Dreaming Versus Doing

How do you approach new projects? Do you start by thinking about what you can do, and then what you could do with those skills?

Or do you think about what would be cool? Or about what needs are there; personal, societal, commercial or otherwise?

Jeff Atwood stated that if you want to be a great programmer, you should put down the keyboard, put up the feet, and take a physical and mental step back from the code too:

cultivate passion for everything else that goes on around the programming

Rather then think from the bottom up, dream for a bit. One day you'll end up drinking a martini (or whatever it is you like to drink) on the veranda on some beautiful island when that wild "only you thought of it" idea was turned into reality, and you made a bit of modest dosh at the same time.

Limit your self by your dreams, not your skills.