... given constantly changing and contradictory specifications, hardware, operating systems, programming platforms and data storage and transmission standards, on unmanaged, obsolete and malfunctioning endpoints, while preserving full round-trip data and binary code compatibility with legacy systems dating back thirty to fifty years, in an international multilingual ennvironment with competing national legal frameworks and data stewardship requirements, over an inherently hostile network, managed by uneducated users, while leaking no data or access rights in the face of organised crime and state-level espionage agencies.

And that's just to begin to meet basic functional square-one user requirements on today's Internet.

No pressure.

... except for everything that is a potential security risk, which is everything that potentially faces the Internet, which in 2013 is everything.

Congratulations. You've just demonstrated in two sentences why the entire Internet ecosystem today is a huge gaping security disaster.

("But my script will never see the Internet! All our workstations are behind a firewall!" Let me introduce you to my little friend called "Bring Your Own Device". Did I say friend? I meant the kind of friend who likes hurting people. Good luck!)

Um, are you sure you don't mean the imperative programming paradigm? Imperative is "do this, then do that" - ie, C - and doesn't scale to parallel. Declarative programming - "this is what I want, I don't care how it gets done" is precisely what modern programming languages don't do, and what I reckon we need more of. Especially if your work involves describing complex interlinked data sets with no particularly well-defined types (hello Internet! hello Semantic Web! hello just about everything on your basic workstation desktop except for the low-level graphics code in your 3D games)

IMO the high-water mark of declarative programming is Prolog embedded in Lisp - but show Prolog to the average C++/Javascript hacker and they'll look at you like you're from Mars. Where are my functions? Where's the curly brackets? What are these "predicate" things? What do you mean, it works backwards as well as forwards? Logic what?

We lost two generations of programming advancement when the AI Winter crashed the Lisp scene. Granted, the Scheme/Common Lisp civil wars didn't help. But we could have had much nicer things to program with than C++. We still could, if we cared.

Racket's semi-okay, I guess; it would be better if it didn't throw a whole bunch of multi-language complexity in your face right away, though. And unfortunately, its implementation of Prolog (actually Kanren) is Not Really Sufficient for the stuff I want to do, owing to massively too busy syntax. Ironically, the stuff I want to do which turns out to be really hard to do in most modern langauges is text adventures.

Oh, so you're just arguing syntax then?

(Sir Tim Berners-Lee kills a SPARQL query whenever someone uses the word "semantic" to mean "lacking information content". Luckily, no-one notices.)

Er, wasn't that part actually first demonstrated back in 1952?

I had the impression the difficult bit was controlling the fusion energy discharge within a continuously-operating reactor.

People here seem to keep assuming that this is a functioning plant and that the original design specifications hold. It isn't and they don't. It's a debris field with lots of temporary jury-rigged power sources running everything.

I'm guessing, but since main power to the reactors has been out since the tsunami, they're probably routing power to the cooling pumps via emergency generators / power cables. And it's probably one of those temporary/emergency junction boxes that lost power rather than the main control panel.

Heck, I don't even know if they have physical access to the original control rooms and wiring for much of anything anymore. This isn't a functioning nuclear plant; it's a wet, rusty junkjard full of random places that can kill you if you stand too close.

Well, I imagine the process flowchart was rigorously overseen and went something like this:

1. Have an earthquake of magitude 6 local equivalent (earthquakes? in Japan? unthinkably unlikely! we can build cheaper if we ignore this contingency)
2. Have a tsunami overtop our seawall (tsunamis? in Japan? on the coast? see above)
3. Lose local, grid and backup generator power (utterly impossible, see above. and obviously flood isn't a risk so we'll put our generators in the basement)
4. Have a meltdown and explosion (this will NEVER happen because it just can't. So it won't. Ever. Full technical refutation: nananana CAN'T HEAR YOU).
5. Run around screaming with our hands in the air (precise details to be left to onsite implementors)
6. Burn out all our trained staff because of contamination
7. Quickly bodge something together to stop the leaking
8. Oh god it keeps leaking aaaaarrrrgh make it stop it won't stop why won't it stop
9. Quick hire some untrained staff and fake their dosimeters why is it still leaking
10. Toss something to the press about billion dollar frozen ice walls. Don't mention the leak. Try to smile. Randomly punch buttons. Is it hot in here?

