Here's what little I know: A measurement is an interaction of the quantum system under test/consideration with another system / component that is classical (no

Here's what little I know: A measurement is an interaction of the quantum system under test/consideration with another system / component that is classical (non-quantum, e.g. big). Why? Because otherwise we would have to specify the exact details of every measurement device. Model number wouldn't be enough. What is the full quantum state of photon detector with its enormous number of atoms? Unusable. Entanglement isn't some spooky thing. In fact it's required for us to be able to measure anything. The measuring device's quantum state must become entangled with the quantum system it's measuring. We can't really consider each one separately, but we try anyway, and with great success if we do it right. I like James Binney's take on this: QM is adult physics. It's an admission that when you measure something, you disturb it. When you measure something really small, you disturb it alot, and can't idealise that away as you can for classical macroscopic systems. He explains it better than I can: https://www.youtube.com/watch?v=NKPI_wurNlo&t=37m20s https://www.youtube.com/watch?v=NKPI_wurNlo&t=37m20s EDIT> I should add that for a measuring device, we want it to be correlated to the system under test. I.e. I could make a thermometer that always read 5C (uncorrelated to environment), but it would be a bad thermometer. Quantum entanglement is how this correlation is set up. But it means that now the future evolution of the quantum system depends on the (unknown to us) exact state of the measuring device.