Super-conformal symmetries generalized from string model context to TGD framework are symmetries of S-matrix. This is very powerful constraint to S-matrix but useless unless one has precisely defined ontology translated to a rigorous mathematical framework. The zero energy ontology of TGD is now rather well understood but differs dramatically from that of standard quantum field theories. Second deep difference is that path integral formalism is given up and the goal is to construct S-matrix as a generalization of braiding S-matrices with reaction vertices replaced with the replication of number theoretic braids associated with partonic 2-surfaces taking the role of vertices. Also number theoretic universality requiring fusion of real physics and various p-adic physics to single coherent whole is a completely new element.

The most recent vision about S-matrix combines ideas scattered in various chapters of various books and often drowned with details. A very brief summary would be as follows.

- In TGD framework functional integral formalism is given up. S-matrix should be constructible as a generalization of braiding S-matrix in such a manner that the number theoretic braids assignable to light-like partonic 3-surfaces glued along their ends at 2-dimensional partonic 2-surfaces representing reaction vertices replicate in the vertex. This means a replacement of the free dynamics of point particles of quantum field theories with braiding dynamics associated with partonic 2-surfaces carrying braids and the replacement of particle creation with the creation of partons and replication of braids.
- The construction of braiding S-matrices assignable to the incoming and outgoing partonic 2-surfaces is not a problem. The problem is to express mathematically what happens in the vertex. Here the observation that the tensor product of hyper-finite factors (HFFs) of type II is HFF of type II is the key observation. Many-parton vertex can be identified as a unitary isomorphism between the tensor product of incoming
*resp.*outgoing HFFs. A reduction to HFF of type II_{1}occurs because only a finite-dimensional projection of S-matrix in bosonic degrees of freedom defines a normalizable state. Most importantly, unitarity and non-triviality of S-matrix follows trivially. - HFFs of type III could also appear at the level of field operators used to create states but that at the level of quantum states everything reduces to HFFs of type II
_{1}and their tensor products giving these factors back. If braiding automorphisms reduce to the purely intrinsic unitary automorphisms of HFFs of type III then for certain values of the time parameter of automorphism having interpretation as a scaling parameter these automorphisms are trivial. These time scales could correspond to p-adic time scales so that p-adic length scale hypothesis would emerge at the fundamental level. In this kind of situation the braiding S-matrices associated with the incoming and outgoing partons could be trivial so that everything would reduce to this unitary isomorphism: a counterpart for the elimination of external legs from Feynman diagram in QFT. p-Adic thermodynamics and particle massivation could be also obtained when the time parameter of the automorphism is allowed to be complex as a generalization of thermal QFT. - One might hope that all complications related
to what happens for
*space-like*3-surfaces could be eliminated by quantum classical correspondence stating that space-time view about particle reaction is only a space-time correlate for what happens in quantum fluctuating degrees of freedom associated with partonic 2-surfaces. This turns out to be the case only in non-perturbative phase. The reason is that the arguments of n-point function appear as continuous moduli of Kähler function. In non-perturbative phases the dependence of the maximum of Kähler function on the arguments of n-point function cannot be regarded as negligible and Kähler function becomes the key to the understanding of these effects including formation of bound states and color confinement. - In this picture light-like 3-surface would take
the dual role as a correlate for both state and
time evolution of state and this dual role allows
to understand why the restriction of time like
entanglement to that described by S-matrix must
be made. For fixed values of moduli each reaction
would correspond to a minimal braid diagram
involving exchanges of partons being in one-one
correspondence with a maximum of Kähler function.
By quantum criticality and the requirement of ideal
quantum-classical correspondence only one such
diagram would contribute for given values of
moduli. Coupling constant evolution would not be
however lost: it would be realized as p-adic
coupling constant at the level of free states via
the log(p) scaling of eigen modes of the modified
Dirac operator.
- A completely unexpected prediction deserving a
special emphasis is that number theoretic braids
replicate in vertices. This is of course the braid
counterpart for the introduction of annihilation
and creation of particles in the transition from
free QFT to an interacting one. This means
classical replication of the number theoretic
information carried by them. This allows to
interpret one of the TGD inspired models of genetic
code in terms of number theoretic
braids representing at deeper level the information
carried by DNA. This picture provides also further
support for the proposal that DNA acts as
topological quantum computer utilizing braids
associated with partonic light-like 3-surfaces
(which can have arbitrary size). In the
reverse direction one must conclude that even
elementary particles could be information
processing and communicating entities in TGD
Universe.

For more details see the new chapter Hyper-Finite Factors and Construction of S-matrix of "Towards S-matrix".