It's more a question of which of the (tens of) thousands of various different healthcheck pings that GitHub has undoubtedly implemented across their infrastructure should be used to determine the status page status?