I’ve spent some time over the last week looking into computing devices with touch interfaces. It has become clear to me that touch is not just the next way of interacting with a computing device. As has been the case with earlier input mechanisms such as the punch card, keyboard, and mouse, the interface defines the experience and the whole platform. I have been in the industry long enough to have started on punch cards. When you wrote programs in that environment, you made sure that everything was right before you even attempted to run the program. (You also tried really hard to not drop your deck of punch cards!)

Keyboards and monitors drove command-line interfaces. In those interfaces, I learned how to enter really complex commands to do things like repeat or modify earliercommands. (If !! or !$ mean anything to you, that probably means you understood how to get the most out of the C shell command-line interface.)

The mouse, as an input device, worked naturally with windowed environments. Click, drag, and drop all took on new meaning in the mouse-centric world. The mouse lends itself well to precision but can be cumbersome for something like turning the page in a document—I still often use the page-down key instead.

Touch-based computing devices are beginning to define their environments. Not only does touch determine what is easy (turning a page is trivial, but picking an exact point on the screen is hard), but a way of working. Applications tend to be less complex, and cheaper and easier to get. Consequently, applications (and their usages) are much more disposable. I used the same family of keyboard-based text editors for at least fifteen years (EMACS) before moving to a mouse-based, windowed word-processing environment and have not changed that (Word) for even longer. However, I have used multiple note-taking programs on my iPad in just the last couple of months. In the world of the mouse and windows, deciding on a particular program required a large investment of time and money. In the world of touch and application markets, I try a few programs, pick one I like, and do not hesitate to change when a new one comes out.

All of this, of course, will need to come into play in any attempt to benchmark a touch environment such as Windows 8 Metro. For example, there may be less emphasis on particular applications than on categories of applications. It also means that it will likely be important to have more small, targeted usage scenarios than a few large scenarios. What do you think will be different (or the same) in a touch benchmark?

I, and the HDXPRT team, wish everyone a great holiday! Enjoy some time at home with family. And, enjoy playing with whatever new devices you get this year!

Bill

Comment on this post in the forums