If it were (as many people assume) just a case of people sitting down and writing articles which remain in that state for all time - then, yes, I'd guess closer to 5% too. But that's now how Wikipedia works. Pick an article at random - hit the history button - see how many people have worked on it? For plagiariasm to stand, it requires that none of the subsequent editors noticed it. That's much less likely - but still possible - but in addition to that, the general churning up of text tends to change sentences and paragraphs around until they bear no resemblance to the form they came in as...this would 'unplagiarize' text fairly quickly in most cases.
So - yeah - I'm a little surprised it's as high as 1% - and probably it's not.
Some are also instances of people writing something on their own web site and then later deciding to put it on Wikipedia - so even the instances where the other web site predates the Wiki article may not be copyright violations. Without discussing the matter with every single original author, it's hard to know.
I guess the only thing this study tells us is that an UPPER limit on the number of plagiarisms is of the order of 1%. That's still an alarmingly high number.
What?!? Why is this news? We have libraries - those are the big chunks of useful code - packaged and (hopefully) documented for re-use and widely distributed.
When you design new applications you look for libraries that do the bulk of the work - and the application becomes mostly 'glue' to hold the libraries together. Scripting languages are the epitome of this - where large piles of carefully written and optimised libraries make up for the overhead of interpreting the actual library code.
Dunno why anyone finds this surprising - it's what we've been doing almost since the dawn of programming.
The tricky part is noticing when you can't find a suitable library - so rather than dumping a bunch of code into the application, design it as a reusable library. There is a small overhead to doing this - but rarely more than (say) 5% of the development cost. So when writing OpenSourced code this is generally done well. The hard part is in commercial code where you have someone with a chequebook and a severe lack of foresight breathing down your neck who would rather save 5% now than save 30% on the next project.
No,no,no - wikipedia doesn't "own" everything they do for the year - they have to agree to OpenSource everything they do for the year.
And what if one or two do goof off - we'd be picking the 1000 most respected, smartest, people and making a big deal about the honor and aspects of perpetuity "The deepest thoughts of the thousand smartest people" - would such people really be stupid enough to goof off given that degree of public significance?
If only 900 of them produced anything - if just 1% of them produced something amazing...wouldn't it still be a wonderful way to spend a hundred mil?
I guess the 3,000 deaths due to terrorism are somewhere in the 105,000 'unintentional injury' group or else they don't figure in the top ten things we should be worrying about at all. 43,000 of the unintentional injury group were car accidents incidentally.
Terrible though that attack was - it was not statistically significant as a thing for rational beings to worry about. Losing everything we've lost in an undoubtedly vain attempt to 'fix' this problem is totally out of all proportion to the injury we suffered. Do you seriously imagine that for the billions we spent (and the billions more we'll undoubtedly spend) on the pointless invasion of Iraq, we couldn't have gotten a half percent reduction in the rate of heart disease? If the terrorists could somehow pull off a twin-towers incident every two days throughout the entire year, year in and year out - it would still rate well below heart disease and cancer as a cause of death in the USA.
Yeah - I said "statistically insignificant" and I meant it.
Paying for things that already exist (even if copyrighted) is a waste. Books full of science can be read, summarised and written about with the existing rights we all have to that material. Paying to release the actual documents is unnecessary.
Let's pay for something new.
I'm betting most academics don't earn much over $100,000 a year. Take the $100M and pay the thousand smartest people on the planet to each spend an entire year writing about everything and anything they feel is important for the future of humanity - with the stipulation that every word they write in that year goes immediately into the public domain.
Think of the qualitative improvement in Wikipedia if we added tens of thousands of new articles by the smartest people in their fields.
Hold on - the linked article says that this scheme is proven to work because in the Yeoville area alcohol related violence had dropped 48% over the trial period. It then went on to say that over that eight month period there were only TWO major incidents. So if there had been (say) four major incidents over the preceeding eight months - which reduced to two during the trial - that would have been a 50% reduction.
(Note that one of those two major incidents wasn't even anything to do with pubs - some kids were at an under-18's disco and obtained alcohol "somewhere else" - it shouldn't even have been counted).
