It was a different world back when I first ventured onto the internet. There was a definite feeling that it was something apart from one’s normal life; a world in which nobody knew you were a dog. I also recall a sense of danger. The cautionary voices of the older generation outnumbered the enthusiastic calls of the more adventurous, and those with actual experience to draw on were all but non-existent. It seemed natural, then, to insist on an anonymous identity.
Things have moved on, and crucially the positive effects of one’s online life to real life are much more visible, whether you are making friends on Twitter or networking on LinkedIn. More and more I’m getting to the point where I don’t want to use anything but my real name for an online identity. Apart from anything else, hours of my life have now been spent sitting at account creation screens trying forlornly to find a name that is amusing enough to make me seem fun, serious enough to be seen by colleagues and above all else not already taken. My Twitter username, @timmartin2, is a sad testament to the eventual abandonment of this process.
My Twitter username is also a good illustration of what might be a real problem with internet applications. My first and last name when combined together are enough to uniquely identify myself among everyone I’ve ever spoken to—not just almost everyone, but literally everyone. Even for people with much more common names than mine, genuine duplicate names are quite rarely encountered in practice. Crucially, rather than try and pre-empt any problems of name uniqueness at birth, the preferred human solution has been to muddle through until a conflict occurs and then solve it ad hoc.
The design assumptions underlying internet technology seem to let us down here. The easiest way to deal with this uniqueness issue in designing a software solution is simply to declare that names must be unique and let the user figure it out. This isn’t just slightly easier but is orders of magnitude easier. The problem of identifying someone with a handle that is locally-unique but globally-duplicated hasn’t been solved, but then it hasn’t really been tackled. We’ve just been happy to let the technological tail wag the usability dog, as with so many usability issues.
I wonder whether the reasons for this are not wholly technical, but also partly born out of accidental properties of the early internet. The anonymity and the domain-name goldrush of the first tech boom have set in our minds the model that identities must be unique in a way that doesn’t serve us well in the era of the more social web. The Facebook URL land-grab of recent months seems like a good example of where the ordinary user is losing out.
One important result of socialising pseudonymously is that it changes your behaviour. Trolling is rare in real life, and mostly restricted to children who haven’t yet learned the importance of keeping a good reputation. Sites like Stack Overflow and Hacker News do well because they tilt the balance of payoff away from the troll and allow people to benefit from a positive reputation, but this still doesn’t spill over from the site into other facets of life. It would be impractical (and probably counterproductive) to attempt to port human social structures directly onto the web, but maybe real life still has a few tricks left to teach.