I suggest nobody can seriously think a QM system behaves differently when the experimenter goes off to get a cup of coffee.
"The questions with which Einstein attacked the quantum theory do have answers [does the moon exist only when it is looked at]; but they are not the answers Einstein expected them to have.
We now know that the moon is demonstrably not there when nobody looks."
Mermin, N. D. (1981). Quantum mysteries for anyone.
The Journal of Philosophy,
78(7), 397-408.
In point of fact quantum theorists like Mermin are perhaps in the majority here. Many adopt the approach that the wavefunction is Ψ-epistemic, and therefore it makes no sense to talk about its behavior independently of an observer of sorts (and indeed the contextualists, who oppose this notion vehemently in many respects would tend to agree here). Fuchs, Mermin, and the other QBists assert that that their position is realist yet maintain that QM is inherently subjective and it is nonsensical to think otherwise. The ontological interpretations in which the behavior of quantum systems is independent of what the observer is doing are actually in the minority. The question isn't so much that quantum systems are defined in terms of observation or that their properties make sense only with respect to the observer (and in a manner fundamentally different from that in special or general relativity), but how.
What would be supposed to happen if the experiment is "observed" by the laboratory cat? Or a passing wasp?
Put a little differently:
"It would seem that the theory is exclusively concerned about 'results of measurement', and has nothing to say about anything else. What exactly qualifies some physical systems to play the role of 'measurer'? Was the wavefunction of the world waiting to jump for thousands of millions of years until a single-celled living creature appeared? Or did it have to wait a little longer, for some better qualified system... with a PhD?"
Bell, J. (1990). Against ‘measurement’.
Physics world,
3(8), 33.
Bell had basically the same troubles with the standard approaches to understanding QM as you express. Yet the actual answer is "we don't know." Not everyone is anywhere near as extreme as Stapp's approach to observation (which builds upon and extends that of Wigner, among other founders) nor as extreme as Wheeler's participatory universe in which the past literally can be understood as not existing until information about it is observed. But the fundamental role that measurement and observers play in QM remains, despite decades of attempts to solve the problem.
The "observer" in QM is no more than the "observer" in relativity. An entity, in a particular frame of reference or informational environment, that interacts with the phenomenon in question.
In relativity, one simply uses the appropriate transformations. In QM, perfect knowledge of a system corresponds formally to a kind of statistical distribution of properties that are typically mutually exclusive and collectively exhaustive (which is the big problem modal interpretations have- they need to explain how properties that are described mathematically but never observed and which can be more or less "probable" and therefore more prevalent, important, or some similar kind of description are somehow made so such that e.g., the deadness of the cat can be smaller or largle depending upon the the probability density associated with the relevant amplitudes).
The difference played by observers is fundamental. In one, it is a matter of reference frames and units. In the other, it is fundamental properties of the system.