Oh boy! I have Opinions about this subject!
I admit, I've skimmed the thread, so some of this may have been addressed. Since nobody has been able to find a "proper" standard, it's safe to assume one does not exist.
So, what would scienticious testing need to do?
For folding knives, we're mostly interested in lock failure, I suppose.
Prior to testing, all of the physical properties of the blade should be known, obviously.
1) Known and repeatable grip strength.
2) Knife being held the same distance from the lock interface.
3) Edge-strike tests, from blade-on to 20 degrees of deflection on either side. Repeat until failure, or a known, "acceptable" number of ft/lbs. There is data about how hard people can hit things with other things, but most likely, people would want to know exactly how much it takes to break the blade, or the lock.
4) Strain tests, slowly increasing the weight on the blade, at a known and repeatable speed. It should be done with the edge up, and the edge down.
5) Side-loading tests. Same as 3 and 4, with the flat side of the blade taking the shock and the weight, at some repeatable distance from the lock interface, with the handle being clamped and the blade unsupported. If the lock is asymmetrical (framelocks and linerlocks for example), then testing should be done from both sides.
6) Leverage testing. Essentially the same as 5, except that now the point of the blade is driven a constant distance into a material with known qualities, and the handle is unsupported. The results may not be any different from 5, but somebody is going to say "you did it wrong!" because "hard use" includes prying.
7) Reverse shock-load test. The usual nonsense of giving the spine a sharp whack to show that the lock can (or can't) hold up to being asked to do the exact opposite of it's job. It needs to have a technical sounding name, because we're doing science. Keep all the other distances and forces constant while increasing striking force at a constant rate, and so on and so forth.
8) Somebody too much time and money who's willing to fund the purchase and programming of a robot to do this.
9) Somebody who's willing to put on a lab coat and safety glasses, then narrate the testing process with just the right mix of knowing what they're doing, and not taking the whole thing too seriously.