I have two observations:
Firstly: I would submit that whether there were two or four major incidents over a period of eight months is not a statistically valid sample. Especially because the preceeding 8 months would have included Xmas and New Year - both notable occasions for serious drunkenness. No competent statistician or conductor of scientific tests would sign up to these conclusions from such a ridiculously small sample - so we should either conclude that they are invalid - or that they were actually counting something else...which leads me to:
Secondly: For a number like '48%' to have come about, we cannot be measuring a reduction from four to two major crimes - that would be a 50% reduction. This MUST have been taken over a vastly larger sample of incidents. We must conclude then that they are not talking about 'major' incidents such as the two described (a sexual attack in the toilets and a fight between two kids that erupted into a major street brawl). So what this fingerprinting exercise is all about is reducing MINOR incidents.
So let's call this what it is. It's not about cutting down on serious offences - it's about reducing MINOR offences by banning people from pubs who happen to have lost their tempers or done any of the usual things that drunk people tend to do.
Is that worth the loss of privacy that this entails?
I was watching the movie "V for Vendetta" again last night - as a graphic novel that was written in 1982 it's eerily predictive. For a movie made two years ago, it's practically a documentary.
I'm a Brit who has been living in the USA for the past 13 years and it's hard to say which is more like the movie. Britain with more spycams per person than anywhere else on earth - and soon you can't even have a beer without being fingerprinted! Or perhaps it is the USA in which the faceless secret police can monitor what books you check out from the library, bug your phone without judicial oversight and swoop down on you, merely accuse you of being a terrorist (no proof required) and on that pretext lock you up, torture you, ship you off to god-knows what hell-hole - and all without any right of trial or appeal?
Hmmm - hard call. Between the two countries - it's difficult to say which comes closest to the nightmare that V opposes in the movie. As he says: If you want to know whose fault this is - just look in the mirror.
Our own fear of statistically insignificant terrorist violence (or avian flu or WMD or drunk drivers or...you name it) induces progressively higher tolerance for the State to ratchet down the human rights of the entire population. There will come a point when we realise that this has been a terrible mistake - but will we do that before or after the point where we can no longer reverse it's effects?
Better get that bulk order for Guy Fawkes masks in before the rush. Amazon have them for $5.99.
> some form of genetic isolation (which may or may not mean geographical isolation,
I think the author (who is an economist by the way) was presuming that rich, tall, glamorous people meet each other in their exclusive up market settings and have rich, tall, glamorous kids. This provides the necessary isolation for genetics to do their thing - and in some ridiculously short period of time, it's a case of Morloks and Eloi.
Sadly, this guy knows nothing of value about genetics - and not too much about society either.
Remember that he also believes that these new humans will have coffee-coloured skin due to racial mixing. It's not clear to me why he believes that racial self-segregation will vanish while other forms of segregation are on the increase.
TFA is all just ridiculous speculation that no true evolutionary biologist would buy into.
The problem with predicting human evolution is that we are rather more in control of it than is the case for other animals. Genetic failures can be fixed up by medicine such that genes that would have been fatal (and thus eliminated from the pool) are now viable if the symptoms they produce are treatable.
Consider this: if there were a strong reproductive bias for taller men - then genetically shorter men would come to be considered 'abnormal' by our society. This would cause them to be treated with growth hormone at a tender age - hiding their genetic predisposition and thus allowing the gene for 'shortness' be available as procreators of the next generation. This would make genetically induced tallness a hard trait to take hold.
If you doubt this could ever happen - please note that it has already happened:
"In July 2003, the Food and Drug Administration (FDA) authorized pharmaceutical companies to promote human Growth Hormone (hGH) for use in children who are very short but not suffering from any specific illness or medical condition. Parents are now using hGH in record numbers, hoping that hormone treatment will give their kids happier childhoods and more prosperous adulthoods."
QED.
What about the deep, icy crater theory?
on
No Ice on the Moon
·
· Score: 3, Insightful
Wasn't there also talk about ice in deep craters situated where the sun can never shine on it?
The theory was (and I hope I have this right) that cometary ice must impact the moon from time to time - so there is water there from time to time - but whenever the sun shines, in the absence of an atmosphere, the water will evaporate (sublimate?) away quite quickly during the lunar day - then freeze out of the atmosphere during the night.
This mechanism would generally keep whatever water molecules there is up there moving around...*UNTIL* (by chance) it lands somewhere where there is never any sunlight - inside a cave or a deep crater. At that point it must settle - and there is no longer a mechanism to move it around again. With no atmosphere to scatter sunlight, permenantly dark places will be profoundly cold.
It follows then that whatever water there is will always end up in these relatively rare places EVENTUALLY - so given enough time, all of the moon's water would end up stashed away in just a few easy-to-predict places.
