Ok, sure, more exploits in Chrome. I suppose that could be the case.

But a very important thing is how big? 15 exploits that let you crash the browser, compared to 1 that lets you root the target... I'd rather take the first option on the user end.

Even if all those failed, there's still the containment building which remained intact- the part that blew up was an external shell that doesn't matter in terms of containing a meltdown.

Worst case, they'd have a pool of molten uranium encased in a concrete shell.

I voted slightly less safe, I expected the cooling failsafes to work without power. Apparently, on newer reactors, they do.

That being said, even if the cooling had failed completely, including the emergency seawater flooding, with the main concrete containment structures surviving, the worst case scenario might be slightly worse than Three Mile Island(which destroyed a reactor, and that was about it). Sure, there's a bit of radiation released, but the toxic chemicals that could be released from a fossil fuel plant going up would probably be at least as bad.

They should look at upgrading the emergency cooling systems to newer standards that are gravity fed, that would have made this nearly a nonevent and might have even let them bring the damaged reactors back online, which doesn't seem likely now. A review of the containment structure design would also be in order- the spec called for surviving an 8.2 quake, this was an 8.9- significantly more powerful. They did survive so they might not need upgrades, but I'd still review the design to be sure they weren't just incredibly lucky.

All in all I'd say that this shows that while nuclear power has risks, these risks can be managed through a solid design and well trained operators.

It's not so much that milspec stuff is better, it's that the extensive testing is supposed to ensure a level of quality to a high level of confidence.

Sure, the regular civilian stuff might be superior in nearly all situations, but you don't want to be finding the one situation where it is not while on the battlefield with people trying to kill you. Sure, the odds of finding that situation at all might be smal, but knowing the odds, which those test seek to ensure, is a huge help in managing the risks.

At least, if they restrict this to those serials that were originally shot in color. I would be a bit uncomfortable if the older, black and white originally, serials were colorized.

Windows Service packs, at least cumulatively over the course of a major version, are pretty close to equivalent to Mac OS X new versions. And they are free upgrades. Even if you buy the Professional Edition full install, your wallet will probably come out ahead of a Mac users over the same timeframe, even if you both got your hardware free. Of course, value for your dollar is another matter, and different people will call this different ways and that's fine.

Windows needs an infrastructure focused release too. Like OS X up through Leopard, there's some ugly that has collected and they need to focus on that. Windows needs a Snow Leopard style infrastructure update. Windows 7 service packs should probably focus on this, and hand out new user facing features when Windows 8 ships.

Exactly.

I buy that post-9/11, a review of the balance we have struck between freedom and security was warranted.

However, the freedom side of that is vital. Yes, if we don't look after security, anyone will be able to come in and take any freedoms we theoretically have. But without freedom, all that security has exactly no point whatsoever.

Reviewing the balance is fine. But it must be an open, public debate with equal weight given to both sides. Working from an assumption, made before the review, that security must improve, is ridiculous and dangerous.

At least as applies to simply using the protected work.

I wouldn't be shocked if it reaches the Supreme Court and they uphold the ruling in the most narrow manner possible, leaving fair use copying uncovered while permitting simple use and viewing.

That doesn't necessarily mean the Court says fair use shouldn't be covered, just that it was not explicitly part of the original case so it does not get answered in their ruling.