2000 posts • joined Wednesday 10th June 2009 11:28 GMT
Hard-wired safety limiters needed?
Surely critical infrastructure ought to be designed so that there are limiter systems which cannot be over-ridden by any computer? (Or indeed, by human operators following any procedure short of using screwdrivers and wirecutters).
I once worked at a synchrotron light facility. I'll spare you most of the details, but the light is generated by relativistic electrons circulating in an ultra-hard vacuum, and the various experiments took light, X-rays, etc. out through "beam lines" which at one end were open to that ultra-hard vacuum. If any air ever got into a beam line, a series of vacuum sensors had to cause valves to slam shut faster than air could travel down the line. If that ever failed there could be expensive damage to repair and days, weeks or even months of down-time while the vacuum was re-established.
What controlled this emergency safety system? Hard-wired relay logic (with its power-fail fail-safe). No digital decision-making. This system could not and should not be overridden. If it tripped when it shouldn't have, it was an annoyance. The converse, a disaster. Relays are fast enough, fail safe on power failure, and have extremely high noise immunity so they never "glitch". The right technology for the job!
One line was operated by a big computer company who refused to use relays. They had a special multiple-redundant computer system doing the job. They claimed it couldn't fail. One day, it did: spectacularly so. The facility was down for weeks and the big computer company was on the hook for all the bills. Hubris and Nemesis at its best!
Abuse of monopoly
I expect that might be viewed as illegally monopolistic behaviour. A hardware manufacturer isn't competing when it supplies competitors. It's just supplying, which is its business. What it supplies, is deterrmined by what it can make and by what its customers ask for.
Yup, that icon looks like a lawyer to me.
"Absolutely no sign" that a super-erruption is looming?
I thought there was quite a lot of evidence that the magma chamber underneath Yellowstone park is currently filling, causing the ground above it to rise. When that ground cracks, all hell will break loose. Of course that's not evidence that it'll errupt next year or in the next 100 years or 10,000 years, but it's rather less reassuring than "absolutely no evidence".
The geological evidence shows a fairly regular 600,000 year cycle for Yellowstone erruptions, and the last erruption was ... about 600,000 years ago. Odds-on the next erruption will be within 50,000 years meaning there's about a 1 in 1000 chance it'll happen in my lifetime.
If the opposition has been smart, they'll not just have constructed their facility a mile underground, accessed by tunnels (multiple) going sideways into a mountain with blast doors and several changs of direction. They'll not just have stocked it with enough supplies to wait out having all the entrances collapsed. They'll also have put it in a massive concrete shell, decoupled from the rock of the mountain by air bags.
Unless you can generate enough of a shockwave to cause the mountain to move by more than the width of the airbags, they'll completely decouple the precision quipment from any shockwaves travelling through the rock. Methinks your chances of damaging such with conventional explosives are infinitessimal, and your chances of managing to do so with half a megatonne of nuclear explosion are slim.
Is USA SAC HQ still under that mountain in Colorado? If so, they think like I do.
I've always wondered why (outside of the movies) no consideration seems to have been given to a launch where the first stage of aceleration uses external electric power to drive the rocket up a ramp. I'd have thought that there are considerable advantages to firing the rocket when it's already travelling at (say) 300mph along a ramp up the side of a mountain.
I find it depressing that people don't insist on ECC even for low-end servers. For anything that's a repository of valuable data, it's really important that the data doesn't get corrupted in transit!
RAM doesn't often fail, and usually fails hard enough to be noticed quickly (before the back-ups have been recycled). Not always, though. LAst year I saw the results of a slightly flakey piece of non-ECC RAM on a busy filesystem, and it was not at all pretty.
Reactor launch not a hazard
Launching a reactor is far less of a hazard than launching a radioisotope power unit. The latter is highly radioactive at launch. The former would be non-radioactive at launch. Its fuel rods (enriched uranium) would be slightly radioactive, but vastly less so than the isotope power unit. They only become highly radioactive once the reactor is activated and has been in operation for some time, at which point we can be sure it won't be returning to Earth.
Give Brown some credit where it's due
He was an idiot in political terms, but not so intellectually. I'm willing to believe that he simply knew where the Euro project would end up, and decided that we'd be part of that shipwreck only over his (politically) dead body. I think history may be kinder to him than his comtemporaries. Blair was the worst PM we've ever had. Gordon was by no means the worst chancellor. If Blair had had his man (boot-licker) as chancellor, it would be the UK at the centre of the Euro crisis, not Italy.
