per hour
Tweets per day
Tweets per month
So some of the data is a little strange-looking primarily because up through August I was working a grave shift and since as you can see from the daily chart I do most of my Twittering during work hours, I have some activity in the wee hours of the morning. However, I’ve begun Twittering much more earnestly in the last couple of months, mostly due to me finally convincing several friends to join, which means the activity seems heavily weighted toward the regular daylight hours.
You’ll note a fairly fitful start at the beginning of the year. I’ve noticed that there seems to be kind of an adjustment period when people first get on the site: You update for a few days, forget about it for a week or so, remember again and make a single update and so on. Until at last there comes this sort of moment of clarity where you either get the critical mass of friends and acquaintances involved enough that it becomes a real means of communication (as opposed to an interesting toy) or you become so enamored with the idea of following the activities of other people in tiny bite-sized morsels delivered throughout the day that it just “sticks.” You’ll note that October was the moment of clarity for me and the dip in November was a sad side effect of me being dumb and forgetting that I can update from my phone while we were in Seattle visiting Fast-Track.
Incidentally, it would have been awesome to have Twittered from Seattle and I sincerely regret my lapse; I lamented the oversight on Twitter hours after we got back.
The script also tracks the ‘@reply’ usage as well, but since it took me the better part of the year to get anyone on the site that I know well enough to reply to, my stats in that category are dull and unimpressive (hence the omission). It’s also misleading because I’ve noticed that with Nik I tend to use direct messages (‘d DixieGirl’ for example) when I need to speak to her directly versus ‘@DixieGirl’ which everyone can see. However I tend to have semi-public conversations with Ryan so my top ‘@reply’ listing is @corvock even though I communicate with Nik via Twitter at least 3-to-1 compared with Ryan. But obviously these stats can’t collect info on the direct messages which aren’t publicly visible (that’s sort of the point).
What I find most interesting about all this is how it took Twitter to really make me see the benefit of text messaging, but because of it I now have a pretty steady stream of messages coming into (and out of) my phone. For sanity’s sake and also for the sake of my cell phone bill I’ve had to limit the people who’s tweets update directly on my phone to Nik and Scott, but I sincerely regret not being hip to what’s happening with Red, Gin, Whimsy, Lister, Ryan and the rest simply because I can’t afford to have 25 messages per day.
However, with Twitter facilitating so much of my daily communication now and with discoveries like the sublime Oh, Don’t Forget, I may have to simply call AT&T and crank up my text message plan to the next level and just be done with it.