My one experience "coding assembly" was 20 years ago as an undergrad visiting one of the experiments at Fermilab. They had electronic detectors triggered various ways sending data to an old Digital PHP system that was supposed to analyze each event as quickly as possible, decide whether it was interesting enough to save to magnetic tape, and then go on to the next event a few microseconds later. The data acquisition code was, naturally, in assembly - and boy they had that pared down to the absolute essentials, not a wasted instruction. My job was to try to, instead of recording to tape, to send the data over a wire to a new VAX machine that had just arrived.
Not sure I ever ran into Foster though - I wonder what experiments he was on? Actually, I have met him since then, but that's another story...
Here's the text from the "release 6" developer preview 1:
Java for Mac OS X 10.4, Release 6 delivers improved reliability and compatibility for Java 2 Platform Standard Edition 5.0 and Java 1.4 on Mac OS X 10.4.10 and later. This release updates J2SE 5.0 to version 1.5.0_13 and Java 1.4 to version 1.4.2_16.
Actually you have that backwards in several respects. The report goes into this a bit.
First, the amount of energy humans use is minuscule compared to the main flows of light to the Earth and heat back out - about 1 part in 10,000. Even if we replaced all human energy sources by space solar power, that added 1 part in 10,000 incoming energy will change temperatures at most something like 0.01 or 0.02 degrees - really irrelevant.
Second, solar cells absorb essentially all the sunlight that hits them. Most of that energy goes into waste heat of the ground since the cells are only 20% or so efficient. That *increases* the albedo of the ground area they cover and increases Earth's net absorption of heat. If we replaced all Earth energy use with solar cells that were 20% efficient, we'd be adding something like 2-3 times human energy use as heat input to the planet, where the energy is being absorbed instead of just reflected back out. With satellites, the energy is captured in space so the waste heat has to be dealt with up there, not down here. Further, the ground receiver doesn't care what it's properties are in visible light, assuming we're transmitting energy at a non-visible frequency, so it can be as reflective as you like and actually decrease the ground albedo. So an SSP system could use its ground systems to completely compensate for any added energy to the planet, something that a ground solar array cannot do.
250 is the average across Earth's surface, per unit surface area. If you were able to keep the array pointed directly at the sun all day long, you'd get 500 (1/2 of 1000, since half the time is still night-time). But that takes extra equipment, additional expense on the local side, and requires extra ground-area one way or another (that's a long shadow you're casting near sunries/sunset).
Well, if average sunlight falling on Earth's surface is 250 W/m^2 and average in Earth orbit is 1400 W/m^2, then 36% of 250 is a lot less than 36% of 1400. That's the difference there.
The main proposed orbit is geo-stationary; these are very rarely shaded (for about 1 month of the year a satellite there gets about an hour's worth of shade every day, the rest of the year it's clear).
Inexpensive space launch is definitely one of the technical challenges. The report calls for large-scale development and deployment of reusable launch vehicles and development of inexpensive orbital transfer vehicles (solar electric space tugs) to handle the launch challenge. A lot of people think it can be done, but nobody's succeeded yet obviously.
No weather, and a clear view (no atmosphere at all in the way).
That gives you a factor between 5 and 10 over on-the-ground systems to start with.
If you really lose 50% in transmission *and* 50% in receiving the case is harder to make - most estimates seem to have higher numbers for overall system end-to-end efficiency, but of course nobody's buit one yet.
On the right hand side of the main DailyKos page you'll see "recommended" diaries and "recent" diaries. Anybody can sign up to the site and write diaries (1 per day) - basically political essays (or on anything if you like). Get yourself recommended by other users, and you're there.
Some things it doesn't mention though, that I recall from my brief summer there 20+ years back:
* the radioactive groundhogs. Every national lab I've been to seemed to have a colony of groundhogs, I guess they like the security.... At Fermilab, there was a burrow in the middle of a mile-long berm of dirt that acted as a beam dump to generate neutrinos (only neutrinos make it through that much matter without being stopped).
* Wilson's artworks - I assume they're still around. Robert Wilson was the instigator of the lab, and got it built on time and under budget. He was also a bit of a sculptor, and a number of his artworks were on the grounds around the administration building. In fact I think he designed the rather unique admin building too.
* the annual "race around the ring" - actually, maybe that's gone away since Leon Lederman's no longer the lab director. It was quite a challenge when I was there though; you can imagine a bunch of desk physicists and engineers trying to make it around the 3+ miles of the ring road in a reasonable amount of time...
Plus, they fed us great buffalo burgers a few times when I was there as an undergrad student helping with one of the projects:-) That was a loooong time ago though, I wonder where the bison go these days.
But it sounds like that is what they're proposing. As far as I'm aware, the natural flow of geothermal energy from below the surface is only 45 TW, and the world already using close to 15 TW, so the total available is 3 times world energy use, not 250,000 times ???