Furthermore, we'd never be able to see those places from earth-bound or low orbit telescopes because any place we can see must also collect sunlight at some point in the lunar orbit....at least that's what I recall. It sounds kinda plausible.
The US Nimitz class Aircraft carriers each carry a single 194MWatt nuclear reactor. The USS Enterprise has a total of eight nuclear reactors onboard. All but two US carriers - and absolutely all US military submarines are nuclear powered. Even ships as small as cruisers have been nuclear powered in the past.
French Rubis class submarines each have a 48 MW reactor.
Russian Typhoon class submarines carry two 190MWatt reactors.
Russian Arktika class ice breakers carry two reactors of 171MWatts each. The Taimyr class have 135MWatt reactors. There are a total of ten Russian civilian nuclear ice breakers in active service.
So a couple of measley 60MWatt reactors on a barge somewhere isn't really the huge news you might think.
Having reactors on a barge that's moored someplace has gotta be safer than having them than crashing off through icebergs or sitting off the coast of countries full of terrorists who would love nothing better than to drive a boatload of explosives into them - or yet putting them on submarines whose safety record doesn't really look so great.
It's been suggested that the two reactors they are using on this barge came from decomissioned Russian submarines anyway...so we're probably better off having them used for peaceful purposes and being moored someplace where we know exactly where they are!
Yep - and the so-called 'Clean Coal' approach concentrates naturally occurring radioactivity to the extent that the waste produced by even the most modern coal fired power plants has comparable amounts of radioactivity to nuclear plants.
Nuclear power has problems - but they are all solvable within our technological reach. The problems of irreplacable fossil fuels combined with the bad consequences of dumping CO2 into the atmosphere are not in any way solvable with technologies we currently have - or even expect to have. Windmills, wave power, solar power , biofuels and others aren't likely to produce the quantity of power we expect to need over the coming years. Fusion looks cool - but we can't do it yet.
So whilst nuclear power is *HARD* - it has the huge benefit of not being *IMPOSSIBLE* like all of the other power sources we have.
A modern GPU might have as many as 48 "fragment shader" processors inside it - so there are really around 21,500 processors versus 282,000 CPU's. Then each GPU processor works in full four-way arithmetic parallelism - so it can do arithmetic and data move operations on four numbers just as fast as one. So with the right mapping of algorithm to processing, you have 86,000 floating point arithmetic units...they only need to be about 3x faster than the CPU's.
But these are not general purpose processors - in some respects they are horribly, horribly limited.
So if you have an algorithm that is enough like rendering polygons with textures - the GPU is just insanely fast...if your algorithm isn't enough like that - then you may not be able to run it on the GPU at all.
Look - FireFox is OpenSourced - right? So for chrissakes let them do what they want with it - that is THE ENTIRE POINT!!! If the Debian guys (who are not exactly complete Klutzes at this stuff) mess up, you say "Hey the Debian guys screwed up - come download the real one from the usual places."
In fact, the cost in environmental impact per load carried to orbit can be arbitarily small - if the climber goes slowly enough, you could power it with solar power. The problem is that the capital cost of the system is non-trivial so you have to get a lot of usage out of the thing in order to pay your investors. So it might not be acceptable for the climber to ascend at a snails pace. Faster climbers mean more intense fuel sources - so we end up with HUGE ground-based lasers pumping light energy into solar panels.
The first task of a space elevator is probably to make a second one.
It does seem close to impossible - but the benefits if it IS possible are huge.
The big thing about getting a rocket into space is that almost all of the launch weight is the fuel - almost all of which is consumed in propelling fuel.
The elevator cab would carry no fuel at all. Because it can be slow, it doesn't need much power - so it can be powered with solar panels illuminated from below using powerful lasers. The fuel for the lasers doesn't have to be propelled into space - so there is VASTLY less energy required - and it doesn't have to be in a dense, portable form like rocket fuel. You could power the system with windmills or something.
Furthermore, the cab can trundle slowly down the cable when it wants to come home. No reentry heat - no expensive insulation tiles - no wings, no undercarriage. Again, this makes the cab smaller, lighter, cheaper - it really just needs to be an airtight box with some wheels to grip the cable, solar cells for power and some oxygen to breath (for manned "flights").
The cab should be 100% re-usable (maybe needs new tires once in a while!).
This makes space VERY cheap. Once you are at geosynchronous orbital height, you can launch satellites by pushing them out of the cargo bay with a spring or something...again, very, very low-tech.