Rather than thinking about mechanisms for throwing the weakest to the wolves (markets), wouldn't the best fix be for Germany to leave the Euro and go back to the D-Mark?
This wouldn't cause currency flight out of Germany and a banking crisis there. Their biggest problem would be preventing currency flight INTO their country. A problem which they'd probably have to solve the same way the Swiss recently did - a currency peg to the Euro, which amounts to buying the weaker countries debt. Isn't this just what the weaker countries need?
Also with the Germans outside the Euro (perhaps along with a few other smallish strong Northern European countries that want to be part of the DM blok and don't mind doing what they are told to do by the Germans), there won't be any opposition to a proposal to allow the ECB to be the lender of last resort to the remaining Eurozone. Of course the Euro will plummet, but better that than the alternative of no longer having a working economy at all?
If that's a photo not an artist's impression ... why is there so much space between the rows of panels? I appreciate you need access for maintenanance and to clean dust off the panels occasionally, but the space they've left seems totally excessive.
I doubt your premise
Inter-species communication is extremely common. Of course they can't talk in a human sense, but ...
Birds "understand" other species alarm calls. It's in their common interest to know that there is a predator on the prowl.
Yellow-and-black stripes mean "leave me alone, I can inject venom". Of course sometimes it's a bluff.
The bee-eater (a bird) can't break into beehives. It has a special song and dance solely for attracting the attention of other species that can. It finds the hive and leads them to it. They break in to take the honey. It eats the bees.
On coral reefs there are "cleaning stations" manned by fish whose role in nature is to eat parasites off bigger fish. The bigger fish could but don't eat the cleaners. Co-evolved communication for sure, but also a wider "neutral zone" around the cleaning station. Something similar around desert water-holes.
Dolphins not infrequently rescue drowning humans to dry land. This is also evidence of high intelligence and inter-species altruism. They know that humans come from dry land, a place where no dolphin ever wants to end up.
When human scientists went to study wolves in the far north of Canada (so far north no humans lived there, not even Inuit) they were surprised that the wolves were neither hostile nor fearful, but highly curious. Inter-species communication was so effective that in one season, the humans had been integrated into the wolf pack, and were trusted to look after the cubs while the pack went hunting! I expect this is how dogs first became "man's best friend".
And so on ....
(Bacteria talk to each other in RNA code packages. Does that count? :-)
I'd postulate the counter-argument that any civilisation stupid enough to develop Von Neumann robots is wiped out by the consequences, long before it is able to develop them to the point where they can propagate over interstellar distances. You've seen the movies ("Terminator" etc.) The reality would probably be worse.
Ken Macleod (SF) came up with a neat answer t the Fermi paradox. The solar system, and just about every other solar system, is already teeming with alien intelligences ("deities"). They don't like hot, they don't like oxidizing atmospheres. They live very long and very slow in the Oort clouds. They talk to each other, very slowly and quietly, across interstellar space. Another thing they don't like is radio noise. If one of those hot oxidizing planets starts emitting noise, they drop a comet on it. This has already happened several times in geological history.
Actually the last two sentences are mine. It's a bad set-up for writing SF, so Ken gives them alien ethics, and a way to sort-of get around the speed-of-light limit . The book is "Cosmonaut Keep", featuring intelligent dinosaurs and intelligent giant squid, amongst others.
I've just realized a calculation of such odds is a good illustration of the fact that we inhabit a 4D spacetime, not just a 3D space. Adding another dimension (time) is a very good way of reducing the odds of a near-coincidence of all coordinates. It may also be the reason that almost all civilisations think about it, realize how infinitessimally unlikely it is that they'll find other intelligent life, and give up. Especially if they try hard enough to get some observational certainty as to the probability of a star having a planet bearing life. If the mean distance between life-bearing planets in this galaxy is 4000 light-years, it's a near-certainty that the universe is teeming with other intelligences. It's also a near-certainty that no two civilisations will ever get to talk to each other, provided the speed-of-light limit holds.
More than "likely"
It's a thermodynamic-odds certainty that we can't find any dead alien probes that are drifting in the solar system. We can just about notice rocks tens of meters across ... if they happen to come close enough to Earth's orbit to attract out attention.