It would if human energy use grew by several orders of magnitude - but relative to current fossil or nuclear options, space solar power adds *less* energy to Earth (see discussion above). And hopefully we'll move most of our energy use into space by the time we're using that much energy.
The energy involved is tiny - total human primary energy use now is 13 TW; the Earth receives 174,000 TW from the Sun. Even if we increase human energy use 10-fold thanks to space solar power, we're adding less than 0.1% to what the Earth receives.
And note that burning fossil fuels or nuclear power also adds heat (that 13TW is almost entirely nonrenewable) and the conversion to electricity on the planet is only 35-40% efficient, so 60% of that energy is waste heat that we don't even use. With space solar power the waste is almost all in space - the Earth-side would have less than 20% waste heat; space power would cut the net addition of direct energy to Earth by 1/2 directly.
And even more important is the CO2 effect - that adds far more heat to the planet thanks to greenhouse trapping than the heat released in burning the fossil fuels.
So, hands-down, space power adds far less energy to the "ecosystem" than any other realistic alternative.
Well, the idea of bootstrapping is that you can start with a device in orbit of, say, 1/10th the final mass, and then it can build itself from there by pulling up pieces starting at 1/10th the final payload mass. That's reasonable with a mass ratio of 10's to 100's, maybe even 1000. But totally out of the ballpark for ratios in the millions, as I was saying.
But that's pretty much what Laine was saying too - an elevator right now is "difficult", but not "impossible", which was the question he was responding to there. There are materials that you could build it with today, but it would cost you trillions of dollars, at least, for very limited capacity. So it wouldn't serve any useful purpose, but it wouldn't be impossible.
Up to you if you need it, but that's not the lower bound on building an elevator with spectra. The lower bound is the safety-factor-1 number, which with your values is 95 billion kg, or 95 million metric tons. That's well within plastics manufacturing capacity of today, though a lot more than the current annual market for spectra itself
The problem with this sort of number is it takes tens of millions of trips to lift itself up, so the bootstrapping technique that is normally assumed just doesn't help. So you could build it with spectra in principle, but what would be the point? Having an elevator with a mass ratio down in the thousands rather than the millions seems to be essential, and that means materials with at least 20 GPa strength. We're obviously not there yet.
None of this is hidden. Laine worked with Brad Edwards on "HighLift Systems" (google it for some background) which did a NASA-sponsored study of space elevator engineering with nanotubes, but the basic engineering isn't that different for other materials, you just have to taper the ribbon more aggressively. You could get a copy of Liftport's book - of course I'm slightly biased since I wrote one of the chapters. It's a mix of fiction (some really good, some not so good) and essays on the basic engineering challenges. "Liftport: Opening Space to Everyone", you can find it in their "store" or Amazon, etc.
The 7.303 billion tons is carbon, not CO2. CO2 has those extra 2 oxygen atoms that makes it about 3.67 times heavier, so 7.303 billion tons of carbon is 26.8 billion tons of CO2. Changes your numbers a bit.
Amdahl's law still applies - look at Gustafson's graphs. The assumption is that the serial part stays constant while the parallel part grows with the number of processors (because you are handling a larger number of problems). Therefore the serial fraction shrinks, and the speedup available is larger. If you tried to run the large problem on a single CPU it really would take that much longer. But that's only in this sort of case where the serial fraction is independent of problem size. Seems like a pretty special case to me, but in any case, Amdahl's law is as valid as ever here.
I've worked with parallel software for years - there are lots of ways to do it, lots of good programming tools around even a couple of decades back (my stuff ranged from custom message passing in C to using "Connection-Machine Fortran"; now it's java threads) but the fundamental problem was stated long ago by Gene Amdahl - if half the things you need to do are simply not parallelizable, then it doesn't matter how much you parallelize everything else, you'll never go more than twice as fast as using a single thread.
Now there's been lots of work on eliminating those single-threaded bits in our algorithms, but every new software problem needs to be analyzed anew. It's just another example of the no-silver-bullet problem of software engineering...
stop/go traffic lets the Prius run purely off the battery. If the top speed is not too high, the ground is flat, and you don't have much in the way of accessories, you can go through a lot of traffic without the gasoline engine turning on at all - effectively very high mpg, though it gradually drains the battery. The braking system does more than "capture heat" - it runs the electric motor in reverse, doing just the opposite of what the motor did to get the car moving. Not quite a perpetual motion machine, but far far better efficiency-wise than anything you can do by burning gasoline.
There have been a lot of responses here, but I don't think they quite answer your question. The fundamental problem is that human causation of global warming is market failure on a massive scale. It profoundly violates the free-markets-solve-everything government-is-the-problem individual-greed-is-all-we-need concepts that are the bedrock foundation of modern conservative belief.
