Is the UbiComp conference getting a bit too in-turned?

ubicomp07“Using an “eating your own dog’s food” approach, the UbiComp2007 Challenge is seeking for submissions of how to implement an audience voting system to finally determine the winner of the “Best Presentation Award” [Read more here]

I’ve seen so much Ubiquitous Computing (UbiComp) research that looks at supporting social interaction at conferences (including my own). Why? Are conferences the best scenario for UbiComp? Well, they do seem like a good scenario. Conferences exhibit lots of interesting issues to do with providing technological support to people who are mobile, in unfamiliar territory, and hoping to learn new stuff and socialize. However, I expect the main reason it was chosen was because it was very practical. “Er, we need a social setting to investigate for our UbiComp conference… Oh I know, why don’t we choose the conference as the setting? – problem solved!”

On the positive side, it will be a great way of getting researchers to get together and genuinely experience each others’ designs, and get that serious level of enthusiasm that you can only get from hands-on activity. However, my main reservation is not the target scenario, but it’s the target user group: designing UbiComp for UbiComp researchers. It seems a bit too in-turned, like a snake eating its own tail. If you design any system for enthusiasts of that system, then you are very likely to end up with a particular, specialist style. For example, if you design a home automation system for home automation enthusiasts you end up with something a normal person would perceive to be weird and crazy.

I’m sure some really interesting stuff will come out of this conference, but I can’t help wondering about the limitations of this method. Maybe it’s OK for a new field. Eric Raymond said in The Cathedral and the Bazaar “Every good work of software starts by scratching a developer’s personal itch.” So maybe it is OK to start off by designing for yourself. The risk is, though, that if you go too far down that road you might suddenly find out that after all that work, you find you’ve made something that nobody else wants to use.

People at the UbiComp conference will be, basically, quite nerdy gadget lovers. Because of their love of the subject matter, they will be probably be willing to forgo a bit of privacy and take on “odd” social behaviors for the purposes of a cool new UbiComp concept. And they are definitely going to go to some effort to use the systems deployed at the conference that might simply get ignored if deployed at a normal conference.

What do you think- should today’s UbiComp research be aimed at satisfying the needs of normal people?