boffinwannabe said:
yes but once you define simultaneous as zero elapsed time and not an approximation then you are stuck. You can never demonstrate this experimentally because you have no way to measure it.
You seem to think that a physical model is only demonstrated if the exact prediction from the model exactly matches the outcome of the experiment.
That is not what we do. All an experiment has to do is distinguish between different models. It does not need to be 100% accurate to do that.
If model A predicts an exact value X=5units and model B predicts an exact value X=6units ... wee need our experimental design to result in a measurement of X which is accurate enough to distinguish X=6 from X=5.
From the design of the experiment we can work out an uncertainty "e".
To find out which model, if either, is "on the right track" (note: we van never tell if a model is "True" <---<<< deliberate caps.) we have to design the experiment accordingly. eg.
Lets say the experiment got a measurement for X of 5.2 units:
If e~1 then we could say that the models appear to be on the right track even though the experiment fails to distinguish them: we cannot tell which is better. If the measurement was 100 ... this tells us that both the models are way wayy off and we need to rethink them.
If e=0.1, then the experiment lends more support to model A than to model B (model A is accepted to 95% confidence, while model B is rejected).
The uncertainty is in
nature - not the model.
i.e. We only need the stopwatch button-press to be "simultaneous" with the event being timed to the extent that this distinguishes between different theories about what we are timing. The model tells us how close-a match that needs to be. The confusion you are experiencing is doe to not being precise enough about your terms.
It is this sort of attention to detail - separating the models from the reality - allows us to refine our experiments so they can tell us more detail about the phenomena under study. It let's us, also, refine our models to more closely imitate nature. It is what allows us to spot Heisenberg's uncertainty as something that has to be accounted for in the design of really fine detail experiments.
If you like to think in terms of absolutes, think of it this way: we have a model for the physics we are studying, which may be absolute if we want; and we have
another model for our measurements, which
has to be statistical. We have to apply both to make an experiment mean something.
However - you should read more about "empiricism": we can never actually prove a physical theory ... your observation about the impossibility of an absolute measurement boils down to this in the end. But we can disprove them.
So much like string theory its all works mathematically but is untestable and fails Newtons grounds for being science. Working mathematically does not make it a reality.
The trouble with string theory is not the impossibility of measuring anything: it predicts every result of every experiment ever performed after all. The trouble is that it is difficult to come up with some result from string theory that can
only be modeled by string theory. i.e. we have not come up with an experiment to distinguish string theory from any other model.
We need (re above) some value X which string theory says is 5 when everything else says it is 6 and we just happen to have not measured X accurately enough yet. IFAIK: there is no such prediction... as yet. We are able to do this in the case of special and general relativity though.
If you want to discuss string theory and why it is not just dismissed as a pseudo-science; that's a big topic: you want a new thread in another forum.
Special and General Relativity, this forum, is only a small modification of classical physics. Within the framework for S/GR, the model allows absolute measurements and provides a scheme for comparing them between different observers.
We have another model with statistics built-in to it - Quantum Mechanics.