I was going through (yet again) the quantum mechanics sequences. I got new perspective about being a mechanic of configuration spaces. I am still at a loss on what kind of mathematical entity a wave-function is and couldn't compute anything with them. I guess they are somehow an animation of complex points in 3 real dimensions?

There was a lot of talk about splitting but I kinda gathered that there must be a counterpart ot it. If you can't compute the next state of a "world" from how it looks now but have to look at the neighbours in configuration space doesn't this mean that world evolution is inherently not a "private" fact? As in your viewpoint world could look exactly the same but depending on the neighbouring worlds state things would go different. Now usually those neighbouring worlds depend on the same past configurations as your world so they can't be completely orthogonal. But on the whole the wave-function that when sampled collapses, does evolve in a very determined way. And in stationary wavefunctions while each individual wave moves outward the shape of the wave function is the same after one cycle. That is such wavefunctions don't "shatter" and they have the quality of reforming the same shape. I don't have a good detail to point to but it seems it must converge just as fast as it splits.

Woudn't the "converging worlds" be us much of a big deal as "splitting worlds"? And wouldn't it be quite possible that instead of looking like a tree or balloon on the macro level your world line would look more like a line. Not a classical one of pointlike width but one with actual width. Doesn't the preservation of measure mean that worlds don't "dissipate into ambivalence". A common take on many worlds where each decision splits your world would in my mind imply such a dissipation. So there must be some mechanic where tracking an ensemble of worlds behaves differently than tracking each world separately resulting in an ensemble smaller than if each world would split into independent trees. I guess one suspect would be to arrive at the same state throught different paths? Ie instead of 2 states branching into 4 individual states they branch into 3 states one of which has double measure.

The bomb tester experiment reads to me as if you blow up the bomb in another world to gain info on the version in your world.

There was mention of mangled worlds and that similar states only meaningfully interact when close in configuration which usually means "until shielding fails". However it didn't seem like world-view implications were worked out. If my decision can affect my other fates making the spread of my measure have "connected fate" doesn't this have very different impact to meanigfullness of decisions. I would like to believe that such effect would just affect random thermal noise. But it might be possible for there to be "quantum magnifying glasses" where insignificant details in many individual worlds contribute to have macroscopic effect in some of the future worlds. I would not like to be blindsighted by them!