65 I was just watching some YouTube videos about quantum mechanics and had a bit of a eureka moment.
Consider the double-slit experiment. You take an electron (or a photon, or any other particle, but electrons are easy) and throw it at a wall with two very narrow slits in it. Some distance behind that wall is some kind of detector- maybe a phosphor sheet like those found in the screens of old televisions and computer monitors. When you throw your electron at the wall with the slits, you might expect it to pass through one of the slits and hit the phosphor behind, which will then emit a pulse of light that you can see. And indeed it does- whenever you throw an electron at this contraption, you'll see a spot light up on the phosphor. However, due to wave-particle duality, the spot on the phosphor that lights up isn't usually directly behind one of the slits. Turns out a lot of electrons will wind up right between the slits, and a good many off to the sides... Go look it up, you'll see what I'm talking about.
Now, add some kind of sensor to one of the slits that will detect whether or not an electron passed through it without changing that electron's trajectory too much. It could be a little piece of wire connected to a very, very precise and fast-reacting voltmeter. Just something to detect the electric field of electrons whizzing by (which, incidentally, is why I suggested using electrons in the first place). Whenever an electron goes through that slit, you'll see a little blip in the readout; if the elctron goes through the other slit, you won't. As it turns out, when you do this, you no longer get weird diffraction patterns at all- the spots on your phosphor sheet will always line up exactly with one of the slits. The system now behaves classically, just like you'd expect based on Newton's laws.
"Ok, so why does that happen?" you might ask. It looks like when you go to measure the electron, its stops acting like a wave that can spread out and interfere with itself and more like a classical particle. In other words, when you measure the electron, its wavefunction stops being all spread out like a more familiar wave and collapses into something that looks like the Dirac delta function (which is zero everywhere except at one point, where its value is infinite). When you measure the electron, its location stops being a kind of random, probabilistic bit of quantum weirdness and suddenly becomes a definite point.
Anyhow, the earliest quantum physicists called this process "collapsing the wavefunction," and figured the simplest explanation was that the wavefunction just instantly collapses into a definite point when you try to measure it. This is called the Copenhagen interpretation of quantum mechanics, and remains the one found in most QM textbooks to this day. However, the Copenhagen interpretation has its weaknesses, some of which are well exemplified by Schrodinger's Cat (look that up, too). What happens if you open the box, but don't look in it? What if a virus observes the cat? Does the cat observe itself?
One way to solve these problems is to just abandon the Copenhagen interpretation and some up with something else that works better. One such interpretation was proposed around fifty years ago by a guy called Hugh Everett, who noted that whatever device you're using to measure whatever quantum system you're trying to measure (such as the electron in the double-slit experiment) is also a quantum system. So a "measurement" is really just what happens when two quantum systems interact- in this case, the electron and the probe. And when two quantum systems interact, they become entangled.
Now, quantum entanglement... is a weird thing. The classic analogy is that if two coins were to be entangled, then flipping one would instantly reveal what would happen when the other is flipped, even if the person flipping it is located on the other side of the galaxy, because whenever one comes up heads, the other will always come up tails. I don't really like that analogy, because it just doesn't make any sense. A more applicable one here might be a weather forecast. I can go onto the internet and see what the weather is like right now in Singapore, say, without having to call up someone who lives there and ask them. Say I go to Google and type in "weather in Singapore" at the exact same moment that sgbros looks out his window. The moment the words "mostly cloudy" pop up on the screen, I know exactly what kind of weather he would see outside his window, yet during that time, there was no communication between us at all. In fact, there was no way he could've possibly told me about the weather in that time, due to the speed of light.
I think that's a pretty decent analogy.
Anyhow, this is where my eureka moment comes in. In the modified double-slit experiment, when the electron either goes through the left slit or through the right slit, and the sensor either blips or does not, there is no hand-wavey instantaneous wave function collapse. The electron and the sensor simply interact with each other, and become entangled, because that's what happens when two quantum systems interact, and then the sensor knows exactly whether the electron went left or right, because that's what entanglement does.
TL;DR: Yay, physics!