It's a question of magnitudes. There will be at least one kid who does what you're saying, but how many? My strong intuition is that it'll be a small number, to

It's a question of magnitudes. There will be at least one kid who does what you're saying, but how many? My strong intuition is that it'll be a small number, too small to cancel out the benefits. The appeal will be largely gone when the network effects are gone. So I say run the experiment in one country and observe the outcome and adjust accordingly. That is the least idealistic position.