If you want to launch something out of earth orbit then climb further up the cable beyond geosynchronouse orbital height and (once again), push the vehicle out of the window and it'll shoot out of earth orbit with no fuel needed.
So the capital costs of getting this thing up there might be a LOT higher than a shuttle launch - or perhaps more expensive than dozens of shuttle launches. But once you've built it, the cost to get a ton of stuff into orbit will be comparable to driving it across the USA in an 18 wheeler. Assuming there is a market for getting lots of stuff into orbit cheaply, this thing will pay for itself very quickly.
BUT - can it be built at all? Even in theory?
That's all down to materials science in the tether. Is there any material that's light enough and strong enough and reliable enough and cheap enough? If there is then this will happen for sure - and within not many years of a suitable tether material being discovered- if there isn't, it won't. The feasibility and economics of the thing are a no-brainer if you can solve the tether material problem.
The problem is weight. It takes a massively strong material (possibly impossibly strong) to support it's own weight over all of those miles. If you cut holes in it, it gets weaker and makes the problem much, much worse. If you stick extra bits onto it, it gets heavier - and that means it has to be stronger.
In the end, it's much simpler to build the leanest, lightest, stongest tether (then engineer the heck out of the elevator cab so it can climb it) rather than engineering the tether to make it easier to climb.
The project is right on the edge of being possible - the tensile strength to mass ratio is a couple of orders of magnitude more than that provided by any known materials *except* carbon nanotubes - which could *MAYBE* handle the forces if you could make unbroken, flawless strands that were long enough (I think the longest ones we can make right now are under a millimeter long - and we need a few thousand miles of the stuff).
100% of this problem is the tether - if a strong enough tether could be devised, the rest of the elevator would be more or less easy to do.
A modern graphics chip (GPU) has 48 fragment processors and a dozen more vertex processors - each of those does full scale floating point in four way parallelism at fairly respectable clock rates. So if all you need is a couple of hundred floating point units - you already have it.
The problem is the memory bandwidth into and out of this beast. The number of pins you can have is fundamentally limited - the bandwidth down each pin is fairly sharply limited. RAM chips are creeping up in speed V-E-R-Y S-L-O-W-L-Y compared to CPU speeds. So in the end, I think it's pretty much irrelevent how many processors you have on a chip because even with infinite compute power, if you can't get the data in and out, you're limited to RAM speeds.
The GPU's only get their speed and the ability to use all of that floating point horsepower 100% of the time (with no nasty Ahmdahl's law consequences) because they are working on Red/Green/Blue/Alpha or X/Y/Z/W in parallel - and millions of pixels are being computed using the exact same algorithm in perfect lockstep with few (if any) branches or loops. That means that a single instruction stream can feed all of those processors in parallel - and the data goes in and comes out in highly memory-friendly ways.
That's not much use for general computing though.
I glaze over when I hear about increasing numbers of processors - if the memory interfaces aren't designed very carefully PS-3 then very bad things can happen. Ask about how much RAM bandwidth this gizmo has.
My 1963 Austin Mini gets 55mpg - designed by one guy in just a few months. It'll go at 70mph and you can actually get four people into it.
Fast forward to my 2005 Mini Cooper - designed by a huge team of engineers with big computers. It weighs more than twice as much as the '64 version, gets less than half the MPG and whilst it'll go 140mph - it's a real squash to get more than two people in it.
Why is that?
* Safety. * Air conditioning. * Emissions standards.
The safety thing is arguably justified - and in many parts of the world, so is the A/C - but the idea that you burn twice as much fuel so that the concentration of pollutants goes down - but the absolute amount coming out of the tailpipe goes up - is entirely an artifact of broken laws. If the laws regulated the total amount of emissions rather than the percentage of CO, NOx, etc - then SUV's would be a thing of the past in pretty short order. But right now, the laws keep out a number of very low-polluting smaller cars BECAUSE they can't meet emissions standards.
Slashdot Mirror
If it were (as many people assume) just a case of people sitting down and writing articles which remain in that state for all time - then, yes, I'd guess closer to 5% too. But that's now how Wikipedia works. Pick an article at random - hit the history button - see how many people have worked on it? For plagiariasm to stand, it requires that none of the subsequent editors noticed it. That's much less likely - but still possible - but in addition to that, the general churning up of text tends to change sentences and paragraphs around until they bear no resemblance to the form they came in as...this would 'unplagiarize' text fairly quickly in most cases.
So - yeah - I'm a little surprised it's as high as 1% - and probably it's not.
