Spectral triples of holonomy loops (Q865068)

The authors, starting from some works by Connes where an unification of the standard model with general relativity were proposed, try to find a quantum meaning interpretation of this approach in the framework of quantum field theory. So they build a suitable ``Dirac-operator'' acting on a ``suitable Hilbert space'' with the purpose that its ``quantum content'' should be described by a \textit{Connes spectral triple}. This is a triple \(({\mathcal A},{\mathcal H},{\mathcal D})\), where \({\mathcal A}\) is a \(C^*\)-algebra, \({\mathcal H}\) is a Hilbert space on which \({\mathcal A}\) acts by bounded operators, and \({\mathcal D}\) is an umbounded self adjoint operator on \({\mathcal H}\) with compact resolvent \(({\mathcal D}-z)^{-1}\) for \(z\not\in{\mathcal R}\), and such that the commutators \([{\mathcal D},a]\) are bounded operators for all \(a\) in a dense subalgebra of \({\mathcal A}\). This object should encode a \textit{noncommutative space} in the sense of Connes. The motivation is that the Riemannian geometry of a commutative manifold \(M\) is just encoded by such a type of triple, where \({\mathcal A}\) is the algebra of smooth functions on \(M\), a dense subalgebra of the \(C^*\)-algebra of continuous functions on \(M\), \({\mathcal H}\) is the Hilbert space of square integrable spinor sections, and \({\mathcal D}\) is the Dirac operator. The spectral triple that they aim to obtain consists of equivalence classes of loops acting on a Hilbert space of sections in an infinite dimensional Clifford bundle. However, they claim that the ``Dirac-operator'' acting on this Hilbert space does not fully comply with the axioms of spectral triple. Reviewer's remark: Even if the notion of spectral triple introduced by Connes fits many generalizations of Riemannian geometry to noncommutative spaces very well, and could also work well in some quantum manifolds, it cannot be a fundamental machinery to encode quantum gravity. In fact, the quantum meaning for such a theory must be necessarily of dynamic type, and cannot be interpreted in a rigid sense, as given a priori, without any relation to dynamic issues. Quantum gravity must be encoded by means of a quantum PDE. (Similarly the classical gravity is encoded by the Einstein PDE \dots does not simply coincide with Riemannian geometry !!!) In this respect the present paper appears to well understood the problem with the Connes approach, even if the solution adopted to overcome it is yet inadequate. In fact, it necessitates a geometric theory of quantum PDE to implement a satisfactory quantum gravity theory. This is just the line followed by the reviewer of this paper.