[ILOVEPOKER929]

"Not exactly. I could be wrong, but I'm pretty sure that in vacuum photons do not interact (or if they do because of some weird vacuum fluctuation / quantum field theory reasons that I don't understand, it certainly doesn't have a big cross section.) So the issue isn't that you're measuring one photon with another, but that any photons that get to the camera represent ones that did not hit the screen, and so you don't know anything about what's going on with the things that are hitting the screen."

I never looked at it this way before. My understanding of quantum mechanics is not deep enough to know if what youre saying is true or not. I will assume for now that what you said is true though. However I do think it would be better for didactic purposes to assume that the camera is using other photons to observe the photons passing through the slits even if this is impossible.

"Looking at the Wiki article (and I should read the paper at some point), the way that real quantum eraser experiments are done usually involves parametric down conversion, which is a nonlinear process in a medium where an incoming photon of frequency 2f gets changed into two entangled photons of frequency f. One of these photons gets sent to the "screen," and the other ends up going into an interferometer, where through some cleverness in design you can extract which way information for some fraction of the photons that make it through. So the trick is that you turn your single photon into two photons, and then use their entanglement so that measurements on one photon tell you something about the other."

What youre talking about here is called the "Delayed-Choice Quantum Eraser" experiment. This is an extention of the beam-splitter experiment. Since we are talking about a double slit experiment in this thread, a simple version of the quantum eraser experiment is all that is necessary to convey my point. Both the eraser experiment you mentioned and the one I delineated are valid in explaining this counterintuitive phenomena.

Again the main point I was trying to illustrate is that once "which path" information on the photon has been established, and these photons have already hit the screen, nothing in the future can change the particle pattern on the screen for these photons, so Sklansky's scenario 4) cannot happen. Again this is based on what little I know about quantum mechanics.