The problem is that, anecdotally, no one seems to use Lists. Twitter is filled with users who have carefully made a few lists, and then promptly forgot about them after they realized their clients don’t make it as easy to read List tweets as it is to read tweets from people you follow.
This is why I was never fascinated by Google+ and its concept of Circles.
Holman goes on to say that Facebook has the better idea, in that it can automatically intuit which groups are preferred by the user by using it’s existing networking data.
I call bullshit on this. Take the examples given in Holman’s screen shot: “close friends” (presumably a list), the city he lives in, the university he went to. I don’t know about you, but I have plenty of friends and followers in my city of residence whom aren’t exactly close friends, and I’m hard-pressed to think of hyperlocal content that I’d be interested in posting that would use a city limit as an interesting boundary.
On the other hand, Google+ does one thing that crucially escapes the shit work designation: you don’t organize your friends into circles, you are forced to add them to circles from the get-go. Create a new Google+ acquaintance and you can’t just add them to the master list; you’re forced at the point of sale to make a purchasing decision and categorize them into a circle. This mediates the work of organizing through UI design.
The downside is that this is still a pain in the ass and not particularly useful. Everyone who is in my college fraternity circle has to also get added to a friends or acquaintance circle; everyone who is in my scientific advocacy organization is subject to a brain fart that’s been going on since college where I might accidentally put them into my fraternity.
On the other hand, consider some heuristics that could automate this process. For example, how would you easily define “close friends” in a way that an algorithm could recognize?
- How many emails have I sent this person? Over how long of a period of time? What was their length? How often is this person the sole recipient?
- Did this person have an association with me in the past (i.e., college or fraternity), and have we kept reliably in touch since then?
- What was the size of that associative group? My university had a population of 30,000 students. My fraternity had a population of 20-30. Ergo, if you were in my fraternity, you’re much more likely to be a closer friend than if we wear red and blue colors on the same anniversaries.
Who has all of this data? Google. Who could use data-mining to come up with these suggestions automatically? Google. Who has access, additionally, to my phone records and calendars? You get the point. There’s some additional work to be done here where everyone needs to be incentivized to provide a more thorough curriculum vitae listing all of their associative groups to provide the seed data, but I expect that ice has already been broken by Facebook.
This references an old issue that arose as far back as the 80s, which few people get concerned about: different relationships leave different data trails. You probably consider information about whom you sleep with to be intensely personal, but it’s revealed in your data trail. People who sleep together have different data patterns than people who meet casually for dinner every so often.
This would be a stunning invasion of privacy, but depending upon the implementation, could also be a stunningly useful stunning invasion of privacy. To date, the only heuristic that’s been released into the wild, as best as I can tell, is Gmail’s guess at which email messages are deemed important. It seems to me that it’s only a matter of time before these heuristics become a standard feature of social networking, limited only by the data you’ve shared by the company, and by the social conventions that would lead users to riot in the streets if they’re released indelicately.