I think that you make a lot of interesting points until your final paragraph. If I'm reading the details properly (and I may not be; you seem to be far more informed than I am), then the particular test in question may have detected the problem about which Mr. Barnett is complaining, where other tests might not:

> Mr Barnett says that when he was decommissioning systems which had suffered minor cosmetic damage, he found that some of the oxygen bottles were not discharging when they were meant to. He subsequently arranged for a controlled test to be carried out by Boeing's own research and development unit.

> This test, which used oxygen systems that were "straight out of stock" and undamaged, was designed to mimic the way in which they would be deployed aboard an aircraft, using exactly the same electric current as a trigger. He says 300 systems were tested - and 75 of them did not deploy properly, a failure rate of 25%.

...or am I reading this incorrectly?