Some are also instances of people writing something on their own web site and then later deciding to put it on Wikipedia - so even the instances where the other web site predates the Wiki article may not be copyright violations. Without discussing the matter with every single original author, it's hard to know.
I guess the only thing this study tells us is that an UPPER limit on the number of plagiarisms is of the order of 1%. That's still an alarmingly high number.
What?!? Why is this news? We have libraries - those are the big chunks of useful code - packaged and (hopefully) documented for re-use and widely distributed.
When you design new applications you look for libraries that do the bulk of the work - and the application becomes mostly 'glue' to hold the libraries together. Scripting languages are the epitome of this - where large piles of carefully written and optimised libraries make up for the overhead of interpreting the actual library code.
Dunno why anyone finds this surprising - it's what we've been doing almost since the dawn of programming.
The tricky part is noticing when you can't find a suitable library - so rather than dumping a bunch of code into the application, design it as a reusable library. There is a small overhead to doing this - but rarely more than (say) 5% of the development cost. So when writing OpenSourced code this is generally done well. The hard part is in commercial code where you have someone with a chequebook and a severe lack of foresight breathing down your neck who would rather save 5% now than save 30% on the next project.
Haiku in six words?
Multisyllabification!
Non-impossible.
Slashdot effect strikes. Bandwidth very limited.
1) Just six words in story.
2) ???
3) Profit!
Picture worth 1k words. Story: six.
Six notes? RIAA copyrighted everything already.
No,no,no - wikipedia doesn't "own" everything they do for the year - they have to agree to OpenSource everything they do for the year.
And what if one or two do goof off - we'd be picking the 1000 most respected, smartest, people and making a big deal about the honor and aspects of perpetuity "The deepest thoughts of the thousand smartest people" - would such people really be stupid enough to goof off given that degree of public significance?
If only 900 of them produced anything - if just 1% of them produced something amazing...wouldn't it still be a wonderful way to spend a hundred mil?
Once a precedent has been set - it's hard to overturn.
According to the CDC, the 10 leading causes of death in the United States (in 2002 as it happens) were:
1. 696,447 Heart disease
2. 557,197 Cancer
3. 162,555 Cerebrovascular disease
4. 124,777 Chronic low. respiratory disease
5. 105,796 Unintentional injury
6. 73,248 Diabetes mellitus
7. 65,418 Influenza & pneumonia
8. 58,866 Alzheimer's disease
9. 40,801 Nephritis
10. 33,569 Septicemia
I guess the 3,000 deaths due to terrorism are somewhere in the 105,000 'unintentional injury' group or else they don't figure in the top ten things we should be worrying about at all. 43,000 of the unintentional injury group were car accidents incidentally.
Terrible though that attack was - it was not statistically significant as a thing for rational beings to worry about. Losing everything we've lost in an undoubtedly vain attempt to 'fix' this problem is totally out of all proportion to the injury we suffered. Do you seriously imagine that for the billions we spent (and the billions more we'll undoubtedly spend) on the pointless invasion of Iraq, we couldn't have gotten a half percent reduction in the rate of heart disease? If the terrorists could somehow pull off a twin-towers incident every two days throughout the entire year, year in and year out - it would still rate well below heart disease and cancer as a cause of death in the USA.
Yeah - I said "statistically insignificant" and I meant it.
Paying for things that already exist (even if copyrighted) is a waste. Books full of science can be read, summarised and written about with the existing rights we all have to that material. Paying to release the actual documents is unnecessary.
Let's pay for something new.
I'm betting most academics don't earn much over $100,000 a year. Take the $100M and pay the thousand smartest people on the planet to each spend an entire year writing about everything and anything they feel is important for the future of humanity - with the stipulation that every word they write in that year goes immediately into the public domain.
Think of the qualitative improvement in Wikipedia if we added tens of thousands of new articles by the smartest people in their fields.
Hold on - the linked article says that this scheme is proven to work because in the Yeoville area alcohol related violence had dropped 48% over the trial period. It then went on to say that over that eight month period there were only TWO major incidents. So if there had been (say) four major incidents over the preceeding eight months - which reduced to two during the trial - that would have been a 50% reduction.
(Note that one of those two major incidents wasn't even anything to do with pubs - some kids were at an under-18's disco and obtained alcohol "somewhere else" - it shouldn't even have been counted).
