For the past several weeks, I've been working on a very complicated app. I've been working with Maciej, my colleague, and together we've been struggling to keep the whole thing going in our heads.

At the same time, I've been reading a biography of Einstein that concentrates on the Great Man before he had attained his now-exalted status. What I find fascinating is how much Einstein struggled with the different, competing observations and theories before arriving at his breakthrough understanding of relativity.

Einstein was not content with a theory that would explain observations; he wanted to understand the fundamental nature of the universe. Once that was done, he was convinced, theories for individual phenomena would arise naturally. He found that understanding by hewing to a belief that reality was, at its heart, beautifully simple. And, as far as equations go, there's not much simpler than e=mc^2. Ah, but getting to that simplicity nearly broke him.

When his theorizing led him into convoluted explanations, he rejected them: the theory wasn't simple enough. And simple, Einstein realized, is beautiful. Not mindlessly simple. Not artifically simple. But beautifully simple. He avoided the warning given by H.L. Mencken that, "for every complex problem, there's a solution that's clear, simple...and wrong." So, time and again, the patent clerk working in Switzeland rejected his ideas and cast again for a solution that would have the simplicity he believed was requisite for truth. Even after his ideas were lauded and the word "genius" was being associated with him by other physicists, it was Einstein who rejected the easy comfort of adoration, sensing that his beautiful theory was not beautiful enough because it was not simple enough.

It's safe to say that none of us is likely to overturn our understanding of physics -- or programming, for that matter. But what is to guide us when we're out on the edge--at least what is the edge for us? Maciej and I have found time after time that, when finally emerging from bouts of deep confusion, the right answer, the elegant answer, the correct answer turns out to be beautifully simple, so much so that when the problem seems to demand a complex solution and our code winds back on itself and starts sprouting branches going in all directions, we stop, look at each other and say: "We're thinking about this wrong. Let's see if there's some underlying simplicity we're missing." And you know, it works.