As someone that is strongly into metal working as a hobby and that has read about some of the physics of the metallurgy in the field this bit by RW about cleaning the casings and chambers is a key point. It shows just how important it is to increase the frictional coefficient between the two metals by ensuring there isn't ANY lubrication at all. Combine this with the rather slack shoulder angle of the .30-30 compared to other casings and it's pretty apparent why bottle neck cartridges simply don't work well in revolvers. And in those where it does, such as the .30-30 BFR, just how marginal it is that it works at all.
First off it really doesn't matter if the case is rimless or not or if it headspaces off a rim or shoulder. The brass is going to balloon under pressure and try to jam the case head against the recoil shield at the same time that the brass deforms the shoulder to fill the space. Keep in mind that under these sorts of pressures the brass is pretty much as strong as silly putty. It's the steel of the cylinder and recoil shield that keeps it all together. So other than a minimal amount of spring back the thin walls of the brass is going to be blown out to fill whatever void it can regardless of rimmed or not. And with a sharp shoulder angle that spells "locked up".
The one saving grace to a .30-30 is that the shoulder is at a longer taper angle. So even a minimimal amount of spring back means that the shoulder let the casing move ahead a trifle more than a steeper shoulder.
THen on top of this if the frictional coefficient can be increased by de-greasing the surfaces between the case and chamber you get just enough less movement and a trifle more spring back out of the sidewalls of the case. This all spells just enough play that RW's gun works when he does all these details to assist it.
So.... What is needed to make bottle neck cartridges work in a revolver? Seems to me that there are two ways.
First option is that the cylinder needs to be wedged back against the recoil shield so there's no play to allow it to jump forward from the pressure at the shoulder. That's going to take some mechanical trickery that works in connection with the cylinder timing. I suspect that the resulting revolver would look quite different from what we have now in order to make such a locking device work. And I don't even want to think about how it would make the trigger feel to have to wedge the cylinder back as part of the trigger pull.
The second option would be some way to divert part of the chamber pressure to push the cylinder back against the recoil shield. I'm thinking that if the front of the throats were recessed so there's a flat recess of around .03 inch depth and around .5 inch diameter and that if the back face of the forcing cone were increased in size that the pressure in that recess could be enough to force the cylinder back. Due to mass and timing it may not prevent the case expanding but perhaps it would halt the expansion before full lockup or be enough to compress the case shorter by some amount so that the cases are not locked solidly in place between the two parts like a door wedge.
This last idea would certainly be something to play with if I were working for an actual firearms maker. If it worked could you imagine the number of small bore varmint cartridges that could then be used in hunting revolvers?
Of course I'm great for thinking up new solutions that someone else thought about many years before me....
I suspect that someone tried this and found that it simply didn't work long before this posting.