In special relativity lets say I'm sitting next to someone and then I go for a walk and come back. When we compare clocks they will be the same since otherwise there would be symmetry breaking and we could establish a preferred inertial frame.

You don't need to factor in acceleration to have the clocks get out of sync, just remember that changing directions means you won't be at rest the whole time in any reference frame:

Inertial frame of your friend: First you move away with say 1 m/s. Halfway through you move towards him at 1 m/s. You're moving, so your watch will be slow when you return.

Inertial frame of you walking away: First your friend moves away at 1 m/s. Halfway through you start moving in the same direction at 2m/s. You're moving faster, so your watch will be slow when you catch up with him.

Inertial frame of you walking back: First you move at 2m/s, while your friend falls behind at 1 m/s. Halfway through you stop moving, and wait for your friend to catch up. You've been moving faster, so your watch will be slow when he reaches you.

I agree it's stupid how browsers show self-signed certificates as more dangerous than plain HTTP.

The difference between paid-for certificates and self-signed certificates means more than just who promises authenticity though: The certificate's signature can be checked against the certificate shipped with the browser, thus preventing MITM attacks.

Basically:

1. HTTP: everybody on the network can read your stuff, including passwords etc. They don't even need to perform a MITM attack. With a simply MITM attack they can also alter content.
2. Self-signed HTTPS: your traffic isn't that easily sniffable anymore, but an attacker can perform a MITM attack to read/alter your data. He'd intercept all your browsers' requests, including the certificate, and replace them with his own.
3. CA-signed HTTPS: an attacker can't perform a MITM attack, because intercepting the certificate request means it's signature won't match with the CA-cert that your browser shipped with.

Thus paid-for certificates mean you won't get MITM'd, the part where the CA also verifies identities is just bonus.

"Always leave a multiple of four plus one" is hardly statistical math/deep AI.
It was only unbeatable because you always let it go first, but at least it seems to have accomplished its task of being fun to program.

FTFY

Also, the reflections are all wrong.

When I read the headline I was hoping someone finally made a Chess Boxing robot.

Maybe they could run a "Windows Malicious Software Removal Tool" every month, and it would remove all software classified as malware by Microsoft?

Article with better picture (the second picture in there only looks like a still from a low rez video of a photograph)

Forbidden

You don't have permission to access / on this server.

Maybe they just took the next logical step after forbidding visitors to look at more than 1 page?

Only the switching is mechanical, there's still current running through the switches so you still get electromagnetic radiation.

No, the Latin plural would be "priora". Toyota has said that in English it is simply "Prius" and also that owners are welcome to use whatever they like. (http://en.wikipedia.org/wiki/Toyota_Prius#Name)

That's way too much if you're planning on scanning the entire population looking for autism, but as this article notes it could be useful for diagnosis:

Although modded funny, parent actually has a good point.

Sure, some people might have a strong preference for a specific drink, but others would likely not really care: they're thirsty, want something to drink from the machine, but don't really care what, and the recommendation is probably easiest to order, so they go with that.

And of course the machine might be influenced a bit by customer demographics, but a smart machine would try to sell more expensive drinks, or the ones it has most left of, saving on refilling costs.

With blackjack, and hookers. In fact, forget the network!

They only solved each position in 20 moves or less, not necessarily in the least possible moves for that position. Since there already was a known position that takes 20 moves this was sufficient to prove 20 moves is the maximum needed, and this sped up the solving by a factor of 1000ish.