This is why broad averages aren't always the most useful statistic. The number of complete mirrors required to make the data trove fairly robust is much lower t

This is why broad averages aren't always the most useful statistic. The number of complete mirrors required to make the data trove fairly robust is much lower than the number of people in the world, and the prior commentor's qualifier was "people reading this", which is likely a biased sample and includes many people working in technology in the richest countries in the world. It's unlikely that said population's income is clustered around overall national averages, and only a relatively small number of outliers are required to set up adequate mirroring.