Columns

It struck me afresh the other day that the primary input method for every computer on the planet is still the QWERTY keyboard, invented in - wait for it - the 1860s. The ironic thing about this type of keyboard is that nobody ever claimed it was actually any good for fast, efficient typing. Its designer, Christopher Sholes, actually came up with the QWERTY layout to slow down typists and prevent them from jamming the mechanical typewriter bars that hit the paper to imprint letters on the page.

For the most efficient typing possible, your fingers should never have to leave the home row, the middle "asdf" row of the keyboard. The QWERTY layout doesn't allow this for the vast majority of English words. It's easy to see why: for starters, the most common letter in the English language, e, is on the top row; Sholes deliberately sprinkled all the most common letters around the keyboard to introduce a delay in typing.

The most ridiculous thing is that a man called August Dvorak devoted his entire life to producing and promoting a vastly more efficient layout, which allowed thousands of the most common English words to be typed without your fingers leaving the home row. But the Dvorak keyboard wasn't designed until the early 1930s, when the QWERTY layout had already been firmly established. Despite every study of typing speed and efficiency showing it to be a far superior system, here I am nearly 80 years later on a QWERTY keyboard with my fingers all over the place, slowing me down.

My train of thought about keyboard input was prompted by Windows Vista Beta 2. The situation with the keyboard

ADVERTISEMENT

seems the same when it comes to modern OS interfaces. Like the QWERTY system being a hangover from the mechanical technology of the time, it seems the current 2D windowing system - which was all computer hardware of 20 years ago could manage - is actually an interface designed to certain limits, which, through widespread acceptance, has become an albatross around all our necks.

Firing up Vista Beta 2, it's a conclusion that inevitably comes to mind. Putting aside what's under the bonnet and all the many improvements that aren't visible, the actual method of interacting with the machine is exactly the same as XP, which was exactly the same as Windows 2000, Windows 98 and Windows 95. I suppose you could say things have improved since Windows 3.1, but only insofar as you now have right-click context menus and the taskbar.

Microsoft has been singing from the rooftops about how the WinFX API, along with WDDM (Windows display driver model), uses the power of modern GPUs to render the interface. But to what practical end? Partially see-through windows that do absolutely nothing to improve information comprehension or productivity. They look pretty, and that is the absolute extent of their worth.

Microsoft, a company historically led by its software engineers not its interface designers, has been trying to improve things by focusing very much on "UX" - a stupid way of referring to user experience. It's produced an interface design guidelines document for Vista apps, instructing developers on presenting information clearly and without extraneous interface clutter. But it hasn't prevented Vista's basic interface - Explorer windows for example - from being, frankly, a confusing mess.

The basic problem is what I've cleverly dubbed the Curse of the Tiny Arrows, a design proposition that somehow manages to convince itself that the way to keep an interface looking clean(ish) while remaining functional is by relegating extra features to drop-down, slide-out and fade-in widgety bits, activated by clicking, yes, Tiny Arrows. These arrows might be pointing down or left, or right - it's such fun finding them - and you don't know what they're going to do until you click one.