It is wrong to say science seeks truth, because it gives a false impression of what science is. Science looks for theories that make the most accurate predictions. If a particular theory tuns out to have problems, that's part of the process of finding one that does.

Too many people, including a lot of not-so-good scientists, regard "scientific truth" as something that actually exists. Some experiment, or a journal article, or a hundred years of experience seems to show something, therefore it's true. Way too many people also insist that science is just another kind of religion.

He (or more likely his scientific advisor) has it right. You've mixed up the explanation of it a bit.

The idea with a superposition is that something, usually something small, can be in more than one state at once. If I take your unpowered cell phone camera and expose it to some weak radiation, such as the CMB, some of the molecules in the photo sensitive layer will donate an electron and some won't. Before you look, you can say that each of those molecules is in a superposition of electron-yes and electron-no. That means that each of the electron wells technically contains every possible combination of electrons, which means that the ccd itself is simultaneously in a superposition of representing every single image it can, at the same time. Naturally if you actually measure the image the superposition will collapse into one of the more likely states, with overwhelming probability, a black image with some sparse noise.

You can also say that the photo-sensitive layer is encoding every book that has ever been or ever will be written, in the handwriting of every person who has ever lived.

While it's technically true-ish, at least if you believe certain interpretations of quantum mechanics, it's also pretty meaningless.

Certainly. I beat him to producing this art last time I turned my phone off.

Nope. They're taking an unpowered CCD and saying that the little bit of energy hitting it from the CMB technically puts each sensor site into a superposition of all it's possible states. You can duplicate this art by turning off your phone. And people have been doing so for ages by storing their unexposed film in dark canisters.

If you look at it, it destroys the "art" of course.

Nonsense. I've had leaders of at least three major Christian sects tell me that the bible is the literal word of god, written down by divinely inspired men. Including the pornographic poem in the middle of it.

There are plenty of radical Christians who interpret the bible literally, including the nasty parts. There are just fewer (by no means none) who care to die for their convictions.

Christianity is older than Islam. It had some pretty nasty times, then it mellowed. It seems to be getting more violent again though.

Murder rates per 100,000 per year (2012):

France: 1.0
USA : 4.7

Looks like France is doing something right in the overall department too.

"Queue" means to line up. You "queue up" people. You cannot "Cue up" people. You can "cue people" though, which means to give a signal. In summary, you are the idiot, not the GP. The metasummary is that it's entertaining to watch two ACs call each other idiots.

I wonder how many posts each of those would have if you removed all the ones complaining about the irrelevance of the story.

Let me introduce you to my little friend, the hash table.

Your disbelief is a common human trait. It's why lots of people who play in casinos insist that the can win.

Solving a game means that the consequences of every possible situation are known. You can't beat a perfect player if the assumptions that make it perfect are met. In this case those assumptions are long play heads up limit hold 'em.

You can't determine the robots cards because you don't have enough information. Playing heads up limit poker long term against this bot, the best you can do is tie, by playing a perfect game just like it does. Any deviation from that perfect strategy will cause you to lose.

You aren't smarter than the scientists who created this bot. They brute forced the game.

You know if the robot checks or raises or folds at each opportunity. How do you expect to figure out what its cards are based on that information? We call that an ill-posed problem.

It depends on the game. In a game like rock-paper-scissors the perfect strategy, complete randomness, gives exactly 50% probability of winning. It's unknown what kind of game chess is because it hasn't been solved, but in the limited situation you give the perfect strategy isn't to prolong the game as long as possible. The computer would first look for a way to force a draw and, if that failed, choose the branch that led to the most win or draw end states, i.e., the most opportunities for the opponent to make a mistake.

Heads up limit poker, which is what the story is about, is by definition a two player game.

The house always wins in poker because it takes a rake and doesn't risk anything. That's the most obvious one of all.

In blackjack it's technically possible to play a winning strategy good enough to overcome the house advantage. When people (such as the MIT group) started doing that, the house changed the game. Adding multiple decks reduces the advantage from counting and makes it much more difficult. You can still beat the house by using a computer, but that's cheating.

The house always wins. If they don't, they shut down the game.