How about a concave payout function? What if someone with 10 million streams didn't expect to get paid a million times more than someone with just 10 streams? How about paying out some of the money randomly or as grants, that are more likely to end up in the tail? This could counter the economics presented in the article. I'm not familiar with these kinds of contracts so it might already be a thing or I may be naïve, but I'm not convinced linearly proportional payouts has to be a law of nature.