Keyboard

I've been thinking a lot about how we could go about revamping keyboards. The story is that when the QWERTY keyboard was invented, it was designed to slow typists down. That may or may not be the case, but it is the case that common letter clusters were split up to avoid the striking bars sticking to each other and making the machine fault. If it's true that the QWERTY layout does slow people down, even, say 10% (more than 1, less than 100) of what they could easily do on a better layout, and if there are about 100,000,000 US adults who use computers at least, say, 3 hours a day (more than 1, less than 10), if we used a better keyboard layout, we could save or redirect 30 million person-hours a DAY. Even if the efficiency gains are only 1% (most of my typing time is thinking of sentences, not the typing), that's still 3 million person-hours a DAY or 600 million person-hours per working year. We could probably also include some things in there to help carpal tunnel - moving common letters like "a" away from weak pinky fingers. Come to think of it, moving common letters to the central keyboard position would speed up two finger typists considerably too.

But do we do that? Nope. There's a Dvorak keyboard that tried to do these things but never caught on in large scale. Why? Because there's huge sunk costs into QWERTY keyboards. They're everywhere - and they're often shared. I think it's the sharing of keyboards that's the killer. It's more efficient to use a slow keyboard layout you know than a fast keyboard layout you don't. But technology has come a long way, and more and more people are using keyboards, and more and more people are using them expertly, if inefficiently. So until a great many people use a new style, the old style will dominate. I think with so many keyboard variants popping up - virtual on a flatscreen iPhone or doubled letters on a texting phone, it's a good time to effect the layout shift.

It hit me this morning that if we can have a Number Lock key that switches the number pad from one set of characters to another, and a shift key that does similar things, why not have a layout key or switch as well? The obvious answer is that the letters and numbers are printed on the keys already. Again, tech to the rescue. The touchscreen layouts are a nobrainer.

For physical keyboards, we have strong, clear plastics. We have low energy, low heat LEDs. Why not make keyboard keys with a window over a selection of colored LEDs. When you turn on the red, the QWERTY layout shows through. When you turn on the BLUE, you only see the CrankyOtter layout. Similar with a green or white - or reserve the other colors for shift/control/alt/etc keys. As long as the keyboard sends out the correct code, the computer it's attached to doesn't care what order the letters are in. And for industrial and educational keyboard with multiple users, each one can sit down, choose the layout of their choice, and get to work, faster. Sure, they will cost more to start out, with custom filters/fluorescing patterns with lights being more expensive than screenprinting ink. But did I mention that 600 million person hours? Not every keyboard has to be both, but just enough. We could start with the touchscreen keypads that have no physical existence, test them, and move on.

Other keyboard modifications:
Put an ENTER key on the left side of the keyboard already! If I'm using the mouse (or other peripheral) right handed, it's handy to have an enter key at the ready with my other hand so I can leave them in place while I work through something.

There's so much online text now that it should be easy like falling off a wet rock to examine not just commonly used characters, but also commonly used character clusters. I'd particularly want "ing" and "ion" to be spaced to favor both combinations. Either make keys for those clusters in addition to regular keys (think of the old 1/2 and 1/4 keys), or position them so that the average user finds the combination smooth and pleasing as well. Or both. Or have them toggle-able. Like a "." can be either just "." or ". " or ". _" where the underline is a second space like they teach in typing class. "The" could be a single key. Similarly a comma could be set to ", " - then in the number pad, they would lose the space.

Speaking of the number pad, that could use some help as well. Any number of programs have alternate uses for those keys, I could handle having an equals sign (=) there and maybe other page/spreadsheet cell navigation keys like backspace/delete. Maybe adjust the spacing to be not quite rectilinear so that natural finger bending lands your digits in the right place making it easier to add useful, ergonomic peripheral keys (like "t,r" and "y,u" share the same extended finger and are easy to find because the diagonal layout puts both in easy reach. Then allow a user to move around the keys programming, if they like, to better win at MMORPGs or input the data for their thesis.

I was thinking of this particularly because I'm still being pained by the transition from Microsoft of old to MS 2007 which has totally messed things around, put commonly used things together to create icons that trigger more than one thing, and we're all expected to get used to it. As long as they're re-inventing everything else, why not the keyboard? MS has enough resources at their disposal that they could try various likely versions with new and old typists to correct major grievances before rolling out a final product. With a toggled layout both expert users of QWERTY and the kids who have no problem texting with a phone keyboard can use the same hardware. Then we can be more efficient when we surf the web at work.