[shoegumfoot]

> In fact, you can always observe which slit the particle passes through, so measurement has no effect on the wave interference pattern.

The first half of that sentence is true, while the second half is not (unless I am misinterpreting what you are trying to say).

Empirically, nobody has (yet) managed to come up with an experiment that measures which slit the photon/electron/other quantum object passes through AND preserves the wave nature of the double-slit effect. In other words, when you set up an experiment to measure which slit is traversed, the characteristic double-slit interference pattern disappears. Again, I'm talking about this purely in practice, or what people observe when they try to perform this experiment[1].

Theoretically, this is result is completely expected. To determine which slit the photon traversed, you must interact with the photon to observe it. Any observation has two results: First, you know the state of the quantum object. Second, the quantum object is in an eigenstate of whatever operator you used to observe it. Since you have forced the photon into an eigenstate, the interference pattern disappears.

[1] Not that you can still see two superimposed single-slit interference patterns, which might confuse some people into thinking they are seeing a double-slit pattern.