How Livescribe created a reliable control interface

The Livescribe Pulse has created a reliable control interface by both separating and combining affordances to create a tool that assists human cognition. It's approach is to employ assistive intelligence rather than artificial intelligence. Here's the background.

Yesterday I referenced Mencius Moldbug's contention that user interfaces needed to be reliable. Mencius' contention is that user interfaces need to be predictable, precluding intelligence being applied by the computing device.

Because any attempt at intelligence is inherently complex, the UI is effectively byzantine and incomprehensible… In the department of intelligent control interfaces, everyone above a certain age will be reminded of one great fiasco of the past: the Apple Newton, and its notorious cursive handwriting recognition… Again we see an intelligent algorithm attempt to insinuate itself into the control loop. Again, we see risible disaster.

That doesn't, of course, mean that handwriting recognition is wrong, or bad, or useless. It just means that handwriting recognition cannot be part of the user interface control loop. Removing handwriting recognition from the control loop was key to Livescribe's success.

With the Pulse smartpen, Livescribe designed a set of discrete and reliable facilities to support human recall. Firstly, the digital ink works just like a pen. You write. You see your writing on the paper. It's just like using a dumbpen. Secondly, the pen records sound simultaneously with the writing. The audio recording provides the notetaker with perfect recall of what was said. These two functions work reliably and predictably. There's no super-intelligence or algorithmic guessing involved. The Pulse is a tool the notetaker can rely on.

With the Pulse, handwriting recognition has been shifted from the control loop to a post-processing stage. The Pulse doesn't attempt to turn script into text. Instead, the smartpen loads its data into the host computer. Once there, the human can interact with the data in a variety of ways—including handwriting recognition.

Take a look at the handwriting recognition capabilities in the MyScript for Livescribe Pulse product. The recognition is not perfect. Imperfection would be a capital crime were the functionality included in the user interface control loop. But being located in a post-processing step—where you're preparing rough notes to be imported into the tool for notes—the handwriting recognition gives you a helpful headstart.

MyScript for Livescribe Pulse, recognizing notes from a First Aid course. (Click for larger image.)
Incidentally, MyScript for Livescribe Pulse was released for Mac OS X today. The Windows version has been available for some time.