I have two observations:
Firstly: I would submit that whether there were two or four major incidents over a period of eight months is not a statistically valid sample. Especially because the preceeding 8 months would have included Xmas and New Year - both notable occasions for serious drunkenness. No competent statistician or conductor of scientific tests would sign up to these conclusions from such a ridiculously small sample - so we should either conclude that they are invalid - or that they were actually counting something else...which leads me to:
Secondly: For a number like '48%' to have come about, we cannot be measuring a reduction from four to two major crimes - that would be a 50% reduction. This MUST have been taken over a vastly larger sample of incidents. We must conclude then that they are not talking about 'major' incidents such as the two described (a sexual attack in the toilets and a fight between two kids that erupted into a major street brawl). So what this fingerprinting exercise is all about is reducing MINOR incidents.
So let's call this what it is. It's not about cutting down on serious offences - it's about reducing MINOR offences by banning people from pubs who happen to have lost their tempers or done any of the usual things that drunk people tend to do.
Is that worth the loss of privacy that this entails?
I was watching the movie "V for Vendetta" again last night - as a graphic novel that was written in 1982 it's eerily predictive. For a movie made two years ago, it's practically a documentary.
I'm a Brit who has been living in the USA for the past 13 years and it's hard to say which is more like the movie. Britain with more spycams per person than anywhere else on earth - and soon you can't even have a beer without being fingerprinted! Or perhaps it is the USA in which the faceless secret police can monitor what books you check out from the library, bug your phone without judicial oversight and swoop down on you, merely accuse you of being a terrorist (no proof required) and on that pretext lock you up, torture you, ship you off to god-knows what hell-hole - and all without any right of trial or appeal?
Hmmm - hard call. Between the two countries - it's difficult to say which comes closest to the nightmare that V opposes in the movie. As he says: If you want to know whose fault this is - just look in the mirror.
Our own fear of statistically insignificant terrorist violence (or avian flu or WMD or drunk drivers or...you name it) induces progressively higher tolerance for the State to ratchet down the human rights of the entire population. There will come a point when we realise that this has been a terrible mistake - but will we do that before or after the point where we can no longer reverse it's effects?
Better get that bulk order for Guy Fawkes masks in before the rush. Amazon have them for $5.99.
> some form of genetic isolation (which may or may not mean geographical isolation,
I think the author (who is an economist by the way) was presuming that rich, tall, glamorous people meet each other in their exclusive up market settings and have rich, tall, glamorous kids. This provides the necessary isolation for genetics to do their thing - and in some ridiculously short period of time, it's a case of Morloks and Eloi.
Sadly, this guy knows nothing of value about genetics - and not too much about society either.
Remember that he also believes that these new humans will have coffee-coloured skin due to racial mixing. It's not clear to me why he believes that racial self-segregation will vanish while other forms of segregation are on the increase.
TFA is all just ridiculous speculation that no true evolutionary biologist would buy into.
The problem with predicting human evolution is that we are rather more in control of it than is the case for other animals. Genetic failures can be fixed up by medicine such that genes that would have been fatal (and thus eliminated from the pool) are now viable if the symptoms they produce are treatable.
Consider this: if there were a strong reproductive bias for taller men - then genetically shorter men would come to be considered 'abnormal' by our society. This would cause them to be treated with growth hormone at a tender age - hiding their genetic predisposition and thus allowing the gene for 'shortness' be available as procreators of the next generation. This would make genetically induced tallness a hard trait to take hold.
If you doubt this could ever happen - please note that it has already happened:
http://www.thenewatlantis.com/archive/7/fox.htm
"In July 2003, the Food and Drug Administration (FDA) authorized pharmaceutical companies to promote human Growth Hormone (hGH) for use in children who are very short but not suffering from any specific illness or medical condition. Parents are now using hGH in record numbers, hoping that hormone treatment will give their kids happier childhoods and more prosperous adulthoods."
QED.
Wasn't there also talk about ice in deep craters situated where the sun can never shine on it?
...at least that's what I recall. It sounds kinda plausible.
The theory was (and I hope I have this right) that cometary ice must impact the moon from time to time - so there is water there from time to time - but whenever the sun shines, in the absence of an atmosphere, the water will evaporate (sublimate?) away quite quickly during the lunar day - then freeze out of the atmosphere during the night.
This mechanism would generally keep whatever water molecules there is up there moving around...*UNTIL* (by chance) it lands somewhere where there is never any sunlight - inside a cave or a deep crater. At that point it must settle - and there is no longer a mechanism to move it around again. With no atmosphere to scatter sunlight, permenantly dark places will be profoundly cold.
