There's nothing magical about nuclear power that makes it inherently more dangerous than other industrial processes.

Name one non-nuclear industrial process that can kill millions and render an area of thousands of square miles uninhabitable for a hundred thousand years.

Just one.

Can't find one?

That is why nuclear power is different from any chemical or industrial process. If you don't, then you get a Chernobyl incident. In fact, it's the cavalier belief that nuclear power isn't different from other industrial processes that caused the Chernobyl incident - the only nuclear reactor in history to have a steam explosion. The plant managers thought they could treat a nuclear reactor like a steel foundry or oil refinery - that you can just "turn it off". They hid their activities from their nuclear commission, and even their onsite inspector. They didn't ask anyone for advice, because it was just a heat source - like coal stove. Then they willfully put the reactor into a state that was known to be disastrous. They had many warning signs, and ignored them. When all else failed, they figured they could just "turn off" the reactor like any other industrial machine. When they SCRAM'd the reactor, they rang their own death toll - as well as killed many thousands of innocents, and will continue to kill them until long after we're all dead and forgotten.

Nuclear power isn't like other industrial processes. It's not intuitive, and you can't just hit the 'e-stop' and shut everything down.

Molten-salt reactors are nothing new, by the way... the first reactor of any kind was the Chicago Pile-1 in 1942 - it had no cooling at all, and was used to create plutonium for the manhattan project.

Liquid-cooled reactors followed, with the PWR, BWR, and molten salt reactors all being developed around the same time, circa 1954.

Conspiracy theories exist as to why PWR and BWR reactors became the standard, especially when molten salt reactors are of the same vintage. Either way: molten salt reactors aren't any "newer" or "more advanced" than PWR or BWR reactors.

It'll be nice to have a Gen IV molten salt reactor - in 2040. I also look forward to seeing the DEMO reactor producing net power before I die. But for now, molten salt reactors are relatively unproven - especially from a safety standpoint.

In the meantime, I want to see Gen III+ reactors like the ESBWR and AP1000 replacing the 30-40 year old reactors we currently have - both of which require nothing in a disaster for 72 hours, and only require topping off the passive water reservoir until cooling isn't required (ie. a week or two). I also hope to see our coal-fired power plants replaced, and nuclear looks like the only reasonable option. We'll probably need all of the wind & solar power we can get as well.

Suppose the Hindenburg accident had never happened and the hydrogen blimp survived its perilous infancy, only to have several spectacular Hindenburg incidents decades later, on much safer designs replete with the benefit of experience and refinement, but also against a much larger operational fleet.

I've got a better idea... let's go with the fact that the Hindenberg went up in a very dramatic, very public fireball.

Now, let's look at current R&D. Hmmm... you know what? There are currently rigid airships being designed that use Hydrogen as a lift gas.

You see, Helium has a few fatal flaws: It's horribly expensive, it's in very limited supply (the world is likely to run out of helium entirely by 2030 or so), it weighs more than Hydrogen, and it lifts significantly less.

Hydrogen, on the other hand is abundant, cheap, is a better lift gas, being around 30% more effective than helium as a lift gas, and, in a serious enough emergency, can be used for fuel. The flammability of hydrogen ranks fairly low on the list of dangers on an airship, and one that's manageable by comparison.

The problem with Airships wasn't unique to the Hindenburg. The US had a number of large airships, which were helium filled: The Shenandoah, The Los Angeles. Even larger were the Akron, and the Makron, which were aircraft-carrier zeppelins.

Only the Los Angeles survived to be decommissioned; the rest lasted less than a couple of years and were lost. The idea was scrapped because the technology was not viable; the Hindenburg was a flashy disaster, but was no more a failure than nearly every other heavier-than-air airship in history.

The big problem with airships has little to do with the lift gas, but with buoyancy control in general, combined with inability to weather storms. Air density changes constantly; this requires a change in the volume (and hence buoyancy) of the zeppelin's gas bags.

We're still working on reliable buoyancy control.

PWR and BWR reactors are the main operating principle of the reactor - in both cases, obligate water cooling for a full week after you slam on the brakes.

Every nuclear reactor requires cooling for a full week after you SCRAM the reactor. It's as unavoidable as gravity. Every nuclear reactor produces heat for a week after all uranium reactions have stopped. This happens in every Uranium reactor - no matter how it's cooled, no matter how it's designed.

Critics point to water's tendency to evaporate in a nuclear reactor in an emergency, which their pet coolant doesn't do. The thing they conveniently ignore is that water is trivial to obtain should more coolant be required. In the end, every reactor drives a steam turbine, meaning every reactor has to have a ready supply of water to make the steam. This in turn requires each reactor to have a large supply of water - like a river, lake, or ocean. In the event of an emergency, additional "coolant" water is already at hand - perhaps not as pure as desired, but it's still there to prevent catastrophe.

