The statistical claims in the article make the assumption that tweets are being sampled uniformly at random, which is most likely false.
The fact that 3 machines handle 20% of tweets suggests that tweets are not in fact assigned to machines in a uniformly random manner. I would guess that there is a geographic bias as to which machines handle which tweets.
The fact that 3 machines handle 20% of tweets suggests that tweets are not in fact assigned to machines in a uniformly random manner. I would guess that there is a geographic bias as to which machines handle which tweets.