It follows then that whatever water there is will always end up in these relatively rare places EVENTUALLY - so given enough time, all of the moon's water would end up stashed away in just a few easy-to-predict places.
Furthermore, we'd never be able to see those places from earth-bound or low orbit telescopes because any place we can see must also collect sunlight at some point in the lunar orbit.
The US Nimitz class Aircraft carriers each carry a single 194MWatt nuclear reactor. The USS Enterprise has a total of eight nuclear reactors onboard. All but two US carriers - and absolutely all US military submarines are nuclear powered. Even ships as small as cruisers have been nuclear powered in the past.
French Rubis class submarines each have a 48 MW reactor.
Russian Typhoon class submarines carry two 190MWatt reactors.
Russian Arktika class ice breakers carry two reactors of 171MWatts each. The Taimyr class have 135MWatt reactors. There are a total of ten Russian civilian nuclear ice breakers in active service.
So a couple of measley 60MWatt reactors on a barge somewhere isn't really the huge news you might think.
Having reactors on a barge that's moored someplace has gotta be safer than having them than crashing off through icebergs or sitting off the coast of countries full of terrorists who would love nothing better than to drive a boatload of explosives into them - or yet putting them on submarines whose safety record doesn't really look so great.
It's been suggested that the two reactors they are using on this barge came from decomissioned Russian submarines anyway...so we're probably better off having them used for peaceful purposes and being moored someplace where we know exactly where they are!
Yep - and the so-called 'Clean Coal' approach concentrates naturally occurring radioactivity to the extent that the waste produced by even the most modern coal fired power plants has comparable amounts of radioactivity to nuclear plants.
Nuclear power has problems - but they are all solvable within our technological reach. The problems of irreplacable fossil fuels combined with the bad consequences of dumping CO2 into the atmosphere are not in any way solvable with technologies we currently have - or even expect to have. Windmills, wave power, solar power , biofuels and others aren't likely to produce the quantity of power we expect to need over the coming years. Fusion looks cool - but we can't do it yet.
So whilst nuclear power is *HARD* - it has the huge benefit of not being *IMPOSSIBLE* like all of the other power sources we have.
A modern GPU might have as many as 48 "fragment shader" processors inside it - so there are really around 21,500 processors versus 282,000 CPU's. Then each GPU processor works in full four-way arithmetic parallelism - so it can do arithmetic and data move operations on four numbers just as fast as one. So with the right mapping of algorithm to processing, you have 86,000 floating point arithmetic units...they only need to be about 3x faster than the CPU's.
But these are not general purpose processors - in some respects they are horribly, horribly limited.
So if you have an algorithm that is enough like rendering polygons with textures - the GPU is just insanely fast...if your algorithm isn't enough like that - then you may not be able to run it on the GPU at all.
Oh good grief Mozilla guys!
Look - FireFox is OpenSourced - right? So for chrissakes let them
do what they want with it - that is THE ENTIRE POINT!!! If the
Debian guys (who are not exactly complete Klutzes at this stuff)
mess up, you say "Hey the Debian guys screwed up - come download
the real one from the usual places."
Geez - just make it happen and get over it.
On Apollo 12:
"Whoopie! Man, that may have been a small one for Neil, but that's a long one for me." -- Pete Conrad
On Apollo 14:
"It's been a long way, but we're here." - Alan Shepard
Hmmmm - not *quite* so memorable.
In fact, the cost in environmental impact per load carried to orbit can be arbitarily small - if the climber goes slowly enough, you could power it with solar power. The problem is that the capital cost of the system is non-trivial so you have to get a lot of usage out of the thing in order to pay your investors. So it might not be acceptable for the climber to ascend at a snails pace. Faster climbers mean more intense fuel sources - so we end up with HUGE ground-based lasers pumping light energy into solar panels.
The first task of a space elevator is probably to make a second one.
It does seem close to impossible - but the benefits if it IS possible are huge.
The big thing about getting a rocket into space is that almost all of the launch weight is the fuel - almost all of which is consumed in propelling fuel.
The elevator cab would carry no fuel at all. Because it can be slow, it doesn't need much power - so it can be powered with solar panels illuminated from below using powerful lasers. The fuel for the lasers doesn't have to be propelled into space - so there is VASTLY less energy required - and it doesn't have to be in a dense, portable form like rocket fuel. You could power the system with windmills or something.
