JeffPT: The basis of the uncertainty principle is that you cannot know the current position of a subatomic particle and simultaneously predict its future motion. In essence, every measurement made of a system subject to quantum mechanics degrades the knowledge obtained via previous measurements.
The easiest way to visualize it is to imagine an object flying through vacuum, and someone uses radar pulses to locate it. Now, if the radar pulses were powerful enough to exert some level of force on the object, they would change its momentum by some amount. In other words, every time someone measured its position using radar, the object's path would be altered, so you can't predict its future path with the same level of accuracy as you can determine its current position. So as you can see, it isn't something that's necessarily limited to the purely quantum level.
The uncertainty in the universe doesn't mean that absurd things can happen. A ball being thrown into the air is subject to gravity, and it would be absurd for it to simply hang in the air on its own or to start accelerating upwards on its own. It would be equally absurd for the ball to fall through your hand when you try to catch it. What it means is that you can't necessarily predict what will happen before it actually happens, though you can predict a range of possible things which can happen, and you can sometimes get a very good idea of what will happen. Uncertainty keeps the universe from being deterministic.
The two-slit experimentWiki
shows that observing something can change it. In other words, uncertainty is affected by the mere fact of observation. Observing something alters its future behavior in ways that cannot be predicted except by more observation, which then changes its future behavior even more, and so on. Another thing to consider is that the results of the two-slit experiment do not predict exactly where an individual electron will hit, but instead, they predict the broader pattern. Even though the pattern is the same, each result is essentially unique, though very similar. If you're observing it, you can predict which slit it goes through, which is not the same as predicting where exactly it impacts. In other words, the uncertainty is based on how accurately one quality is being observed (if you observe its position, you increase the uncertainty of its momentum, and if you observe its momentum, you increase the uncertainty of its position).
One of the reasons I was arguing with Anfauglir is because of his insistence on using the save-state thought experiment to try to prove that starting from the same initial state would guarantee a deterministic result. It took me a while to figure out why it bothered me so much, but I think I can demonstrate it now. So you have your saved-state. The exact position of everything in that universe - including photons - is recorded with an extremely high degree of accuracy - which means it has to have been observed. The uncertainty principle thus ensures that knowledge of the accuracy of their momentum will be extremely low. In other words, by saving the current state of the universe you have dramatically increased the uncertainty of the future of that universe. It doesn't matter if you rewind and try again, the mere fact of having the current position of everything recorded ensures that the uncertainty of its future positions would be very high. Sure, this might only be at a quantum level, but the quantum level pervades everything everywhere, so those differences at the quantum level would affect things at higher levels, and you'd end up with a situation where you'd always have differences in how things played out.
That doesn't say anything about free will, or lack thereof. If you can't possibly guarantee the same result because the uncertainty principle prevents it, then you can't really be sure of what factors created that result.
Also, the fact that there's only a range of possibilities to work with doesn't mean that the exact specific possibility is locked in. If there's any freedom to make decisions, that's where it must be.