An alien interstellar probe is unlikely to be bigger than a coke can. (I'm assuming a modicum of technological advancement, but no way to beat the speed of light.) It might once have had a solar sail, but if it hadn't evaporated within a few years of its deployment, those aliens were wasting an awful lot of energy sending it here!
I haven't had enough coffee to think about the physical limits on detectability, but there are such. Radar is probably the best bet. How far away does a coke can have toi be, before its radar reflection becomes undetectable in the universal background noise? I'd hazard a guess, inside the orbit of Mars. Tht leaves a LOT of solar system outside. Anyway, the solar system is chock-full of coke-can-sized rocks to detect. How to spot the needle in the haystack?
BTW alien probes that landed on Earth will long ago have been destroyed by geological or biological activity. Or if neither of those, just undetectably buried in kilometers of solid rock. (I'll ignore the extra energy budget needed to send something that could land on a planet across interstellar distances!) Best hope for aliens that really, really wanted to leave a probe where the developing life on earth would find it many millions of years in their future, would be the moon. But even there, it would get covered in dust.
An active probe, sent across interstellar space to carry on bleeping at probably non-existent locals rather than devoting all its energy budget to communicating back with its makers? Seems unlikely. That it can stay "live" for as small a span of time as a million years? Improbable squared, and we've probably still missed it by more millions of years.
About the only thing we can rule out is hegemonising swarms of Von Neumann robots. The Fermi paradox definitely does apply to them; they aren't here so they don't exist in this galaxy.
The electrician didn't chuck in the day job and change his name to Banksy, by any chance?
Meta-viruses, not viruses.
OK, perhaps I should have made that clearer. Yes, of course I know that life has co-evolved with natural viruses, and that these days it's possible to genetically modify viruses. Some day a bad one will escape, by malign intent or accident, and the world will be faced with a new plague. ONE new plague. Nature does the same thing all by herself on occasion.
What is being proposed seems to be an analogue of inserting an interpreter for something else into life's operating system. And you know the trouble we have in IT with programmable entities that are supposed to be confined to a sandbox, escaping from it when the bad guys provide the input?
The thought of someone genetically engineering one killer plague is scary enough. The thought of someone engineering a generator that iterates through all the permutations of nastinesses is far scarier. Which is precisely what might happen, if this idea is put into practice and if it then escapes from the labs. Would we long survive as a species, if someone created a mutation-and-permutation generator for (say) new flu virus variants?
Why the fuss?
It amazes me how many people don't get it.
It's not because Gnome 3 exists. If they'd taken the Gnome 2 code-base, forked it, announced that Gnome 2 needed a new maintainer because they wouldn't be working on it any longer, then the world would be a happy place. The maintainer would have been found. Some of us would stick to Gnome 2. Others would enthusiastically embrace Gnome 3. System managers could install both on multi-user systems, and let their users decide when they log in.
This is the strength of open-source. Diversity, choice, and no-one forcing us to use new software that we don't like.
The Gnome people broke the "rules". They developed Gnome 3 in such a way that you couldn't choose. They pretended it was an upgrade, when it was a brand new interface with little if anything in common with its predecessor. They did to us for the sake of ego-building, what Microsoft did to us for the sake of making more money for Microsoft. And frankly, compared to the gulf between Gnome 2 UI and Gnome 3 UI, the jump from XP to Win 7 is across a mere crevice.
That's why the fuss. If the open-soure community had any laws, Gnome would have broken a lot of them. Thank heaven that XFCE exists. It feels like a step backwards, but only a step. And now it's attracted a whole lot of new users including Linus, it might even get improved.
This sounds horribly dangerous.
I don't know enough about biology to comment on any proposed safeguards, but I do know about computing and human nature. If it can run code, it can run viruses. If it's there to be cracked, someone will try to crack it. And if it's a microbe, there's a chance it'll get out of the lab and start thriving in the wild. And then start expressing those viruses and their o/s in the wild, and mixing and merging with other wild microbes.
What after that, I'll leave to the SF writers. Greg Egan wrote a short story with a set-up something like this - a self-mutating plague that escaped. Fortunately, it mutated randomly and away from its original deadly form, so it only wiped out the population of Arizona, created a global panic, and left in its wake uncountable millions of new diseases.
Most of my CDs of that age are still perfectly fine.
