> This is exactly what hackathons try to do. It doesn't really work because the best incentive is internal.

I agree that the best incentive is internal, but wasn't following that hackathons don't work, as I find them tremendous, where space is made for unusual work to be done. Especially awesome if that environment and time is made by your employer.

Then I realized that the 'it' which doesn't work are the prize incentives, etc. This was also found to be true in hackathons I've participated in, where there was a trend toward more effort being put into the presentation of the work rather than the work itself. Upon recognizing the pattern, the formula was changed without placements/prizes, but that seems too far the other way as good work still should be recognized and shared in some way.