As it applies to fitting a barrel to a receiver the receiver effectively becomes the gauge. It either fits well on the tennon or rattles around like a marble in a can.
Wires, micrometers, etc. become rather pointless. This is only amplified if the action has been chewed on by the smith. Full profile inserts are great when running a pile of parts that need to go to a specific thread quality/class. This is what I use when I'm turning a muzzle for a can for instance. The suppressor manufacturers often use a 1/2-28 pitch thread machined to a 2B standard. I'll use the corresponding insert to cut a class 2A on the barrel. So far so good. No bitching from clients. I'd really like to have the suppressor in my hand, but the paperwork involved makes it unrealistic to do all the time.
With regards to being competent enough to work on guns, we (as gun plumbers) exist in a foggy environment when it comes to tolerances. All too often I'm presented with a receiver, be it std production from Remington or a boutique custom action, and the threads are not what the Machinery handbook says it should be. I have the gauging to check 2B threads for Rems, Winny's, and the 1-1/16x18 stuff used by many boutique makers. So I'm left with a decision. Reject 60% of what comes through the door in the interest of being a purest, or make it work. Making it work is far more lucrative.
Man visited the moon in the 1960's-70's. I'm betting at some point some machinist had too much to drink the night before and goofed at least one thread on the Saturn V rocket.
All the astronauts that made it off the pad returned safely so I can accept a receiver thread that's less than ideal in most instances for I know how to bump things around and make the tennon feel like the thimble on a B/S micrometer when I'm done. I'm betting I'm not alone on this. I just do it with code and offsets instead of a compound.
FWIW to date the nicest/most accurate threads I've ever seen are on Jim Borden's stuff. Exceptional in fact.
When in doubt, turn a sample tennon on a piece of scrap first. This is what I do when proofing out a new tennon program for something I've not run before.
Measure twice, cut once.
C.