I did have one that started decaying. I found that I was able to copy it to CD-R without error using my computer, even though my audio player was unhappy with it. Methinks that modern computer everything-drives have much better optics and error-recovery, than audio systems. The CD-R plays fine on the audio system (although with older audio CD drives, that isn't guaranteed).
Doesn't anyone listen to classical music, or anything else using non-electronic instruments? A solo flute is close to a pure sine wave plus some white noise ... the non-harmonic artifacts that arise from compression are truly ghastly.
If they don't let us download uncompressed or losslessly compressed then I have zero interest. in downloading. Same reason that if they turn off FM transmissions I won't listen to music on (DAB) Radio. (Classic FM take note! )
I don't imagine for a moment that a study of methane hydrate under permafrost would give such a happy conclusion. We already know most of the answer.
Runaway global warming caused by methane released from melting methane hydrates has happened many times in the recent geological past. The evidence is there in the geological record. We're about to trigger it again, for some geological value of "about".
That's important. A run-away over 2000 years is an eye-blink in geological terms, but is slow enough that the human race will adapt to the changes and only historians will notice them. But if it proves to be 200 or 20? The human race is gambling for high stakes.
(Before anyone shouts ... not *that* high. Life will survive as it did the times before. Humans will probably survive. It would just make a big mess of our civilisation. Today's flooded Bangkok, writ permanent and world-wide with saltwater).
Old Science-simulation programmer's lore
If your system has an easily-computed invariant (such as total energy) then compute it after each iteration and compare it with the previous value. You expect small changes because of floating-point rounding errors. An unexpectedly large change meant one of two things:
You'd introduced a bug into the program, or
The floating-point hardware was flaking out.
In the days of the CDC 7600 and Cray-1, it was commonplace for a scientist to phone the computer centre and tell them the latter. They'd assert the former. It was most gratifying to say "told you so" on the occasions that the system went down for hardware maintenance a few hours or days later.
1. the need to check your invariants hasn't gone away. It's just that with modern technology you usually own the whole CPU, and it isn't mend-able any more.
2. Who says that the observable universe isn't just a sim in $deity's computer? And that it doesn't have any bugs?
Observable universe < universe.
The universe may be very much bigger than the observable universe. The latter is the part of spacetime from which light is today reaching us. It's pretty much a 3D section of a 4D spacetime (since we have observations over a few milennia only, and good ones only over a decade or two). The rest might be (a) forever unobservable, (b) inferrable from its effects on the observable at an earlier time, or (c) capable of becoming observable in the deep future if anything can brake the observed (and probably accelerating) expansion of the observable bit of the universe.
I've long had a pet idea that the topology of the entire universe is torroidal . That's based on the non-observation of magnetic monopoles and the everyday observation of magnetic fields. The simplest topology within which those facts do not essentially contradict each other is the torroidal. Such a topology might also eliminate the need for the birth of the entire cosmos from a big-bang singularity. What's physics like in the "hole" in the donut, if the observable universe is on the "outside equator"? Different, for certain. BTW that's a 4-torus, or possibly one of higher dimensionality if string or brane theories are correct. It would imply time and space are both eternal but cyclical.
A torroidal universe would have to be anisotropic (different in different directions) but the observable bit of it might look very close to isotropic. Very close, but maybe not so close that we can't find a slight hint of a built-in directionality?
Who buys on the high street?
Surely it's a no-brainer to buy electronic gadgetry, and even white goods, from an internet retailer wo can offer better prices because he has no expensive high street or retail park premises to rent?
As far as I am concerned PC World is for distress purchases ... something I need immediately not a few days hence. Also PC World is quite smart - you can go online, reserve it at PCW internet prices, and collect it from the store. PCW internet prices aren't always very competitive, but they are often a lot less than the standard in-store price!
The internet is killing the high street, at least with respect to things that one doesn't want to touch or try on before purchasing, and doesn't need in a hurry.
Of course it's just one consequence of a major natural disaster. So? This is a technology site not a general newspaper. Give us credit for being able to supply our own context without needing more than a one-sentence reminder.
We might do better to express a strong desire that Seagate and WDC don't rebuild elsewhere and leave their Thai employees out of a job after the floods recede.
Visible / near IR band is universal
See my post below.
There's a transparency window between where UV starts getting absorbed by ionising molecules, and where IR starts being strongly absorbed by exciting vibrations of molecules. The next transparency window is right down at high microwave-radio frequencies and there's no strong natural source of those for passive vision.
