Tuesday, February 24, 2009

UI in the Future

Face it. Computer interfaces have not essentially changed in years. When did the first mouse/keyboard come out? I remember the Apple II and Lisa from the early 80's (although I know it was actually invented by Xerox Park). I'm sure the info is there in Wikipedia, but I meant it as a rhetorical question. The point is that for the most part we still operate computers the same way we did decades ago.

First step. Many people do not realize that using the mouse with the keyboard is slower than using the keyboard alone - when you know the keystrokes. That is why there is a keyboard. Try to imagine instead of typing, using the mouse on an on-screen keypad (for most of us this is not an attractive option). Many kids games are this way because kids haven't learned the keyboard yet. Despite this many people never learn the shortcuts that could really speed up of their work (try pressing Ctrl-S to save in many applications instead of using the mouse to click File->Save). Point - people don't use the fastest method possible because there is effort required to learn it.

The next step. Because people don't know the keystrokes, and because they are more willing to use the mouse the next step was mouse gestures. Quick, how many of you have installed an add-on or utility to allow you to cause actions by "stroking" with your mouse. (Here is an example addon for FireFox.) Not many I bet. Why not? Because it takes effort to learn. Point - people don't use the fastest method possible because there is effort required to learn it.

Intuitive interfaces. So why does the mouse with the graphical user interface (GUI) work? I say it is all about not learning. There is a minimal learning curve to figuring out how to move a mouse, and how to click on objects. That is enough for many people. Indeed if you have ever had the pleasure/pain of teaching a neophyte you will know the real challenge starts with "right click" (Apple users insert ctrl-click, unless you have the new mighty mouse) or "drag" or "scroll" (although this last doesn't seem to be too bad with a scroll wheel). Why does a steering wheel, a light switch, a pair of scissors, work? Because it is hard to use them incorrectly and easy to figure out how to use them without spending much time learning (perhaps scissors are the worst of those examples).

What Nintendo gets right. You knew this was coming (with kudos to Apple as well). They take what we already do and try to mimic it in their interfaces - so the DS has a touch screen that I did NOT have to show my son how to use. I had to show his younger brother how to use the graphical keyboard part, but he was still learning his alphabet... The Wii allows you to point (almost like pointing your finger). You can put the remote in a plastic steering wheel and it is suddenly as easy to use as a (wait for it ... wait) a steering wheel! The iPhone and iTouch are the obvious Apple examples of what can be done right. Did you know you can use a Wii remote with your computer - as long as your computer has a blue tooth connection (Google it)?

In the future. The takeaways so far are - the interface cannot require serious learning and it should reflect things people already do. Partly this will work with current tools - if you haven't tried Google Earth yet, try it. Navigation is pretty simple and it goes beyond simple point and click. People naturally zoom, pan, and even set it scrolling automatically (give the Earth a spin!). More promising work is seen with "surface computing" which is like a really advanced touch screen. Instead of just mimicing a mouse new actions are being developed. The successful new actions will be the ones that are similar to the things people already do. However I think the iPoint 3D interface direction is even more promising. Cameras are used to track your body movements which are interpreted by the computer. Think Wii - without needing a remote! The developers could still screw up though if they are unable to make it work with intuitive body movements (pointing is perhaps the most obvious). Surface computing, body movements, no matter what we come up with as our interface, if people need to work hard at it they won't bother.

No comments: