Here is a very interesting demo of the new HTC Diamond phone’s sensors. In this game, called Teeter, you move a marble by tilting the phone and you need to avoid the holes along the way.

It’s a very impressive demo, but as opposed to the author of the article, I really don’t think that much more can be done with this kind of technology. People are quick to point out that we could write fantastic games or develop great applications with this technology, but the fact that you need to keep the screen visible (and at a reasonable angle) as you tilt the phone severely limits the range of what you can actually do.

Recently, I have also started encountering limitations in what you can do with touches on a phone screen. Touch interaction allows for great-looking devices that are not encumbered by ugly-looking keyboards, but it also takes a toll on user interaction in two ways:

  • The widgets need to be much bigger so that fingers can accurately tap them, thereby limiting the amount of information that can be shown on the screen.

  • When you tap on a widget, your entire finger will mask a significant portion of the screen, including the very widget you are tapping on.

This last point is particularly interesting because it impacts user interfaces in a lot of subtle ways and it actually makes touch inadequate for a large number of games, for which not seeing the entirety of the screen is simply not an option.

There is no doubt that touch and accelerometers are here to stay, but in my opinion, the most revolutionary applications that we are likely to see in phones in these coming years will come from features that don’t require any user interaction, such as camera, compass and GPS.