Our visible is tuned to the emissions from our star. I'd expect for life around a redder star, their visible would exclude everything above yellow, and include some or all near IR (which we use for line-of-sight IR network links and suchlike).
Visit the British Moseum, and you'll see some very beautiful Roman jewellery made by weaving fine gold wires into a flexible cloth. Similarly from other civilisations. Another technique is tiny interlocked rings, like mouse-scale chain-mail.
These died out because of goldsmiths' hallmarking regulations. Every separable piece had to be individually hallmarked. Obviously, that wasn't possible with suchg thin or tuny pieces.
Some modern jeweller ought to bring it back. These days the hallmarking could be done, usinga microscope and a laser!
Physics is universal, evolution is probably likewise
I'd expect a large fraction of alien animal life to "see" using light (visible or near IR), assuming they originate on rocky planets with somewhat transparent atmospheres.
Reason: solar radiation exists. As life evolves predators and prey, it becomes an evolutionary advantage for both to be able to see by reflection of such light. An evolutionary arms race leads to eyes as good as physics and biochemistry can allow. Physics dictates that other wavelengths are less suitable for sight by passive sensing. There's a window of transparency there in any sensible atmosphere. Ditto in any sensible liquid ocean. There's also a good fraction of the solar output in the visible/near-IR band.
Here on earth eyes have evolved several times independantly. Fish and higher life evolved from them, including humans. Molluscs. Spiders. Insects (compound eyes, not at all like spiders). Trilobites. (or are insects descended therefrom)?
Yes, there may be exceptions. An intelligent species descended from bats might have eyes atrophied to the point of uselessness, and see by sonar. From cetaceans, ditto, though it's hard to see how a marine-adapted species could acquire technology first-hand.
Sight is an essential for survival, and the most obvious route is evolution of sensory cells sensitive to solar illumination, leading to eyes as we know them.
I'm not convinced that a technological civilisation which once felt the need for night-time illumination would ever give that up as long as it remained a technological civilisation. That, even if current human solar and LED technologies are as good as it gets.
Efficiency, yes, but that means using less electricity to generate much the same amount of light. Efficient street-lighting may illuminate only downwards, but a significant fraction of that light must reflect back into space. Eyes would be useless if it didn't.
We can calculate the available solar energy resource. It's greater than the entire current energy usage by humanity (even before you consider the possibility of powersats). Night-time illumination at human levels using efficient light sources does not need an impossibly large fraction of a solar-powered civilisation's energy harvest.
So looking for artificial light is a good idea, if it's detectable over many light-years. Any detection of any technology being deployed outside the solar system would perhaps be the single most significant observation of all time.
BTW, powersats. Would the microwave leakage be detectable lightyears away?
0.997 - It's a fraction of perfection
Obviously, 1.0 is impossible. Or it means that they've done a complete rewrite to put in loads of new features nobody wanted, and the perfection quotient goes back to zero while someone else forks 0.9971
Yup. What's the bet that to de-activate you'll have to provide a serial number that you forgot to write down prior to throwing the dead box away, or which became forever inaccessible the moment it failed?
Surely the AC modulation of the alien-made light emission would also help, if it existed. Physics is universal, so it's a reasonable expectation that the technological evolution of an AC power grid would be like ours, locking them into a frequency not very far removed from the 50-60Hz here. (It would also help that they are very unlikely to use exactly 50Hz or 60Hz, so manmade interference can be filtered out).
Only snag is that AC lighting may be a short-lived phenomenon. Anyone know if LED lights contain an AC-DC converter, or do they economize by not bothering with a capacitor?
And where's the light-bulb icon?
PCL / hplip not bad.
It's not quite that bad. If it's made by HP and supported by (open-source) hplip, I've found that the only remaining question is how easily I can get a sufficiently up-to-date version of hplip integrated with my linux system. And that's not always necessary either.
PCL is somewhat backwards-compatible. Enough so that in a hurry with a distribution that didn't have hplip up-to-date, printing to an Officejet Pro 8000 (current) worked with the driver for an Officejet Pro K550 (2? 3? models back in time).
Second-best to Postscript I agree, but if you want a colour-capable printer for under £100 and reasonably low running costs, then HP's Officejet Pro range is well worth a look. Beats many cheap lasers on running cost, and colour almost for free (if you don't go overboard printing photos and blocky graphics).
The key points are that PCL is open-spec and hplip is open-source.
Cram more in?
