I’ve just read @sookio’s blog entry on Why Twitter’s Who To Follow is missing the point. I half agree, because so far, the feature has half worked for me. Sue’s not the only one to have suggested the feature is useless, and I can see why that would be a common reaction – didn’t I already follow and unfollow that guy? Meh.
On my personal account, Who To Follow has thrown up a bunch of comedians and other minor slebs, most of whom I had already tired of and unfollowed months ago. I don’t think much of a lot of the people suggested to me – self-promoting truculent whitterers without anything interesting to say who conduct conversations with each other over the heads of all those who follow them at worst; just plain unengaging at best. The people I follow are funny, diverse and interesting, day in and day out, and you’ve never heard of most of them before. I do follow well known people on Twitter, but they engage, reply, retweet, argue – they get it.
But it’s not all bad. On @bigwide, another account related to my studies and interest in development issues, Who To Follow provided a stack of interesting suggestions. It did exactly the same thing – recommended new people based on the number of people I follow who follow them. The difference this time was that considering the subject matter of the account, I didn’t have to like these suggestions, but I did need to follow them. Also, my comparatively shallow level of engagement with the other account meant that I had not yet tapped all of the potential contacts out there by a long shot. I am a relative newcomer in a separate community.
So, Who To Follow fails on my personal account because the number of people in my community following someone is not an indicator of their usefulness or likeability to me. Who To Follow is like an overenthusiastic party host, taking wild guesses who I’ll like. Twitter’s no meritocracy if you measure followability by numbers of followers, and Twitter’s algorithm clearly can’t see that, but it just wants to make sure that everyone at the party has mingled. Being based on numbers, it’s also biased against less well-known members of the ‘community’, as Sue points out. To continue butchering the metaphor, if Who To Follow were a real-life party host it would insist on introducing you to the loud and lively character in the corner everyone else had spoken to, while ignoring the poor soul who had just walked in, hung up their coat and wandered what the hell all the fuss was about. Retweets, lists and follow Fridays fortunately do the job of the guests at the party who have enough knowledge to introduce you to people they think you’ll like, while Who To Follow is sprinting around with a tray of drinks and laughing hysterically. Had I just walked in to the party, I would still have appreciated being introduced to people, if only so I knew to avoid them for the rest of the night.
The point is, if Twitter’s suggestions don’t look much use to you, it’s probably because you’ve already met all the people at the party you like, and you’re hanging out in the kitchen, getting a bit drunk and laughing at the host. If that really interesting person walks in looking a bit nervous, give them a drink and make them comfortable, because the host can’t even remember who’s met who any more.

Also here