Do you know the rational behind having override-overrides from the cockpit? For me, this only makes sense under the assumption that Prob(malicious agent who knows emergency codes in cockpit and good agents knowing codes outside ) is much smaller Prob(good agent knowing codes in cockpit and malicious agents knowing codes outside). How is this assumption justified? Under current regulations, the latter is becoming more and more unlikely. Or this could be fixed by letting the protocol to lock the door from inside the cockpit depend on an exchange with a trusted third party like air controllers.
[edit: language]
"real life" as in "they can do it for real in a lab": Alice has two photons, applies a quantum operation to them so that they become entangled, and gives one to Bob. Bob received "quantum information" from Alice. They can use this resource (entanglement between the photons they own) to perform several tasks now, like "teleportation" of the state of a particle, or secure key distribution.
The event is neither in my causal past (it has no influence over the current state of my body) nor in my causal future (I have no influence on it). It's simply uncorrelated with me. If now you remind me of the number again, it becomes part of the causal past of my (new) body. Analogy: if a dwarf dies in a fortress far far away and you don't hear about it, her death is neither in your past nor in your future. You know nothing about her state: in QM, you would say that your brain and her are in an uncorrelated, product state, something like |dunno><dunno| x (|dead><dead| + |alive><alive|)/2 .
The x stands for \otimes, tensor product.
1. When you tell me the result 6296, my brain becomes only classically correlated with it, not entangled. The source of randomness (whether you got it through a quantum experiment or not) does not matter here, as I am only receiving the classical information.
2. After I forget it completely, all I can say is that I (my current body) am not correlated with the event --- but there is no reason to think of the event as being in my future. It's simply not correlated to me any longer. The process of forgetting means dumping all correlations with an event in the environment. For instance, neurons interact with blood stream that interacts with lungs, passing along those correlations to some air particles. So for my current body, the event never happened, although you might have written the number down and will always remember it. In other words, the past is relative.
Good answers. Just a small comment: quantum entropy is a generalization of classical entropy. In particular, if A and B are classically correlated, then S(A|B) has all the properties of a classical entropy (Shannon's, in this case), for instance, it is non-negative.
Performance aside, here is what I hate about Skype compared to MSN / WLM: privacy.
(1) Skype logs you in to your last known status / on MSN you could choose before login (for example, "appear offline"/"invisible"). If I just want to check if a contact is online without my aunt start chatting, I can't.
(2) There is no easy way to block whole contact groups temporarily. When I'm at work, I don't want my friends to see me online; when I'm at home, I don't want to be contacted by co-workers.
(3) Say I'm chatting with Bob, and he goes offline just before I send him a last IM. This IM will be delivered the next time that we are both online. Even if one of us is "appearing offline"/"invisible". So we have a way of knowing when someone is hiding: just send someone a message when they are offline. Bah.
(4) Not related, but the delivery of old messages is awful across platforms.
What annoys me the most about these bugs (1-3) is that they seem so easy to fix, and yet nothing is done. Meh. Do you have similar frustrations?