It got him an off-hour story on Slashdot.

Really? Seems to have been a pretty winning strategy for Google. My anecdotal experience involves hiring only a small handful of people, so I wouldn't expect to draw any serious conclusions, but Google's experience surely can be used as a guide.

Ultimately, when I'm hiring a new person, I want them to be someone who likes working on solving hard problems that may or may not have solutions, and that most certainly includes thinking of new and original ways of looking at long-held beliefs. Having been exposed to brain teasers as a child is a good way of developing those skills.

Simple with a BSD license you can check the code yourself, build the code, and then deploy the code using a signed encoded DFU file with a custom boot loader.

I disagree that it is because java is easy to read. Java is easy to write. A good programer can write an app in Java and have it work really well. A bad programer can write an app in java it will work.
With C++ a good programer can write an app and it will work but you really have to watch for a lot of gotchas. A bad programer can not write a program that works in C++ because it will leak memory, stomp on memory, and have issues with pointers.
Java is better at stopping the little brain farts from blowing up in your face.
C++ is a lot more fun to write in IMHO.

The point of brain teasers is not to prove you're clever enough to know the answer, but to ask a question that you might not have heard before and observe your reasoning and explanations. While the North Pole question is cute, and most interviewees would know the question (at least I hope so), being able to answer it indicates not that you are smart, but that you have a certain kind of background that leads you to have been exposed to such things. Now if we continue with that assumption, then there are other questions that are worth asking.

My personal favorite question is: Explain the answer to the Monte Hall problem in such a way that a high school student could understand it.

A lot of people know the answer to the Monte Hall problem. Most people are confused by it, or get the answer wrong, but let's concentrate on those who know the answer or can figure it out on the fly. A few of them can cogently explain the reasoning behind the correct answer. Even fewer can explain it in such simple terms that a teenager could understand it. Those are the people I want to hire.

OTOH, using "roll your own crypto" is nortorious for individualized holes and weaknesses. It does tend to mean that the "one size fits all" means of breaking the code won't work, however. Or at least may well not work.

That said, if you have good enough communication to share custom crypto programs, you may be better off using a one-time pad....as that can't even theoretically be broken. But it does require a good source of random numbers (e.g. amplified triode vacum tube with no input so you're just amplifying noise). Such things are reasonably easy to build, but for some reason they aren't normal computer accessories. (Video cams watching a flickering flame are another good source.)

But custom crypto is hard to do correctly. AND it requires good communications to standardize the programs. So if you have the communication, a one time pad is better.

Trucks and buses.

There's a stretch of separated two-way road near me in an urban center. Because of the particulars of the roadways around it, one direction is used almost exclusively for buses. The other direction almost exclusively for cars. The road surface until recently was made of brick, a not-very-good choice for road surfaces as it is particularly fragile and needs near constant maintenance. But Holy Surface Deterioration, Batman! The side of the road with the bus traffic was easily ten times worse than the side with the car traffic. And that's despite there being far fewer vehicles passing on the bus side than on the car side.

Heavy vehicles do most of the road surface damage, and that includes buses, at least in urban areas. I'd wager that the ultra-light vehicles like the Cooper Mini and Smart cars do almost nothing. Taxation should be proportional to induced damage, in a pay for what you use scheme, with a baseline offset because even a bicycle rider benefits from the road existing even though bikes likely do not contribute to its deterioration. And, yes, we should tax bicyclists for road use.

Well, with electronic toll-paying that could work, but it would still shift the burden from low MPG to high MPG cars.

The great thing about a gas tax is that it's a simple way to kill two birds with one stone: encouraging higher mileage and paying for infrastructure. The problem is that not everyone agrees that both birds are important. Two-birders think that high mileage vehicles should be discouraged because of externalized costs -- pollution mainly, but also space required in parking lots, greater risk to other road users etc. One-birders don't care about externalities but understand that the roads and bridges need to be repaired. Zero-birders are just idiots.

I'm a two-birder myself, so raising the gas tax is a no-brainer. I'd also issue everyone a flat rebate per driver, because in fact I'm a three-birder: I'm concerned about the effect of a regressive tax on the working poor who have no options but to drive to their jobs.

But I'm also a realist. There are a lot of one-birders out there and the roads need repair. It's also politically easier in one-birder territory to sell something as a fee rather than as a tax, even though from my perspective that's an irrelevant difference if you're raising the same revenue either way.

Well, they're already opting to have damaged natural joints like hips and knees replaced. That's a case of upgrading from natural to artificial to gain function. As the performance of artificial limbs increase, it might become an increasingly commonplace treatment for older people, just like knee or hip replacement.

If we project that trend forward for twenty or thirty years I wouldn't be surprised at all to see artificial legs that outperform natural legs for the purposes of walking or even running. But I don't think people with normal abilities will be trading in their limbs just to be able walk a little longer, run a little faster, or carry more weight. That won't happen until the replacement is subjectively indistinguishable from the real thing; until you can feel the grass under your toes.

I'm comfortable predicting locomotion parity in the next fifty years, but I wouldn't care to speculate on when we'll see sensory parity.

I have no problem with going after people who steal trade secrets, anything more than I have a problem with going after people who steal nuclear secrets. The only thing is that the FBI has a long history of racist paranoia about Chinese scientists, from Quan Xuesen in the early 50s to Wen Ho Lee in the 90s.

Rhwew may well of a legitimate case against these guys and if they do I hope they nail the bastards. But I'm not jumping to any conclusions based on FBI say-so.

You're the one worried about passwords that can be broken in 25 years; that's a non-issue. The issue is security that works well enough for long enough and is workable for the users. Impressive sounding, inflated requirements means something else has to give: price, performance, or usability.

Well, once you've cracked the VPN traffic the password is almost a secondary concern, isn't it?

This is the wrong way to think about security, e.g. for a hypothetical world where users adhere to anything you demand of them no matter how intrusive or onerous that is. In reality if you decide that usability and convenience aren't factors in your planning then that's actually an oversight which will come back to bite you on the ass someday. The only thing you can say for that approach of wishing usability away is that when disaster comes you'll be able to point the finger of blame at the users -- even though their non-adherence is a predictable result of your poor understanding of system requirements.