... after multiple explosions and meltdowns.

You seem to be under the assumption that this is a nuclear plant. It's not. It's a nuclear meltdown site, and the thing about meltdowns is they don't have an off switch. You don't fix them - fixing the problem will require several hundred years for the corium to decay. You don't decommission them - decommissioning is what you do to an unexploded reactor that reaches the end of its natural life gracefully. You sure as heck don't operate them like they're functioning online plants. The best you can do is try to manage them and minimise the releases. And since the only management they can do is pour water onto naked radioactive corium and pump radioactive groundwater out, then try to store it in leaky tanks - all in a mess of jury-rigged pipes, cables and generators with temporary staff recruited by the Mafia who are rapidly burning out their dosimeters - it's not exactly likely to all go according to the textbook, is it?

Well, yes and no. This isn't actually a nuclear plant anymore; it's a nuclear disaster site, full of jury-rigged temporary equipment. Of course they're making things up as they go along, and of course it's all awkward and unsafe, because nobody in the entire nuclear industry has been here before. There really is no long-term plan, but that's not entirely TEPCO's fault because the industry as a whole doesn't have any long-term plan for dealing with nuclear disasters other than to say "they will never happen because we have failsafes, so we don't need to waste any time thinking about that".

Fortunately all that saved time and effort not coming up with worst-case disaster scenarios because they were too far-fetched to imagine is now paying off! If we all just close our eyes, clap our hands and chant "THORIUM PEBBLE-BED LIQUID SODIUM BREEDERS!" this will all magically go away.

Arguably "decommissioning" Fukushima Da-ichi isn't a problem because the reactors are already way, way out of commission. Decommissioning is what you do to unexploded reactors. The problem now is containment (very difficult because of groundwater flow) and then cleanup (a very long-term job).

Actually TEPCO is planning on moving the spent fuel rods from the mostly-unexploded reactor come November, but they're going to want to do it very very carefully. Getting them moved seems like an important thing, but actually doing it is probably the most dangerous part.

And there's the rub. The problem is that there basically isn't any way to clean up a situation like Fukushima where there's a meltdown in groundwater; this has been known to the nuclear industry for decades, and the answer has always been "we know this is an unsolveable problem, but we believe the odds of this happening are so low as to be unthinkable, because we have multiple redundant safety systems." The GE boiling water reactors especially took this philosophy to extremes; they don't have containment to deal with a meltdown because the suppression torus was supposed to make a meltdown impossible.

This kind of tempting-the-wrath-from-high-atop-the-thing reasoning from the industry is exactly why the anti-nuclear-power protest movement started gaining traction in the 1970s. Activists kept asking "yes, your active failsafes are shiny, but what if the unthinkable happens? How will you clean it up?" And the industry kept saying "we can't, but don't worry, it won't! Stop worrying!"

As anyone in IT should know from bitter experience, you can have all the multiple redundant disaster planning you want, but reality always tends to come up with interesting disaster scenarios your plan didn't account for. And "we're betting the worst case will never happen" is your cue to sell stocks and head for the door.

So it's just a really complicated way of selling EVE Online stats?

Ah, that would explain why the entire banking sector collapsed a while back in a heap of unverified, falsified and fraudulent financial instruments. All the incompetent players got competed out of existence during those years of hurly-burly open competition and now only the smartest, most ethical people in the industry remain standing, right?

There's a Mr Shroedinger and a Mr Bohr here to see you about that statement. Oh - sorry. A single Mr Shroedinger-and/or-Bohr.

And a Mr Murphy, who just set both their beards afire to, and I quote, "see what would be the worst that could happen."

... to be really scared.

Like they say, if you want to sleep at night you don't actually want to know how sausages, laws or semiconductor devices are made.