4 ms·
I don't know precisely what you're asking because the whole point is that there are three measurements (measure qubit 1, entangling-measure qubits 2 and 3, and
by drostie 13y ago
I don't know precisely what you're asking because the whole point is that there are three measurements (measure qubit 1, entangling-measure qubits 2 and 3, and measure qubit 3) whose time-order does not matter. There is a nice theorem in quantum computing that you can defer all measurements to the end of a set of operations if you replace all of the classical logic with appropriate quantum gates; conversely, you can measure as soon as you want if those quantum gates can be replaced with classical logic. Here, conveniently, there are no gates; so the theorem says that you get entangled results even though the entangled things don't exist simultaneously, because they could have if you'd pushed the measurement to the end of the experiment.
The linear description of events describes the time-order that the actual experiment used.