Physics says they won't / can't.
As soon as the fuselage gets big enough to have two decks you *have* to give each passenger about twice the area compared to a single-decker plane. You can't pack in twice as many passengers, because the plane couldn't then get off the ground. And even with a single deck there will be more space per passenger in a big jet than a small one. I don't think that wing lift scales up as length squared, whereas passenger deck area does (for the same shaped fuselage).
Of course a lighter plane does mean they'll be able to cram more passengers into it. Sigh.
Not so simple
Point-to-point means more routes means less flights per route means less choice as to when you travel. You may prefer (say) a 10am departure and 8 hours travelling time, to a 7am departure and 6 hours travelling time.
I've paid significant amounts extra in the past, to travel at a time convenient to myself. More generally there are certain flights that are always cheap as chips yet which get few takers. These are at horrible times of the day, which the airlines have to fly in order to position an airliner where the next flight is going to be full.
Well, horrible for most people. I've saved money and had most of an extra day abroad by choosing to be back at Luton at 11.45pm. No bad traffic on the M25 at that time, so easier on the last leg!
Long-term ageing problems
We didn't know about the long-term metal-fatigue problems with aluminium alloys, until Comet airliners started exploding in mid-flight. We didn't know about the effects of sea salt on them, until the open-top Hawaiian airliner incident. Rolls-Royce once almost got sunk as a company by committing to using carbon-fibre turbine blades in the ill-fated RB211, only to discover the fatigue problems once the things were almost in production. And so on. I can think of a number of similar examples in the history of IT. Even using well-understood materials, Alfa Romeo still managed to have a huge problem with a car whose engine fell out after only a few years.
Manufacturers do of course do accelerated ageing tests, but they have to leave out the passage of real time.
It's anyone's guess whether composites or Scandium alloy will be best for aircraft fuselages. And I'll be a little bit more nervous flying in either, since it'll take more than the rest of my years for these materials to be truly well-understood by the industry.
They did make a nuclear shell, called the "Davy Crockett". I think that even during the cold war, they soon realized that a weapon that included the people firing it in its deadly blast radius was rather pointless.
The trouble with getting rid of all nukes, is that it leaves the field clear for whatever evil megalomaniac decides to hold the world to ransom by building a few of his own in secret. Human nature being what it is, our best hope is for some number of hopefully stable countries to have fairly small nuclear arsenals, and for everyone to know that any power which is first to use one, will be on the receiving end of all the rest. We're not there yet, but closer than we were 50 years ago.
For the appropriate value of "safe"
Yes, once the electrically-operated detonators are disconnected, any nuclear bomb is probably incapable of generating a nuclear chain reaction, and certainly incapable of generating any yield in excess of a few kilotonnes.
If you are the person dismantling it, or even the person a few kilometers downwind, this consideration is quite academic. There's a lot of painstaking work to go before you can think of it as safe. A nuclear bomb it may be no longer, a bomb it most certainly remains.
I suspect that a bunker more than 2km underground in hard rock not prone to collapses cannot be busted. The USA put SAC HQ under a mountain for this reason. Iran has done the same with its A-bomb factories.
If you're going to try, you need to make sure that the explosive shockwave is well-coupled to the rock rather than the atmosphere. That's why a third of a megatonne detonated a few meters underground is an adequate (probably much "improved") replacement for this multimegatonne monster.
As for environment turning into "a fair replica of Venus" you've been reading too many Z-grade apocalypse-SF stories. Nature has several times inflicted far worse on the planet than our bombs ever could. The Chixulub impact was about 100 Teratonnes, several orders of magnitude more than the combined arsenals of the USSR and USA at the height of the cold war. It might have wiped out the dinosaurs - if not it helped them depart the stage. It would certainly have put the human race back to the stone age if we'd been around at the time. In the big picture, it was a fairly routine mass extinction event - nothing like as bad as the big one 200Myears earlier.
Not just in Oz
I once watched an outsized load - some sort of big pressure vessel - being manouvered through the middle of Saint Albans, here in the home counties of Blighty. It probably wasn't outsized at all by Oz standards, but this is an ancient town centre with listed and very valuable real estate on both sides. Moving awkward buildings out of the way, intentionally or otherwise, was definitely not an option.
