Jochen, thanks for your comments. I am glad you also appreciate Weizsäcker's work. He is much less cited than Wheeler yet he went much further in trying to formulate how "It from Bit" could actually work. After he won the Templeton prize he was able to supervise students who furthered his work.
For the moment I am working on the principle that the holographic principle applies for any surface bounding a volume. I know this would have cosmological implication. A spatially closed cosmology would be ruled out because a surface encloses both the inside and outside which does not make sense if the surface is small. A flat homogeneous universe is also problematical, but hyperbolic universes and in-homogeneous universes are fine. Perhaps a more complete version of my ideas would imply only the covariant entropy bound but as it presently stands I think I need to believe in the more general rule.
I agree that symmetry can be seen as "some kind of redundancy" but I don't agree that this should be preceeded by the word "just". Consider a toy model where the content of the universe is described by a single NxN hermitian matrix and the dynamics have a U(N) symmetry. There are N^2 degrees of freedom and the symmetry Lie algebra is also of dimension N^2, yet the system is not quite all redundancy. There are N real invariants modulo the permutation group S_N. When we talk about symmetry being a redundancy we need to think about what is left when it is integrated out. With gauge fields there may also be topological invariants as well as conserved charged. I think this is consistent with what you are saying about its comprehensibility.
As to how quantisation builds symmetry, in Weizsäcker's work he starts with a bit which has Z_2 symmetry. The first quantisation is a qubit with SU(2) symmetry. The next quantisation is a larger object that includes larger symmetries and so on. The right definition of quantisation is still lacking but the version that came up in my work builds bigger necklace lie algebras from smaller ones. (I connected this to Weizsäcker's work later after a communication with David Finklestein) I did not have space to describe the details here. I think the anomalies will not kill the important gauge symmetry in the real theory otherwise you can lose unitarity and become inconsistent.
Thanks for you comments which have made me think. I have already noted that your essay is another technical one so I will read it carefully next.