My problem with the 'review' is he doesn't go into specifics. It seems to me he is shooting long form natural conversations and interactions and then making his edit from those, if so I don't believe it was ever Canon's intent to be used that way, but we need data.
Why didn't he say it took x amount of time to overheat on this occasion, it took y amount of time to restart on that occasion, we want ballpark numbers of real user data not blanket 'it overheats I can't use it' garbage, that doesn't further our knowledge at all.
If I had been in his situation, testing the camera, I'd have timed the start to overheat, I'd have recorded the ambient temperature and I'd have timed the restart delay. I'd then have presented everybody with a performance curve. That curve would be different for everybody because heat/humidity/settings/etc are going to impact the results, but they would give potential buyers an indication of performance rather than a personal opinion that doesn't relate to performance one iota.