Furthermore, the cab can trundle slowly down the cable when it wants to come home. No reentry heat - no expensive insulation tiles - no wings, no undercarriage. Again, this makes the cab smaller, lighter, cheaper - it really just needs to be an airtight box with some wheels to grip the cable, solar cells for power and some oxygen to breath (for manned "flights").
The cab should be 100% re-usable (maybe needs new tires once in a while!).
This makes space VERY cheap. Once you are at geosynchronous orbital height, you can launch satellites by pushing them out of the cargo bay with a spring or something...again, very, very low-tech.
If you want to launch something out of earth orbit then climb further up the cable beyond geosynchronouse orbital height and (once again), push the vehicle out of the window and it'll shoot out of earth orbit with no fuel needed.
So the capital costs of getting this thing up there might be a LOT higher than a shuttle launch - or perhaps more expensive than dozens of shuttle launches. But once you've built it, the cost to get a ton of stuff into orbit will be comparable to driving it across the USA in an 18 wheeler. Assuming there is a market for getting lots of stuff into orbit cheaply, this thing will pay for itself very quickly.
BUT - can it be built at all? Even in theory?
That's all down to materials science in the tether. Is there any material that's light enough and strong enough and reliable enough and cheap enough? If there is then this will happen for sure - and within not many years of a suitable tether material being discovered- if there isn't, it won't. The feasibility and economics of the thing are a no-brainer if you can solve the tether material problem.
The problem is weight. It takes a massively strong material (possibly impossibly strong) to support it's own weight over all of those miles. If you cut holes in it, it gets weaker and makes the problem much, much worse. If you stick extra bits onto it, it gets heavier - and that means it has to be stronger.
In the end, it's much simpler to build the leanest, lightest, stongest tether (then engineer the heck out of the elevator cab so it can climb it) rather than engineering the tether to make it easier to climb.
The project is right on the edge of being possible - the tensile strength to mass ratio is a couple of orders of magnitude more than that provided by any known materials *except* carbon nanotubes - which could *MAYBE* handle the forces if you could make unbroken, flawless strands that were long enough (I think the longest ones we can make right now are under a millimeter long - and we need a few thousand miles of the stuff).
100% of this problem is the tether - if a strong enough tether could be devised, the rest of the elevator would be more or less easy to do.
A modern graphics chip (GPU) has 48 fragment processors and a dozen more vertex processors - each of those does full scale floating point in four way parallelism at fairly respectable clock rates. So if all you need is a couple of hundred floating point units - you already have it.
The problem is the memory bandwidth into and out of this beast. The number of pins you can have is fundamentally limited - the bandwidth down each pin is fairly sharply limited. RAM chips are creeping up in speed V-E-R-Y S-L-O-W-L-Y compared to CPU speeds. So in the end, I think it's pretty much irrelevent how many processors you have on a chip because even with infinite compute power, if you can't get the data in and out, you're limited to RAM speeds.
The GPU's only get their speed and the ability to use all of that floating point horsepower 100% of the time (with no nasty Ahmdahl's law consequences) because they are working on Red/Green/Blue/Alpha or X/Y/Z/W in parallel - and millions of pixels are being computed using the exact same algorithm in perfect lockstep with few (if any) branches or loops. That means that a single instruction stream can feed all of those processors in parallel - and the data goes in and comes out in highly memory-friendly ways.
That's not much use for general computing though.
I glaze over when I hear about increasing numbers of processors - if the memory interfaces aren't designed very carefully PS-3 then very bad things can happen. Ask about how much RAM bandwidth this gizmo has.
No! The guys have the PERFECT excuse. In zero g, there is no up or down.
My 1963 Austin Mini gets 55mpg - designed by one guy in just a few months. It'll go at 70mph and you can actually get four people into it.
Fast forward to my 2005 Mini Cooper - designed by a huge team of engineers with big computers. It weighs more than twice as much as the '64 version, gets less than half the MPG and whilst it'll go 140mph - it's a real squash to get more than two people in it.
Why is that?
* Safety.
* Air conditioning.
* Emissions standards.
The safety thing is arguably justified - and in many parts of the world, so is the A/C - but the idea that you burn twice as much fuel so that the concentration of pollutants goes down - but the absolute amount coming out of the tailpipe goes up - is entirely an artifact of broken laws. If the laws regulated the total amount of emissions rather than the percentage of CO, NOx, etc - then SUV's would be a thing of the past in pretty short order. But right now, the laws keep out a number of very low-polluting smaller cars BECAUSE they can't meet emissions standards.
It's automotive Bizarro-land.