Keyboards Are For Wimps

The Age

Tuesday September 16, 2008

Graeme Philipson

How we interact with computers is ripe for change. But not just yet.

THE first commercial computers, in the 1940s, were programmed by plugging cables into sockets. Things improved over time, with punch cards and paper tape punched with holes, but even the first microcomputer, the Altair 8800 of 1975, needed an array of switches to be flicked to make it work.

The history of computing has been about improvements in processing power and storage, and about better and better programming languages, but the man-machine interface has not improved nearly so much.

Keyboards replaced cables and switches, green screens and the coloured screens replaced teletypes, and all input and output devices got a whole lot better. But not very quickly, compared to advances in other areas.

The biggest improvement was the invention of the WIMP interface. WIMP stands for either Windows, Icon, Mouse, Pull-down menus, or Windows, Icon, Menu, Pointing device, depending on who you listen to. It is a closely related term to GUI (pronounced "gooey"), which stands for graphical user interface.

Today, we are all gooey wimps. This style of interface was invented at Xerox's Palo Alto Research Centre in the 1970s, and popularised by the Apple Macintosh in the mid 1980s. It became almost universal with the widespread introduction of Microsoft Windows in the early 1990s.

The evolution was not clean. Apple essentially stole the idea from Xerox, but then everybody stole all sorts of ideas from Xerox (the laser printer and the local area network, to name a couple of examples). But when Microsoft stole the idea from Apple, Apple had the audacity to sue Microsoft for infringing its intellectual property, a suit that was so laughable it collapsed under its own mirth.

We've come a long way since those first GUIs. Or have we? Today's PCs are thousands of times more powerful than the first Macs or Windows-capable PCs, but we all still use a mouse and a keyboard to get information into a computer, and a screen and a printer to get information out.

For many years there have been attempts to move beyond the WIMP interface. Voice recognition has improved significantly, and some of us can now talk to our computers to make them (most of the time) do as they're told.

We have touch screens and tablet PCs. We have devices that project information on to special glasses and which can track an eye's movement around a screen. We have experimental carbon-silicon interfaces that may one day enable our brains to work directly with computer memory. (Fancy eight terabytes at the base of your brain? Never forget anything ever again! Carry Wikipedia around in your head!)

But all these developments are slow. Bill Gates got up in front of an audience in Hong Kong last month and gave a long speech about the future of computing, in which he concentrated on the man-machine interface and where it's all headed. He spoke not of the GUI, but of the NUI (natural user interface), and of advances in robotics and "pervasive" and "embedded" computing that will make current ways of interacting with computers redundant.

Now, many people like to criticise Bill Gates and Microsoft as inhibitors, rather than promoters, of technological change, but it is hard to argue with his predictions in this case. The current model of the man-machine interface is clunky and overdue for repair.

But as is always the case with predictions, the time frame is the problem. These things will no doubt happen one day - but when?

I fear it will be a slow old ride.

Look at any notebook computer, and what you see is not much more than a keyboard and a screen. If they go, is what you have a left a computer, or something else?

How will voice recognition work in an open-plan office, even in a world where we are becoming inured to hearing private chat shouted into handsets in public places?

And we are seeing more screens, not fewer. The "one foot" mobile phone experience, the "three foot" PC screen experience, the "10foot" TV experience. The screen will remain the most dominant output device, no doubt about it.

With all the predictions of the demise of the inefficient qwerty keyboard, originally designed to slow typists down so the keys didn't get in each other's way, a massive amount of typing is now done on even more inefficient 12-key pads of mobile phones. In some ways, we are regressing.

I suppose the day will come when we will interact with computers with a wave of the hand and a glance. We will one day get the eight terabyte extension to our cerebellum.

But don't throw away your mouse any time soon. Just as today's PC looks like the one you were using 10 years ago, so will the one you are using 10 years hence be instantly recognisable as such.

As they used to say, we'll see the paperless office when we see the paperless toilet. I suspect computer keyboards will be with us for a very long time.

graeme@philipson.info

© 2008 The Age

Back to News Index | Back to Home