This is a can of worms. Safety can only be a quantitative value (x events per 100,000 hrs, etc). There is no such thing as a binary "safe condition" vs "unsafe condition".
As has been pointed out before, it's entirely possible for a pilot to fly his entire career without a real engine failure. Is this "safe" or "Unsafe"? How seldom are two failures on a single flight? It has happened, sure, but has a Captain's decision to press on EVER affected the outcome in this unlikely case?
FAA needs some operations engineering expertise, as evidenced by their puerile language.
Agree, safety is truly a quantitative value, but most of us have a lack of comparisons vs other risks that we live with daily. So our tendancy is to use the binary on-off approach to justify our own individual opinions.
Boy can you ever see this in this thread.
However, the rulemaking by the authorities including US and European does consider the quantitative approach in developing the rule, and then decides at what level is "safe enough".
Using this approach, it is generally accepted that if you exceed the rule you have thrown a switch and you are no longer safe.
This is the basis of the rule, but obviously like all rules it can't cover such unforseen circumstances where 500 operators decide to operate right to the edge on every flight. Thus the safety is really assumed on the basis that the minimum level (flying on only 3 engine) will be a rare event.
So we now need to ask ourselves if it indeed was. Once again I doubt that we will ever obtain consensus on this board, so I truly await the outcome of the examination by the authorities of the individual circumstances for this event within this operators operating practices..