[RogerL]

Well, it surely was lacking in the evidence department.

However, it certainly squares with my experience. Heck, I was up at 430am this morning, standing in my garage and giving an impromptu lecture to my clothes dryer about the math behind Kalman filters and Dempster-Shafer theory, because I had some ideas in my brain burning to get out, and talk out loud imaginary lectures do that for me (YMMV). Kind of hard to get that kind of solitude and privacy in a workplace where you have to measure every 15 minutes of productivity, or work in a open work space, or, you get the idea.

[freshhawk]

Interesting, I find sometimes I have to dictate into a recorder app to get the ideas out best, I suppose those end up being said in a lecture style a lot of the time, but that had never occurred to me.

So you aren't alone in that habit. I find listening to those notes something I don't do quickly enough but the number of times I've listened to an old one and heard myself say "... Have to watch out for X I suppose, anyway..." where X is a problem/bug that showed up later and was hard to track down has convinced me this can be a really useful exercise.

[typicalrunt]

In case you want to put a name to this act, this is known as Rubber Ducking [1].

[1] http://c2.com/cgi/wiki?RubberDucking