Dan I agree on the proof testing. Depending on material and stress something can fail anywhere from one load cycle to never. I still like a moderate over pressure proof to check a new gun. Major overloads are for checking a design prototype.
I proof everything. Modern certified alloy steels will tolerate anything BP can produce behind any projectile with no strain.
The things that REALLY need testing are the vent liners and such. The thread tolerance can make an otherwise safe part/installation unsafe.
I have a 1/4-32 tap that seems to be at the upper end of tolerance and I ended up grinding the die so the holder could expand the split die to make something like a reasonable fit.
I could not find a H1 tolerance tap when I looked. This last liner means I need to order another tap and hope its a little tighter if I can't find an H1.
I just proofed a 36 with 75 grains of FFF Swiss and 2 balls. Nothing moved. This is probably still only a 50% overload pressure wise.
The British did pretty heavy overloads for everything except cartridge shotguns according to my 1896 printing "The Gun" .
The proof they would perform on a barrel like I did today is double service powder and a bullet 1 1/3 the ball weight (Second, Fourth, Sixth and Seventh classes). The Second class includes ..."Single Muzzle-loading rifled Arms of every description, except arms of the 7th and 8th Classes".
My load above was probably a little heavy on lead, but might be light on powder 37 grains is probably going to be light for the accuracy load but this will not be known for awhile.
A 20 bore "Carolina" (one of the Third class guns) would be proved with a 316 gr ball and 219 gr of powder. This is the lower "Definitive Proof" the only one used for this class.
When we consider the possibility that guns we make could be used with propellants that are sold as BP replacements but are capable of and even designed to produce higher pressure, like T7, proofing a to the upper end is not a bad idea. Not proving to a high enough level is really no proof and IMO is pointless. Greener's book gives a guide for what is and is not an adequate proof.
Dan