Accidental Innovation, Part 3

I didn’t write the previous two installments so I could build up my ego. I wanted to give concrete examples of innovation and the circumstances surrounding it, to show that it’s not magic or glamorous, to show that innovation is more than sitting down and saying “Okay, time to innovate!”




It’s curious how often the “innovative” stamp is applied to things that don’t fit any kind of reasonable definition of the word. How many times have you seen text like this on random company X’s “About” page:

We develop innovative solutions which enable enterprise-class cloud computing infrastructure that…something something synergy…something something “outside the box.”

I’ve seen that enough that I’ve formulated a simple rule: If you have to say that you’re innovating, then you’re not. Or in a less snarky way: Innovation in itself is an empty goal, so if you’re using it in the mission statement for the work you’re doing, then odds are the rest of the mission statement is equally vacant.




Really, the only way to innovate is to do so accidentally.




In both the examples I gave, I wasn’t thinking about how to do things differently. I was thinking about how to solve a problem and only that problem. The results ended up being interesting because I didn’t spend all my time fixated on what other people had done, to the point where that’s all I could see. If I started designing a puzzle game in 2011, I’d know all about Tetris and all the knock-offs of Tetris and all the incremental changes and improvements that stemmed from Tetris. It would be difficult work within the restrictions of the label “puzzle game” and come up with something that transcends the boundaries of those restrictions.




Suppose it’s the late 1990s, and your goal is to design a next generation graphical interface for desktop PCs—something better than Windows. Already you’re sunk, because you’re looking at Windows, you’re thinking about Windows, and all of your decisions will be colored by a long exposure to Windows-like interfaces. There are icons, a desktop, resizable windows, some kind of task bar, etc. What you end up with will almost certainly not be completely identical to Windows, but it won’t be innovative.




Now there are some interesting problems behind that vague goal of building a next generation GUI. The core question is how to let the user run multiple applications at the same time and switch between them. And that question has some interesting and wide-ranging answers. You can see the results of some of those lines of thinking in current systems, such as doing away with the “app in a movable window” idea and having each application take over the entire screen. Then the question becomes a different one: How to switch between multiple full-screen apps? This is all very different than starting with the desktop metaphor and trying to morph it into something innovative.




(If you liked this, you might like How to Think Like a Pioneer.)

Comments

Add comment