Yesterday a friend of mine was mine was telling a story, to which I replied “well the definition of genius is doing the same thing twice and expecting a different result!”

Everyone had a good laugh. We discussed the meaning of the joke for half a minute, before I concluded by adding in the exact same inflection “like I always say, the definition of genius is doing the same thing twice and expecting a different result!”

At this point someone else in the room, who'd been within ear shot the entire time but hadn't been listening, suddenly piped in with “what? You know that's the definition of insanity, right?”

I had tried to do the exact same thing twice, implicitly proving the real saying as a joke. Ironically however the result was different the second time.

Scientifically, doing the same thing twice should yield the same result, so assuming something different will happen is crazy. In the real world, though, almost nothing is isolated, so nothing is idempotent, and something different usually happens. Getting the same thing to happen twice is actually really difficult, and in fact the only way in which experiments become respected in the scientific community is if they are repeated, preferably by different people.

However, too often people take it for granted that the same thing will happen twice because they mistakenly think they completely understand the causes of what happened the first time. If you acknowledge the fact that most of us never really know what we're doing anyway and those that do rarely have 100% control of the outcome of our actions, there are many things people could attempt repeatedly while getting widely different results. This phenomenon forms the basis of the game “Horse.”

If being insane means expecting a different outcome from the same actions, yet most people are incapable of performing the intended action once, let alone twice, then maybe the real measure of sanity (and competence in general) is the ability to identify why an outcome is different from a goal and adjust accordingly.