[davidw]

> the marginal cost being nearly zero means that the chance of hitting a few out of the park justifies a whole bunch of expensive flops.

Yep. If you're interested in the economics of our field, I can't help but recommending 'Information Rules' ( http://www.amazon.com/dp/087584863X?tag=dedasys-20 ). One of the co-authors is now the chief economist at Google.

[lsc]

heh. I'm a janitor, not a programmer, so the marginal costs in my field are definitely non-zero. As I sadly say when I buy a particularly large amount of hardware "But, that's my porsche!"

But my point was that I think the advantage that the large corporation has is not as big as people seem to think it is. It's a lot more expensive for them to swing the bat than for you to do so, and you don't have to do as well. I think the only large advantage that large corporations still have in this arena is advertising, and of course in software like modern games that is too big for one person to complete in a reasonable period of time.