Agreed. For the benefit of those who don't know, blowback and locked breached operations are the two primary operating methods of semi-auto pistol actions. In a blowback action, the recoil spring and mass of slide is sufficient to keep the breech closed until after the bullet as left the barrel. It's important to wait for the bullet to leave, so that pressure can sufficiently decrease, you else you might get all sorts of nasty stuff - like the casing bursting and possibly damaging the weapon and/or the shooter.

Blowback is appropriate for lower power rounds. As power increases, the mass of slide and stiffness of spring must increase, and at some point may become impractical. So for locked breech actions, some other mechanism is used to lock the breech closed until after the bullet as left the barrel.

In the case of the 1911, is the combination of the locking lugs and and the pivoting link. The locking lugs lock the barrel into the slide, so that when fired the barrel and slide initially travel together rearwards. The pivoting link (that thing behind the recoil spring) allows the entire barrel to drop down as it travels backwards. As it drops, the lugs disengage with the slide, allowing the slide the continue to the rear and open the breech. All of this is calculated so that the pressure is safe when this occurs.

There are a variety of ways to do locked breech. The 1911's action (tilting barrel) is particularly influential.

I assume reading John Moses Browning (PBUH) patents would be enough to actually re-create the 1911 and Hi-Power and thus essentially all modern semiautomatics, with a basic level of machine shop skill.

OTOH, if you were trying to recreate firearms in a "world has ended" situation, an open bolt SMG like the Sten would probably be your first choice.

Firearms, especially pistols, would be pretty far down on the list of technologies I'd want to recreate -- irrigation, communication, and direct production technologies seem a lot more important.

I wouldn't quite go that far with the patents. I just looked at the 1911 patent, and it basically describes a dimensionally parameterized 1911 (which is in itself really cool). Parts and lengths are given in qualitative terms.

For a 1911 design for example, the actual patent and patent drawings would not quite be enough give you the dimensions nor location of the swinging link. It would take either quite a bit of tinkering, or sitting around and mathing (or both) to figure it out. Certainly not impossible, but I think the effort required goes a bit beyond your intention when you said 'basic level of machine shop skill'.

If you were stuck in your position, an optimist in human nature would pick something like a break-open shotgun. Not useless for self defense, but really useful for hunting - which you presumably would have to do for a bit while bootstrapping civilization again. If you weren't so optimistic... oh ya, totally an open bolt SMG.

In a situation where resources were constrained, the traditional lever-action rifle would be an excellent choice. Simple and reliable design with few parts, tolerances aren't so important, can be made to work with centerfire or rimfire ammo.

Isn't a bolt action a lot simpler? And a lot easier to make strong?

I've never thought of lever action rifles as having a complexity worth the gains, I prefer going down to the bolt action or up to a semi-auto. Which, for that matter, can be fairly simple for lighter rounds, in fact, you have to add complexity to keep many (all??) of the designs from being full auto.

> Isn't a bolt action a lot simpler?

Parts count of a bolt action is a bit smaller, but that's potentially deceptive. I'm pretty sure the tolerances are actually much tighter with a bolt action. If that isn't a problem, yeah, it's down to whatever gun can be cheaply produced in volume. (I'm sympathetic to the idea that a stamped-receiver semiauto might be a good idea as well)

My opinion here is influenced by the way lever-actions gained so much popularity in warfare and the old West even though bolt-actions were an older design.

> And a lot easier to make strong?

I imagine that's a function of metallurgy in either case. The grizzly-killing lever-action .45-70s in existence today are 140-year old designs scaled up with better materials.

Similarly with revolvers; way easier to make a semiauto pistol and depot-level maintain it than even to just fix the timing on a double action revolver.

Very true, and that's a good example for another reason. Bolt actions and revolvers have that nasty property of rotation in the action (in conjunction with close tolerances!) that somehow cosmically invites jamming. A lever action is more of a simple push-pull kind of thing.

Revolvers have hung onto their reputation for superior reliability with some people but there's so much that can go wrong, and so much that's pretty difficult to fix. In contrast you can build a semiauto that can be taken completely apart without any special tools or knowledge.

So, a really really good version of this is "How Small Arms Work" from the 40s ( https://www.youtube.com/watch?v=GJzXG7MYX1c ). It was designed to teach, presumably, likely-undereducated draftees, and so it is very clear in its instruction.

A book which is somewhat similar is "The Way Things Work" (the old German one, not the happy mammoth one): http://www.amazon.com/Way-Things-Work-Illustrated-Encycloped...

Also not mentioned - the deformation of the brass to seal the barrel. This means reloading is halfway about reshaping the brass (and the rest is replacing primer, powder, lead). Another example of the graphic explaining without explaining.

The animations are really neat but I was thinking the same thing. It's science porn with emphasis on porn. But I don't think they're meant to be educational - "just" impressive.

But I'd imagine that if a society were starting over there would still be a lot to be learned, even with whatever guides were left behind. These animations could serve as a starting point for engineers to work from, discovering problems and solutions along the way. As far as sheer information density goes, I think these get a large amount done even if there is much left out.