When designing social software, I imagine one of the most challenging things is finding the incentive to interact. Answering the question: How and why should I use this service? I haven’t really designed anything like this (on a serious scale, at least) so I wouldn’t know from personal experience. But being an avid user at least allows me to see the complexity that is there.

I’ve been thinking about this when using Foursquare lately. As I remember it, Foursquare launched as a service to find out where your friends were and what they were doing. Privacy is of course an issue with these sorts of things and it was structured as a social network where you add your friends. If you don’t want people to see where you are, don’t add them as friends. The Facebook-model, if you will – a symmetrical relationship between two people (if I am their friend, they have to be my friend too).

Over time, Foursquare seems to have left that idea a little bit and moved over to become a way to find new places in a city. You can filter by seeing where your friends have been – not necessarily where they are. That little detail makes all the difference, for me. From a privacy perspective, I’m not sure I want people to know where I am – but I’m generally fine with telling people where I have been. Every person has their own rules, I suppose.

Since the intent of the service is now changed, I would much prefer a Foursquare that was asymmetric. The Twitter-model. Since I mainly use it to find new places to go, I’d like to follow people that live in the city that I’m currently in. I understand that they are most probably not interested in me, so a simple “Follow” would suffice and wouldn’t burden them too much. Asking someone I don’t know to be my “friend” just to get a restaurant recommendation seems a little much.

It’s interesting how these relatively small choices have such a big impact. It seems like Foursquare is going through a bit of a rough patch. I wonder if it would be different if it was asymmetric instead.