No the US never admitted it. The White House said it wasn't being done at present (in 2013) and wouldn't do it in the future, but refused to comment on whether it had been done in the past.

Also, it's not a crime for countries to spy on other countries. More precisely, there is no international law making it illegal. Countries have laws against spying, but German courts can't usefully prosecute the the United States. Germany could prosecute the specific people who installed the taps, assuming they could identify and get hold of them (the US obviously wouldn't honor an extradition request).

The whole notion of "crime" is simply irrelevant here. This is a question of international relations, where countries punish those they believe do them ill by other mechanisms, ranging from subtle snubs to nuclear war, with lots and lots of gradations in between.

The ones who see well into the future are the ones that create and operate the lotteries (and casinos, etc.), not those who play them.

Nonsense.

You're conflating several issues. In fairness, this misunderstanding is common.

What happened in the Lavabit case was completely different. There were no NSLs or gag orders, the sequence of legal events is in the public record, and shutting down was Levison's own decision. What happened was that Levison not only refused to comply with narrowly-targeted orders for one particular individual's e-mail, but lied, saying that it was impossible for him to comply. His bad faith ultimately resulted in a court order to turn over the keys, because the FBI successfully argued that they couldn't trust him to provide the requested material, and so had to get everything. I think the court was wrong, but there's no reason to assume that a company that employs attorneys and fights orders in the court system, rather than trying to fool the court, would see any similar order.

Er, the Canadian arctic, I meant to say.

How about additional ice core evidence from Antarctica, the Canada arctic and South American mountains?

From here:

That paper is from 2000. More recent research has further confirmed the evidence of rapid, global change within the recent geologic past.

Unfortunately, the tech companies that didn't cooperate with the NSA are getting it, too.

The terms on those loans aren't kept secret from the students signing up for them. There's no need to rely on hindsight, just a little basic mathematics. As another poster mentioned above, the median cost of a US university education is just under ~$5800 per year. If you're racking up $40K, $50K, $100K or more of debt it's because you're making expensive choices. Go to an inexpensive school, work a part-time job during school, full-time during summers, and live cheaply, and you can graduate with minimal debt -- and learn some valuable lessons about life and responsibility while you're at it.

FYI, you're conflating several things.

NSLs cannot compel turning over keys, or even data. The law that authorizes NSLs limits them to metadata (granted that metadata is still important data).

Lavabit was not compelled to turn over its keys by a NSL, but by a court order (not a secret one, either). Whether or not that order was justified is a subject of debate, but the FBI got the order by successfully convincing the judge that Lavabit was being deliberately obstructionist by failing to comply with previous, appropriately narrow orders. Lavabit did appear to act in bad faith with regard to previous orders.

Google (including GMail) doesn't sell or give your info to others. So if that's all you want, you don't even have to pay to get it.

Not really. Targeting is harder and more expensive to do well than non-targeting. Advertisers really don't care whether they're buying targeted or untargeted advertising, they just want a good return for their advertising spend... it's the same to them whether their dollar of ad spend that generates two dollars of revenue is doing it by displaying a dozen carefully targeted ads or ten thousand untargeted ads.

All of this means that advertisers and on-line ad services are just as happy to use and deliver, respectively, untargeted ads. So why are targeted ads so popular? Because users prefer them. Specifically, users prefer fewer ads and less visually-intrusive ads. This means site owners prefer fewer ads and less visually-intrusive ads. This means users and site owners prefer targeted ads over non-targeted ads, because achieving the same ad effectiveness without targeting means lots more and bigger ads.

Remember what on-line advertising looked like pre-Google? Blinking banner ads everywhere? For that matter, take a look at the typical "36 weird ways to X" web site, with it's massive number of ads per page and content spread out over 40 pages. That's what untargeted online advertising looks like. There are exceptions, because some sites are so narrowly targeted that advertising on that particular site is all the relevant advertisers need to do. But that only works with narrowly-focused products on narrowly-focused sites. In all other situations, untargeted means massive ad volume.

I don't want to see the web go that direction. If we want an alternative to targeted advertising, it should be paid services. Untargeted advertising sucks for users.

Or, tweak Chrome with:

new tab, type "chrome://flags"

find "enable-npapi" and set it to "true"

It will be possible to enable NPAPI in Chrome for some time yet. The reason for disabling it by default is to push plugin vendors to port to better approaches that don't leave your system security at the mercy of whatever web page you happen to hit.

No, we don't. We know what approximate shape it is. Our approximation of its shape is limited to the accuracy of our measurement instruments. That accuracy gets better as we learn more, and thus we continue getting a clearer and clearer understanding of its shape.

This is exactly how the whole of science works. It's an asymptotic approximation to the truth. We never achieve perfect knowledge, but bit by bit we weed out errors in our old understanding and obtain a more accurate -- but still erroneous! -- view. At some point in every area of scientific research we achieve a sufficiently precise understanding that we can use it to engineer all sorts of things, and perhaps even get to the point where further improvements in our understanding are of only academic interest, because they make no practical difference in how we live or what we build. But that doesn't mean we've achieved "truth", merely that we've approximated it closely enough for our needs.

And sometimes the level of knowledge that used to be good enough is no longer good enough, and so what was academic actually becomes important and relevant research again. Our knowledge of the shape of Earth in the 19th century was good enough for effective global navigation, but when we later started putting satellites up we needed more precision. The satellites themselves have provided even more precision, and it's not inconceivable that that precision may soon become important to climate modelling, if it hasn't already.

Science is always moving forward (with occasional diversions into dead ends and sometimes a little backtracking), but never arrives.

Greenland ice core records show that within the last 100,000 years the planet saw a temperature change of up to 10C in as little as 40 years. That's 2.5C per decade, not 0.15C per decade. And while that's the most extreme event in the history we can see, it's not at all isolated. There are many such extreme, rapid changes, and they don't correlate with CO2 level changes, or any other obvious cause we can find.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC34297/

So... don't assume that if we can only stop from altering the climate ourselves that we'll be safe from rapid climate change. We won't. We need to learn how to deal with it, either by adapting to changing climate or (far, far better, IMO) by learning to engineer the climate, to stabilize it and prevent rapid changes, whether anthropogenic or not.

Keep in mind that "natural" climate change can also happen with great rapidity, including at least an order of magnitude faster than what we're seeing. Greenland ice core records show that in the past 100K years we've seen a shift of as much as 7C in as little as 30-50 years, possibly less. That's ~0.15 C per year, rather than per decade. And the ice of the period shows no great increases in particulates or CO2, or any other obvious cause. As far as anyone can tell, it just happened.

Note that I'm not arguing against anthropogenesis with respect to the current warming trends. I'm sure at least some if not all is human-caused. But at the end of the day it really doesn't matter why it's happening as much as what we're going to do about it. And anyone who thinks we're going to halt it just by reducing CO2 emissions is living a dream. There's no way we're going to be able to cut emissions enough, fast enough, even if it is the only cause. We need to start thinking hard about how to actually cool the planet.