I don't entirely agree with you (obviously :) ), but I do admit you have a good point re what you call the free rider problem. If someone could e.g. just scrape

I don't entirely agree with you (obviously :) ), but I do admit you have a good point re what you call the free rider problem. If someone could e.g. just scrape the noun project and undecut them by providing the same thing (without having to put in the effort of assembling the collection), I don't see that being a desirable outcome, because presumably the Noun Project wouldn't have done all that work in the first place. Likewise for Netflix, et al. I'm not convinced the reason this hadn't actually happened is because of IP laws though, but I definitely take your point.