Obviously they will screen the comments BEFORE they get to the worker. Swear words, obvious spoilers, crude drawings, etc. can be dealt with with an algorithm, then the rest would be passed off to the worker.
It does seem inefficient, but it is the most feasible way I can think of doing it. They are a software company, but they're not microsoft or apple. It would take some insane filter to deal with every single kind of improper submission.
It does seem inefficient, but it is the most feasible way I can think of doing it. They are a software company, but they're not microsoft or apple. It would take some insane filter to deal with every single kind of improper submission.