Just noticed this, wondering what is the cause behind a downage this serious - they rely on being constantly up and have huge investments in infrastructure to a

Just noticed this, wondering what is the cause behind a downage this serious - they rely on being constantly up and have huge investments in infrastructure to avoid this. Possible attack or a large change gone wrong? Or even, third possibility - their research lab experiment gone wrong?