Michael
"There are a number of physical arguments in favour of a closed universe"
There is, and must only be, one physical reason for a valid closed system of existence. We can only have knowledge of existence. And we can only know, as opposed to believe, what it is possible for us to know. So we have known and not-known. The absolute being the differential, ie there can never be a known extrinsic absolute reference. This means we can only know one form of existence, ie knowledge is the equivalent of this form.
Put simply(!): given A, there is always the logical possibility of not-A, which cannot be defined from within A, as a reference from within not-A is required for that. All that can be defined is A, from within A, and that that is not not-A. But not what not-A is.
So the question is what physically determines known, ie what form of existence is this? And the answer is, what physically manifests as detectable. From which we can infer what caused that. [Note: detectable/know includes potentially proven so, but not belief]. The form of existence that we can know has certain basic characteristics, which are a definition of physical existence, ie it has physical substance which must ultimately be non-divisible, that exists in a physically existent state, it is sequence, etc. Which addresses your point about my "science cannot presume an abstract concept". The physical existence we can know, is a particular form of existence, it is not a 'blank sheet of paper'. Whether it is a particular form of existence, etc, is irrelevant because we cannot know. I will comment on distance which you linked to this below.
Now, the next point is about referencing and closed systems. There is no extrinsic reference, but every statement has the same logical form, ie a comparison to establish difference, which necessitates a reference. So, within a closed system, any reference within that system will suffice, it is only that, in practice, some references are more suitable than others. Comparability of identified differences must be ensured by maintaining constancy of reference. This means using the same reference, or discounting for changes of the reference, which is the same thing. Assuming due process in measurement, then the differences so identified by comparison are, by definition, correct within the closed system. The point about whether the closed system is expanding is largely irrelevant as it is an omnipresent effect, so it does not affect the vast majority of differences. Indeed, if you think about it, then whatever the closed system is expanding into is really part of the system! The real point here is that there is ultimately a state of not-known.
You say, correctly: "In Relativity, every event has the co-moving reference frame as being the unique reference frame".
This is an indication of why relativity is wrong. There is no reference. This has happened because he equated physical existence as represented in light, with physical existence, ie in effect, observational light was eliminated. The light he refers to is really time, which is a constant against which to calibrate rate of change. In physical existence, the representation is physically different from what is being represented, and physically exsted (this applies to all forms of sensing, not just sight). So there is a time delay between the time at which physical existence occurred, and the time at which a representation of that is received. The timing differential is in the receipt of the representation, not in physical existence. That, by definition just exists in an existential sequence, and at any given time we can attempt to identify what existed.
There is no duration within physical existence, because whatever comprises it can only be in one physically existent state at a time, that being superseded by its successor in the sequence. Timing concerns the rate of change between physically existent states, not of them. There is no such thing as 'cosmic' and 'local' time. Time is a deemed constant rate of change, which is used as a reference to calibrate all rates of change. There is only one, otherwise the measuring system known as timing would be useless. Timing devices just 'tell' the time, and within the realms of practicality, all are synchronised to the same point, and then maintain the same rate (irrespective of how they actually do so). In other words, the whole system is an operable manifestation of a conceptual constant. The physically existent feature being measured is rate of change, which must be between existent states, not of them. It is existent, alteration is caused, which results in a replacement state. There is a rate to this change sequence, timing calibrates it.
You ask: "distance, as measured relative to what?"
The important point here is that distance does not physically exist, it is a difference, and it can only occur between existent states existent at the same time. So the answer is: relative to whatever the distance is a function of, and a specific component thereof, as existent at the same time. As explained above, logically, what is used for the calibration is irrelevant, so long as it is always used so that the differences are comparable. This points to the fact that, like the duration measuring system, the spatial measuring system is a practical manifestation of a conceptual constant. It is important to point out that doing this is impossible in most, if not all, cases. But we must start with a proper understanding of what is involved, then make informed decisions on how to proceed.
In respect of spatial dimension (there is no 'time' dimension within any given physically existent state), the number 3 is just the least ontologically correct number we can have with the highest level of conceptualisation of physical existence. That is: up/down, side to side, back and forth. Existentially, dimension is a specific aspect of spatial footprint (ie the spatial positions 'occupied' by any given entity at any given time). As it relates to the distance along any possible axis of that 'occupation', then the actual existent number of possible dimensions is half the number of possible directions that the smallest substance in physical existence could travel from any single spatial point.
Paul