I've noticed something odd that I can't really explain: lots of people, people who aren't or don't think of themselves as designers, have ideas for things that they might build or like to see built. Now, the quickest, dirtiest prototype you can make at this stage is a sketch. But they seem utterly averse to it: I'm not sure why, maybe it's because these sketches seem like “wireframes,” something designers have laid claim to, but they just really, really don't want to sketch anything out.
I was reading Clive Thompson's new book, “Smarter Than You Think,” examining how technology actually makes us smarter when we use it intelligently. I was particularly struck by his introduction: the idea that much of technology are ways for us to externalize our thinking, and much of the “technology” is older than we think. He brought up long division: we can divide a few things in our heads, but for most division problems, having a piece of scratch paper allows us to keep track of all the little moving pieces involved in division, and if we are serious about solving a problem, we pull out a piece of scrap paper so we can do some long division.
Which basically got me thinking: “wireframes” or “sketches” (the semantics aren't important here) are really like writing out long division. If you're trying to think seriously about a software product, if you aren't sketching things out, then you aren't really thinking about it. Yes, some details are inane (“Where should I put the login button?”) but many more are the details that make up the service: Tumblr linking to its six basic post types wasn't a detail, it markedly effected the way the service felt and was used. And these are the little details that we start to see once we can start externalizing bits of our short-term memory to paper.
As an even more general point, I'd say that confidence and fluency in sketching and diagraming is one of those things that can make us exponentially better thinkers.