Joel Liu's Library tagged patterns →  View Popular, Search in Google

In this freeconomics world, startups still have a chance because startup costs are rock-bottom low. However, it is not enough to build a "killer app". They have to build a "killer honey pot" that uniquely attracts workers/customers that generate the content that both attracts page view "honey" and (virally) more workers/customers.

Is this bad or complex? Not really, just a different skillset. In this "honey pot" world, effective social architecture is more important than sheer quantity of application features. You don't charge (or charge much) for the "application." Instead, you harvest value out of the content/attention of your worker-bee customers.

One of the biggest problems that I have with the "free" argument, which you've mentioned in your article, is the whole idea of marginal costs of zero, or virtually zero. Generally speaking (and I can't think of a good counter-example), the only way that you get to such low costs is through significant capital investment and mass production so that, over time, fixed costs are distributed over a huge volume of product.

Anything that's mass produced, or mass distributed, still requires a rather large up front investment. That takes deep pockets, which many smaller companies don't have. You've indicated this above in your comments about Gmail overpowering Yahoo Mail (although Hotmail is still around...?)