Looking back, over 7 years ago, to when we first began developing DubitChat, we started out with a simple assumption: if we can create a graphical chat world for teenagers they’re going to have a great time, meet new people, and spend hours chatting. This was going to be their world, for teenagers, by teenagers!

It was a fantastic time, teenagers flocked to the site and for a while we had trouble keeping up with growth. But what we saw when they entered the world was worse than any school disco – it wasn’t just boys ignoring girls, it was no-one talking to anyone! In a social world this is pretty disastrous. If people aren’t talking they’re not making friends, and if they’re not making friends their isn’t much of a reason to come back.

This simple problem became our focus; how can we encourage conversation? This article goes through some of the things we learnt so that you don’t need to make the same mistakes.

At first we thought the problem was something to do with the user interface; we changed the speech bubbles, added private chat, and created a instant messaging client.

At the time there we’re very few graphical chat rooms – the whole concept was pretty radical. The plan was to introduce the idea of chatting through an interface inspired by the popular MSN Messenger. It worked. Only the messenger interface worked so well that the kids used nothing but the messenger to chat, the chatting in the rooms actually decreased.

After more tweaks and changes to the GUI nothing we tried would get the kids chatting. Then it hit us, what we were really seeing was not a UI problem, it was a social problem; talking to strangers is really hard! Really really hard! And no matter how much energy we put into UI we we’re just addressing the wrong problem.

Hi, ASL?

It turns out, just being a teenager doesn’t mean you’ve actually got a lot in common with another teenager. This was the first, of several, simple observations that appear obvious with hindsight, but have shaped the fundamentals of how we create virtual worlds.

On realizing that teenagers don’t necessarily have a lot in common the solution was obvious: help them find other teenagers with similar interests. We built cinema rooms so they could talk about film, sports rooms to talk about sports, 3 different music venues, even a drugs alley, you name it, we built it.

Only it didn’t make any difference. No-one was going to the cinema to talk about film.

Christmas

Every Christmas the whole family goes home to ‘Mum and Dads’. In the Warneford family we have quite a spectrum of ages ranging from my 8 year old nephew through to 88 year old grandma, and everything in between.

It turns out that my family and a virtual world are quite similar; similar in our differences. A world full of teenagers doesn’t mean they all have similar interests and lifestyles. A room full with my family doesn’t mean we all have shared interests, in fact the broad spread of ages ensures we don’t – there really isn’t any one thread we can all talk about.

So after Mums turkey and Dad’s stuffing, we all come together to play a game of Monopoly, and even though we have very different lifestyles and interests we all have a great time together! While we don’t have a lot in common the act of playing Monopoly brings us together – Monopoly becomes the shared interest, the activity we have in common. Through creating an activity, in which we can all participate, we catalyze conversation, something to hang the conversation from.

So how does this translate to a virtual world? Trying to match people by their interests didn’t work. Instead, if we flipped the problem on its head, we need to creative activities in the world that would give their shared interests. Don’t try to find shared interests, make them.

This simple idea is the boiled down essence of what we are trying to achieve through our GAML Engine.