When your fundamental beliefs are being so profoundly proved inadequate, denying scientific facts becomes essential.
You'll notice that, once most people accept the science, they cease being conservatives in this American "libertarian" sense. Because regulation and government action to change the basic market rules is the only way we're going to solve this one. The die-hards are going to be harder to change, but they'll be coming. Bush was at 40%, then 35%. Now it's under 30%. There's hope...
The truth is, we've sent far more and better spacecraft to Mars in the last few decades than to the Moon. The only things the US has sent to the Moon since 1972 have been Clementine, a DoD low-cost project that didn't have anywhere near a good enough camera, and Lunar Prospector, another low-budget item that had no camera at all. Galileo swung by briefly, but not enough to take close-range pictures. Europe has sent SMART-1, again decidedly low-budget: it took over a year to get there and was mainly for testing other things besides photography.
But that's the Moon for you - the inner city of the solar system that everybody says they care about but nobody does anything.
It was pretty cool seeing the teams trying to climb the tether. I only saw a couple make it to the top (200 ft), but several got part way. I don't believe anybody beat the 1-minute time limit to meet the goal.
One interesting thing is that, having to power the climbers from beamed power, they had to make them as light as possible, relative to the area of solar panels trying to capture energy. So these were pretty flimsy looking devices, and you could see wind causing trouble. Stripped bolts and computer glitches also caused their share of failures...
It was also nice to see all those young teams of excited people trying to do this - mostly undergraduate engineering students, but there were even some high school students participating.
And having John Carmack hanging out chatting with the crowd while his crew was trying to get his "hover" craft back in shape was fun. They had jumbotron displays for their challenge attempts, but you could also see it just hovering there a hundred feet up (not too close to the crowd, but quite visible). Of course the crashes had a bit of a car-wreck interest too... The most successful things seemed to be some straightforward high powered rocket launches. But there was a big enthusiastic crowd, and lots of sideshows. Definitely worth a trip to the El Paso area if they do this again!
As all the discussion about cheats indicates, "telepathy" is a word for some "magical" form of communication between people; given that we have lots of real-life means of communciation between people, and more and better ones coming out every year, it's almost certain that within a few decades humans will be communicating with one another via means that are essentially indistinguishable from classic telepathy.
That doesn't mean it was likely to have evolved naturally though. There does seem to be a whiff of real "irreducible complexity" in an iPod...
Losses in electric transmission are typically less than 10% from power plant to outlet. Steam-turbine power plants convert the chemical energy of coal or oil to heat (by burning) and then to electricity through an engine cycle that is limited to 35% efficiency or so, but combined-cycle gas turbines can be 60% efficient by making use of mechanical as well as thermal output from the chemical fuel. Batteries typically return over 90% of their charging energy for use. Electric motors are also very efficient, though you'll get some losses with the drivetrain (as you do with an internal combustion engine as well). And regenerative braking is very natural with electric vehicles, so the losses from braking may go away almost completely. So even with coal steam turbine generators as the source you're still getting well over 25% from the initial chemical energy to the energy supplied to move your car (minus drivetrain losses). With internal combustion engines you're lucky to get 20%.
Slashdot Mirror
My one experience "coding assembly" was 20 years ago as an undergrad visiting one of the experiments at Fermilab. They had electronic detectors triggered various ways sending data to an old Digital PHP system that was supposed to analyze each event as quickly as possible, decide whether it was interesting enough to save to magnetic tape, and then go on to the next event a few microseconds later. The data acquisition code was, naturally, in assembly - and boy they had that pared down to the absolute essentials, not a wasted instruction. My job was to try to, instead of recording to tape, to send the data over a wire to a new VAX machine that had just arrived.
Not sure I ever ran into Foster though - I wonder what experiments he was on? Actually, I have met him since then, but that's another story...
No J2SE 6 there.
Ha, my uid is less than your uid :-)
And I started with Linux kernel 0.99, so there!
Is there a Godwin's law for computer nostalgia discussions?
Actually you have that backwards in several respects. The report goes into this a bit.
First, the amount of energy humans use is minuscule compared to the main flows of light to the Earth and heat back out - about 1 part in 10,000. Even if we replaced all human energy sources by space solar power, that added 1 part in 10,000 incoming energy will change temperatures at most something like 0.01 or 0.02 degrees - really irrelevant.
Second, solar cells absorb essentially all the sunlight that hits them. Most of that energy goes into waste heat of the ground since the cells are only 20% or so efficient. That *increases* the albedo of the ground area they cover and increases Earth's net absorption of heat. If we replaced all Earth energy use with solar cells that were 20% efficient, we'd be adding something like 2-3 times human energy use as heat input to the planet, where the energy is being absorbed instead of just reflected back out. With satellites, the energy is captured in space so the waste heat has to be dealt with up there, not down here. Further, the ground receiver doesn't care what it's properties are in visible light, assuming we're transmitting energy at a non-visible frequency, so it can be as reflective as you like and actually decrease the ground albedo. So an SSP system could use its ground systems to completely compensate for any added energy to the planet, something that a ground solar array cannot do.
