[mindcrime]

That's not so different from the way I'm approaching the startup I'm working on. It did start as something that I just thought would be cool to work on, and everything is open-source, and I definitely figure that I win from doing the project no matter whether it becomes commercially successful or not.

That said, I feel like it is something that has a chance commercially, and we are actually employing the Customer Development / Lean Startup model to things. I don't necessarily see that as incompatible with "doing cool stuff." I'm just prioritizing the "cool stuff" a little, based on feedback from potential customers. That's a sacrifice that I, personally, can live with. YMMV. :-)