Hi Nick,
In re-reading your first comment above, you bring up Wigner's "Unreasonable effectiveness..." paper, answering which was a key motivation for my dissertation, [link:www.amazon.com/s/ref=nb_sb_noss?url=search-alias%3Dstripbooks&field-keywords=Klingman+Automatic+]The Automatic Theory of Physics[/link]. The figure on the first page of my essay encapsulates the way in which physical counters, producing natural numbers, can make counter-based measurements (recall that the heart of quantum field theory is the number operator.) The measurement map is the plane at the far left, and the measurements are generated automatically, one way or another. The question is what to do with all these numbers. To answer that, I invoke a robot. This allows algorithmic computations but removes intuition. (Whatever one thinks of the eventual possibilities of robots gaining intuition, it was beyond doubt that when I wrote this, robots did not have intuition. But even in those dark ages we had neural net algorithms and other self-modifying-map architectures for learning.)
My design of the robot anticipates Max Tegmark's concern with 'baggage' - I left the baggage out. The robot has access to resources, including random number generators, arithmetic-logic-language circuitry and systems, and sensors and activators to interface to the physical world, but no consciousness or intuition or beliefs - no baggage - just designed-in algorithms, including learning and self-modifying-map algorithms. This response to Wigner is a vehicle for testing "A theory of theories of physics", and some similar work by others in 2009 extended this idea.
The relevance of this to your being laid back about the incompleteness of QM is that the measurement space on the left of the figure consists of nothing but numbers, all of which can be accessed by the robot who performs cluster mapping on involving intercept and interest set distances, grouping the data into n-dimensional cluster space. This is not necessarily the optimal dimensional representation of physical reality (i.e., the object/source of measurements) so I employ a Karhunen-Loeve transformation from n-space to lower dimensional m-space, from which the feature vectors are formed. The eigenvalues are the vector elements and the eigenvalue equations are the transforms that preserve them - quantum mechanics!
In this view the quantum mechanical formulation arises from statistical clustering of measurement numbers, but what is being measured? What is at the left of the number space that is the real source of measurements? What is its nature? There is no obvious answer.
If one believes all of physical reality is somehow quantized, and it is impossible to measure any other value, then one assumes
A eigenstates exist, and
B ideal measurements yield eigenvalues
But the robot doesn't carry this baggage, the ingrained belief of almost a century of QM. There is no way for him to tell whether the numbers represent eigenstates or random digitized data derived from measurements on a continuum. Not being a robot, I do carry baggage, and my personal view is that reality is a continuum, with only action quantized, not time or length or even energy, except when boundaries are imposed on a subsystem. The magnitude of intrinsic spin is quantized, but its direction is 3-dimensional unless and until it is caused to align with the local field, as in Stern-Gerlach.
This view has led to a local model that produces the same correlation as quantum mechanics, with implications for entanglement and it seems to imply that QM is incomplete. I repeat, not wrong, but incomplete. As you mention, there is EPR baggage that causes some resistance to this notion. Not sure how to handle it.
Edwin Eugene Klingman