Even after a devastating earthquake ten times more powerful than the reactors were designed to withstand, followed by an even more horrific tsunami, replacement coolant was being put into the Fukushima reactors within 16 hours. That wouldn't have even been possible with any other coolant.

And even with other coolants... Fukushima wouldn't have been pretty. Half-meter wide cracks in your coolant tanks tends to deplete your coolant quickly, no matter what coolant you have.

Whoah, that's weird... I thought I shouldn't have moderator access on a thread I'm a part of! Slashdot! BUG!!!

Anyway... for me, the proof that sane management of nuclear energy is made impossible is simple:

The 30+ year moratorium in the USA on processing of nuclear "waste", which is still over 98% unused fuel. A nuclear engineer I talked to likened it to throwing away a gallon of gas after removing an ounce. France, in contrast, reprocesses over their waste, and as a result has dramatically less (I read ~80% less) waste per reactor than in the USA. That moratorium is thanks to an uninformed environmental lobby, and has only recently been lifted.

The assertion that nuclear energy should never be used until it's perfectly safe. It's an asinine requirement and it's impossible to have any sort of sane management when the requirement is as insane as falling upwards. Perfect safety is impossible for any human endeavor, including things we have to do, like breathing.

In other words, you can't have a sane policy when it's being created at the behest of the inmates of an asylum. (I'm actually including both the nuclear companies and the anti-nuclear activists, as both suffer from delusions of competency).

Reactors that aren't cooled by water are essentially in one of two categories: Lab and military.

A number of "silver bullet" nuclear technologies have proven to be non-starters because they have problems that were unanticipated or swept under the rug until it came time for certification.

I recall a few years back "Pebble-bed" reactors were hailed as the solution to everything; they were helium-cooled, so nothing bad could happen. They also conveniently ignored that a coolant leak can result in air getting into the reactor - at which point the graphite pebbles ignite, burn down, and you end up with the uranium inside having nothing to separate it - leading to an unstoppable meltdown. The pebble-bed is a reactor design whose its proponents claimed that the reactor would safely shut itself down in any emergency... except, of course, one of the most likely - a broken coolant pipe.

Your idea of "distributed, sustainable, and carbon-neutral" ignores reality, as does your concept of what is "modern". It's even more mythical than the liquid-salt nuclear reactors.

Let's look at the ages of the methods we have to generate power:
- Burning Fuel: Predates history.
- Hydro Power: Ancient technology, used since the 6th millennium BC in China. Inherently "centralized", devastating to aquatic ecology.
- Wind Power: Ancient technology (17th century BC in Babylon). Due to limits placed by prevailing winds, wind isn't decentralized either. Serious NIMBY and ecological problems as well, as areas with wind turbines are devastating to avian populations (and their prey).
- Solar Power: Older than fire; The most "modern" tech is photovoltaics -- which are horribly inefficient, and require massive amounts of fossil fuels to produce. There are "hopes for the future", but the same can be said for Fusion.
- Nuclear: Less than a century old, capable of producing stunning amounts of power with very little waste if managed in a sane manner. Unfortunately, sane management has been made impossible by anti-nuclear activists and their propaganda.
- Fusion: It'll be ready any day now, just like "clean" photovoltaics.
- "Biofuels" - I'll let you know my opinion of these when I stop laughing. So far, biofuel production at scale are currently complicated ways of creating a liquid fuel using processes that require burning massive amounts of coal. "But in the future!" you say... well, in the future we'll have Fusion too.

The idea of distributed power isn't about generating power; it's anti-corporatism - efficiency (and environment) be damned.

In all honesty, right now, we need to use every tool in our toolkit -- including nuclear power, including cleaner coal.

I don't see how it makes sense to incentivize a power facility to continue running its legally grandfathered, old, horribly polluting coal plant instead of building a new coal plant that emits a fraction of the emissions - but that is precisely what the "green" lobby is doing.

Hard drives don't have the property of being able to depopulate cities should something go wrong. Hard drive failures don't kill or maim kill people. Hard drives don't remain dangerous for hundreds of thousands of years. The same goes for a computer using *any* architecture.

Electrical and Computer engineers aren't generally required to be certified as "Professional Engineers" in the way Civil or Mechanical engineers are for precisely this reason - when they make mistakes, their designs are incapable of killing large numbers of people.

Nuclear power is almost infinitely more dangerous, and has an almost infinitely greater potential for destruction - hence, there is a lot more scrutiny and certification to any design.

It's simply not prudent to treat nuclear power the same way you treat a hard drive - a nuclear reactor has a design life of 30+ years; a computer is generally only expected to last until its warranty expires - and even then, disposing of a computer is effortless in comparison to disposing of a used reactor.

The fact that PWR and BWR have a history that stretches back decades doesn't mean a new water reactor isn't "modern". PWR and BWR reactors are the main operating principle of the reactor - in both cases, water cooling.