It was preceeded by a flotilla of police vehicles, checking the route, and a parking enforcement vehicle, that lifted anything parked where it shouldn't have been out of the way . The big load crawled along at about walking pace, leaving a trail of devastation. OK, I exaggerate. Its driver really was very skilful, even inch-perfect. It left a couple of depressed kerbstones on one corner, and the gas and water companies were rather busy digging up that road in the weeks thereafter.
By the way, somewhere in the Midlands is a graveyard of huge steel cylinders. They're failed precision castings for paper-making machines. If the surface has any imperfection, the cast is a reject, but it's so friggin' huge and strong that no-one knows how to chop them up for recycling. So they are towed off as outsized loads, just as if they were perfect, and unloaded ino a field to rust in peace.
Yes, this is definitely the point which matters. These (unelected, unaccountable) banks and credit companies are demonstrating that they can shut down any e-organisation that needs paying customers or supporters any time they choose. Soon, as physical cash fades into history, that'll be any organisation, e- or otherwise.,Today, wikileaks. Tomorrow ... who knows? They already have enough power to take out a small country.
A reminder from the past
First they came for the Communists / And I did not speak out /Because I was not a Communist
Then they came for the Socialists/ And I did not speak out /Because I was not a Socialist
Then they came for the trade unionists/ And I did not speak out /Because I was not a trade unionist
Then they came for the Jews/ And I did not speak out/ Because I was not a Jew
Then they came for me/ And there was no one left/ To speak out for me
In so many ways it's hard to know where to start.
Lithium Batteries contain (surprise) Lithium, not rare earths.
Batteries are electrochemical devices. The electrochemistry of all the Rare Earth elements is almost identical, which is why it took so long for chemists to work out how many such elements there are. NiMH batteries can utilize any rare earth or mixture thereof.
The physical-electronic properties of the rare earths are as spectacularly diverse as their chemistries are similar. For a good red phosphor, only Europium will do. For a magnet, you need Neodymium (or Samarium for a weaker but higher-temperature magnet). Other rare earth elements have other specialist uses.
Imitation is the sincerest form of flattery!
... which is why software patents should be scrapped, and why Jobs should have been happy rather than angry.
The right form of protection for software is copyright, as for books. Copying chunks of code is illegal. Creating new code that offers the end-user a similar look and feel, or which solves the same problem using the same mathematics, should be completely legal. Imitation ....
I'm sure that a combination of an accelerometer and a heat or ionization detector could trigger self-destruct of a satellite only after its re-entry. Making that fail-safe (in other words, reducing the odds of a premature explosion to the infinitessimal) is surely no harder than making a reliable satellite and its (potentially explosive) launcher in the first place. Although you would need some interesting studies to make sure that multi-year exposure to space and space radiation didn't degrade the explosive into instability (studies that I'd wager have already been done by the military, and the results filed "secret" or above).
The problem is economics. This plan would make the satellite quite a few kilogrammes heavier (the weight of a sufficient quantity of explosive). This would increase the launch cost very considerably and/or reduce the available mass for the useful payload. Better to rely on favorable statistics, and pay out compensation to any very unlucky person (or their heirs), at least until death by falling satellite happens for the first time.
Using their monopoly while they can
They know that the high price of REEs has stimulated the development of many new mines outside China. Most are still a couple of years from production. At that time the Chinese monopoly will be broken, and the price of REEs will plummet like that of many other commodities that were in short supply in the past.
The Chinese may then switch to trying to flood the market to depress prices and put the newcomers out of business. Methinks that would fail (and if they are smart they won't try). My reasoning is that there must be a lot of latent demand for REEs at a lower price. An IT example is the lead-acid battery in UPSs, with an annoying 3-ish year life. If NiMH batteries didn't cost 3x more, they'd use NiMH instead for an effective-infinite UPS battery life (and lighter weight as well).
Rare earths - despite the name - aren't particularly rare. It's just that until the 21st century, there wasn't much demand for them, and the Chinese built themselves a position as near-monopoly supplier (based on foresight, cheap labour, and a willingness to inflict severe environmental damage in the vicinity of their mines which wouldn't be allowed anywhere in the West).
- Xmas Round-up Ten top tech toys to interface with a techie’s Christmas stocking
- Google embiggens its fat vid pipe Chromecast with TEN new supported apps
- Microsoft: Don't listen to 4chan ... especially the bit about bricking Xbox Ones
- Shivering boffins nail Earth's coldest spot
- Exploits no more! Firefox 26 blocks all Java plugins by default