So you make a quantum mechanical system which evolves over time and which only reveals the correct key if observed at the correct time. Observing it at any other time erases (parts of) the required information. Practically difficult to make if we're talking about delays longer than picoseconds probably, but the problem specification didn't include a timescale.

AFAIR, you're not allowed to plan a flight with less than 30 minutes of fuel left at landing. So this is for very short hops indeed. Source: Have a (mostly unused) pilot's license.

a man has a 1 in 9 chance of being murdered during his lifetime.

And usually towards the end of his lifetime.

Navier-Stokes emerges as a continuous model from a classical particle model. Even a direct simulation of the particle model, allowing all particlas to interact with all others, is only O(n^2). A quantum mechanical simulation is more like O(2^n)

anyone can register an account?"

FAQ says:

Can somebody please reexplain this in a way that a dumb child would understand?

The transmission was not shut down at 01:07 as I understand it, that was the last automatic transmission from the engine system. These are half-hourly, so you wouldn't expect another one if the plane disintegrated at 01:21.

A curious thing is that the last contact with the pilots was a handover from one ATC to another. The Malaysian ATC told them to contact Vietnamese ATC, which they acknowledged but never did. Normally, you'd do that right away (I think).

You could argue that patents are unneccessary because all innovation could be directly funded by government. That's another discussion, though. The innovations mentioned in the brief were things like http, gnu/linux and hadoop. One of these was government funded, the two others are knock-offs of things that to the best of my knowledge originated in private companies (unix and map/reduce). Of course there is an exchange of ideas between academia, private companies and open-souce projects. All of these also come up with new ideas, but the questions are: at which rate, and how would these rates be affected by the abolishment of software patents in the USA? I'm not at all convinced that the overall rate of innovation would suffer, but the argument in the brief does not contribute to that one way or the other.

That is a valid example of open source innovation, but can hardly be used in an argument against software patents. If we want a software industry, i.e. companies whose investment in software development isn't recouped through hardware sales, we can't go back to the business model of the 60s. And I think we all want a software industry.

The reason why the very start of a field is important to this debate is that this is where innovation happens, and the main argument for software patents is that it allegedly fosters innovation. That's also why the very first origins of httpd matter when it's being used as an example. Those things that originate in academia are usually not patented, since academia has a culture of publishing without patenting.

The important thing here is not the number of users or developers of a product, but its degree of innovation. As mentioned earlier, innovation also happens on a smaller scale within projects all the time, but if open source is to prove that software innovation doesn't need software patents, as the brief claims, then it should be possible to find widely known examples of innovative open source software. Maybe CMS is one, in which case it should have been mentioned in the brief instead of projects that are knock-offs of various patented and government-funded work.

That is evidence that software innovation can happen without software patents. The success of open source (at least as far as it has been exemplified thus far) is not.

Which fields have started off as open source? You mention CMS and web servers. Web servers started out open source, but since that work was government-funded it's a bit tangent to a debate about software patents. I don't know the history of CMS. Maybe that's a good example.

On the other hand, the argument that patents cause innovation is also false, if based only on the observation that innovations (and many things that aren't) nowadays tend to be patented.

I agree that GNU/Linux are high-quality pieces of software and that innovation takes place in large and small ways within every project, open source or not. My observation, which I think you will agree with, is that historically, most software categories (say, word processors, database servers, operating systems etc) start out as proprietary and often patented programs. Usually, several proprietary versions get produced before the FOSS versions start to come along. This is just an observation, and it doesn't mean that truly new software wouldn't get created without patents. The point is merely that you can't just point to some successful FOSS projects as proof that patents are useless, since the proponents of software patents will shoot down that proof unless the FOSS projects in question are truly new and original.

Better examples to support the argument could possibly be found by looking at software innovation done in parts of the world where software patents are not allowed, like pretty much everywhere except in the USA. Unfortunately, most examples of software innovation that comes to my mind are from the USA.

As to my claims about government funded web servers and browsers, I had httpd from CERN and NCSA's Mosaic in mind.

Ok, so there is a open source "unix", open source "office", open source "photoshop" and more or less popular and successful open source versions of just about any common piece of software. The brief specifically mentions GNU/Linux as an example of the overwhelming success of open source in the software industry. The argument, as I understand it, is that we don't need software patents because innovative software gets created anyway, like GNU/Linux. But for all its qualities, GNU/Linux wasn't innovative, its stated goal is to create a knock-off of unix. There are better examples of innovation in the brief, like web servers / browsers, but these were initially government funded, and the government has never needed to patent its inventions.

There are good arguments against software patents, but "the overwhelming success of open source in the software industry" is not a very compelling proof, IMHO.

You seem to be confusing a map and a business directory. OpenStreetMap has a lot of map detail, i.e. street names and such, but no business listings. Google Maps is not only a map, but also a business directory. Certainly, these are useful services to combine, but I wouldn't fault OpenStreetMap for not being a business directory. That would be better handled as a separate project, and the two data sources could easily be combined to produce the service you're looking for.

and Koreans is way higher than the employed (page 226). What's going on there?

The proportion of foreign users of each of these services is GREATER than 50%, so even this infinitessimally more strict criterion is fulfilled. Whether the NSA actually uses such a pedantic and disingenuous interpretation of this rule is, of course, another question. If they are actually looking out for foreign threats, then they shouldn't need to. But you would think that they, with all their access to information and analytical power, could easily set the bar higher than 50% if foreign threats were the only ones they were interested in.