Sometimes it’s fun, even useful, to look out five years or so and try to predict what might happen in the field of computers. The easy predictions are that computers will be smaller, cheaper and faster. That isn’t really a prediction. It’s merely a boring extrapolation from what we already have, mixed with a resolute faith in Moore’s Law. Ho-Hum.
The fun stuff happens when we’re willing to go out on a limb and try and break a few boundaries. If the prediction doesn’t get at least a few people shaking their heads and saying “You’re an idiot. That’ll never happen!” then we haven’t tried hard enough.
So? Let’s choose a starting place. I said the fun stuff starts when we “go out on a limb”…so let’s use the field of modern prosthetics as a starting place. Great advances are taking place in this field. Sensors attached to different muscle groups easily perform simple tasks like opening and closing a prosthetic hand. The electrical energy used to contract muscles on one side of the arm close the hand, while the muscles on the other side of the arm open the artificial hand.
The failing in this process is the user of the artificial limb requires training to use the correct muscle groups to achieve the desired results. The patterns of electrical activity are pre-set in the limb and the user must match those pre-determined patterns in order to open and close the hand. The future of prosthetics is to hardwire individual nerve endings to the artificial limb, offering the promise of much finer movements.
Let’s push the basic concept over the edge a bit. Assume for the moment you’re a touch typist. Each letter you type requires a unique set of muscle contractions in your hands/fingers.…What if.…
What if you wore a pair of gloves, or perhaps just two wristbands, that monitored the electrical activity in your hands as you typed via a collection of sensors? What if that data is fed into a program running in the background that compares the incoming data to the actual characters being typed? I suspect the type of programming best suited to this activity would be some sort of neural net.
Once enough data was collected both from the monitors and the keyboard itself, I’d be surprised if there were no useful correlations between electrical impulses and the keys struck by your fingers.
Once correlations were found the computer could start making guesses as to what key you would actually hit based upon the electrical data received. When those guesses were more than 99 per cent accurate…the computer could start substituting it’s guess for your actual keystroke.
At this stage…you could actually disconnect the keyboard and still type…and then perhaps even get rid of the keyboard altogether? It would finally make the concept of working on a plane in coach a workable idea.
Far fetched? Unlikely? You could be right. Except I’m not aware that anyone has actually tried anything in this direction. Although…there are active attempts, with some success, with interfaces that allow you to merely think at the computer and get some simplistic results. Up. Down. Left. Right. It’s not much, but it’s a start.
For all the computer advances in the past decade or so, the vast majority of it has been along the lines of smaller, cheaper and faster. It would be nice for a change for some progress to be made along the lines of smarter. After all, it’s a new millennium, or so I’m told.
de Jager is a speaker and consultant on management strategies and related issues. He’s just published a book entitled Truth Picks — Observations on this thing called Life. For more details, and to order it online, visit