Complaining that the new reactors are also water cooled is a lot like saying a car's engine can't possibly be effective or safe because it's based on the century-plus old principle of a piston-driven combustion cycle.

Going with the new for the sake of 'newness' ignores a solid foundation that has withstood the test of time.

There are advantages in using modern evolved PWR and BWR reactors - namely decades of refinements and operational experience with the design, as well as technicians that understand the reactor, and safety issues involved.

Any admin that has to manage a large number of machines with common authentication is INSANE to use /etc/passwd or /etc/shadow. It's FAR too much work, FAR too much trouble, and will cause FAR too many problems.

I'd bet the farm they're using some combination of Kerberos + LDAP. Kerberos+LDAP is the standard for holding the authentication and user information - even Microsoft uses it in their Active Directory.

PAM = Pluggable Authentication Modules. This means you just use a Kerberos+LDAP plugin for authentication, and the user/password database doesn't even exist on the servers in the network.

Thing is: Kerberos doesn’t do anything unencrypted; passwords are never passed between machines. The client and server machines are able to authenticate each other as genuine, as well as authenticating users.

LDAP is easily (and properly) configured to use X.509 certificates and encryption as well - so nothing is ever passed "in the clear".

I'm very interested to learn exactly what happened.

Why do you just assume the passwords are cleartext? Do you not understand that even hashed or crypted passwords are easy things to crack?

There are numerous programs specifically designed to crack passwords. They start with the obvious things (dictionary), then adds in misspellings, moves on to simple substitutions (1=l, 4=a, @=a), and finally to more difficult things, like multiple words+substitutions. Many crackers combine clustering with GPUs, which makes for highly effective cracking.

They are terrifyingly effective - primary because password policies that require things like 10+ characters, a number, and a non-alphanumeric character -- are often almost trivial to crack, because they're still based on a word - something like 'h@w7n3s5' is cracked surprisingly quickly.

If the authentication database (even encrypted or hashed) is comprimised, then the only sane course of action is to expire ALL passwords immediately. An admin has no way of knowing which passwords are "safe", so the best solution is to ensure that no cracked password will work for the attacker.

The theory wasn't entirely unsound. Nearly every air to air victory in the past decade has been via missile.

The problem was that during Vietnam, as many as half of the missiles didn't even ignite - they just fell to the ground inert. Of those that lit, fewer than half did anything other than fly in a straight line. And even if you had one a missile whose engine lit, the guidance tracked, and it was near the target - it's all for nothing if the explosives don't detonate. Pilots often ripple-fired 4-6 missiles at once, in the hope that one would work.

The reliability gained by almost 50 years of air to air missile development since Vietnam has changed the situation considerably.

You're putting a lot of trust in a propaganda piece from a country with a perfectly reasonable score to settle. At this point, I'd say the Iranians are (and should be) playing the propaganda for all they can, espescially on the home front.

If you look at propaganda from the past century, you'd notice that the pattern is similar: Make a claim that "your team" outsmarted the enemy, and that's the reason for an event. It's pure BS, but that doesn't mean it won't play well for your allies and at the homefront.

These drones aren't purpose-designed to spy on low-tech countries. They're designed to be able to go into far more advanced countries (ie. Russia and China) that are more than capable of toying with GPS, and in fact, where altering and jamming GPS is expected.

Electronic warfare is 70+ years old at this point, and the questionable reliability of radio signals (espescially in the face of jamming) are well known. Being able to complete your mission with massive radio jamming is a requirement for anything that flies for the US military, and has been for generations.

We don't know whether the drone had inertial navigation or not - given the size and cost involved, there's no reason it shouldn't have inertial navigation - in fact, I find it hard to believe it doesn't use INS. US cruise and ballistic missiles use INS, as do warplanes - specifically because it's well-known that GPS can't be counted on.

If the drone does have INS (as well as terrain mapping, or both which is what I suspect), it's extremely unlikely that the Iranian story has even a hint of truth to it.

That’s why I dont' believe it. Were it less comedic, I'd have no trouble with the story.

It's great political theatre, but it doesn't line up with what I know about the technologies involved. I've built robots for the military. While gimmicking GPS is doable, it's only a piece of the puzzle - and one that can be easily gamed. It makes for great propaganda, but it overlooks the other systems that have to be onboard for the thing to fly.

Drones have multiple independent means of determining its location, because GPS can't be counted on in a full-on war; it's one of the first things to fall in electronic warfare. INS is cheap, has been used for decades, and still is used because it can't be messed with externally. Terrain mapping is relatively new, but is also cheap to implement, and one of the drone's missions is to scan the terrain (with millimeter band radar).

So unless the Iranians somehow found a way to externally modify INS, as well as shift large-scale sections (at least a hundred square miles) geography of the region in real-time, the GPS story just doesn't hold up.