250 is the average across Earth's surface, per unit surface area. If you were able to keep the array pointed directly at the sun all day long, you'd get 500 (1/2 of 1000, since half the time is still night-time). But that takes extra equipment, additional expense on the local side, and requires extra ground-area one way or another (that's a long shadow you're casting near sunries/sunset).
Well, if average sunlight falling on Earth's surface is 250 W/m^2 and average in Earth orbit is 1400 W/m^2, then 36% of 250 is a lot less than 36% of 1400. That's the difference there.
The main proposed orbit is geo-stationary; these are very rarely shaded (for about 1 month of the year a satellite there gets about an hour's worth of shade every day, the rest of the year it's clear).
Inexpensive space launch is definitely one of the technical challenges. The report calls for large-scale development and deployment of reusable launch vehicles and development of inexpensive orbital transfer vehicles (solar electric space tugs) to handle the launch challenge. A lot of people think it can be done, but nobody's succeeded yet obviously.
Saves on transmission and storage.
No weather, and a clear view (no atmosphere at all in the way).
That gives you a factor between 5 and 10 over on-the-ground systems to start with.
If you really lose 50% in transmission *and* 50% in receiving the case is harder to make - most estimates seem to have higher numbers for overall system end-to-end efficiency, but of course nobody's buit one yet.
On the right hand side of the main DailyKos page you'll see "recommended" diaries and "recent" diaries. Anybody can sign up to the site and write diaries (1 per day) - basically political essays (or on anything if you like). Get yourself recommended by other users, and you're there.
The bison are indeed mentioned in the article.
Some things it doesn't mention though, that I recall from my brief summer there 20+ years back:
* the radioactive groundhogs. Every national lab I've been to seemed to have a colony of groundhogs, I guess they like the security.... At Fermilab, there was a burrow in the middle of a mile-long berm of dirt that acted as a beam dump to generate neutrinos (only neutrinos make it through that much matter without being stopped).
* Wilson's artworks - I assume they're still around. Robert Wilson was the instigator of the lab, and got it built on time and under budget. He was also a bit of a sculptor, and a number of his artworks were on the grounds around the administration building. In fact I think he designed the rather unique admin building too.
* the annual "race around the ring" - actually, maybe that's gone away since Leon Lederman's no longer the lab director. It was quite a challenge when I was there though; you can imagine a bunch of desk physicists and engineers trying to make it around the 3+ miles of the ring road in a reasonable amount of time...
Plus, they fed us great buffalo burgers a few times when I was there as an undergrad student helping with one of the projects :-) That was a loooong time ago though, I wonder where the bison go these days.
But it sounds like that is what they're proposing. As far as I'm aware, the natural flow of geothermal energy from below the surface is only 45 TW, and the world already using close to 15 TW, so the total available is 3 times world energy use, not 250,000 times ???
It would if human energy use grew by several orders of magnitude - but relative to current fossil or nuclear options, space solar power adds *less* energy to Earth (see discussion above). And hopefully we'll move most of our energy use into space by the time we're using that much energy.
The energy involved is tiny - total human primary energy use now is 13 TW; the Earth receives 174,000 TW from the Sun. Even if we increase human energy use 10-fold thanks to space solar power, we're adding less than 0.1% to what the Earth receives.
And note that burning fossil fuels or nuclear power also adds heat (that 13TW is almost entirely nonrenewable) and the conversion to electricity on the planet is only 35-40% efficient, so 60% of that energy is waste heat that we don't even use. With space solar power the waste is almost all in space - the Earth-side would have less than 20% waste heat; space power would cut the net addition of direct energy to Earth by 1/2 directly.
And even more important is the CO2 effect - that adds far more heat to the planet thanks to greenhouse trapping than the heat released in burning the fossil fuels.
So, hands-down, space power adds far less energy to the "ecosystem" than any other realistic alternative.
Well, the idea of bootstrapping is that you can start with a device in orbit of, say, 1/10th the final mass, and then it can build itself from there by pulling up pieces starting at 1/10th the final payload mass. That's reasonable with a mass ratio of 10's to 100's, maybe even 1000. But totally out of the ballpark for ratios in the millions, as I was saying.
But that's pretty much what Laine was saying too - an elevator right now is "difficult", but not "impossible", which was the question he was responding to there. There are materials that you could build it with today, but it would cost you trillions of dollars, at least, for very limited capacity. So it wouldn't serve any useful purpose, but it wouldn't be impossible.
Up to you if you need it, but that's not the lower bound on building an elevator with spectra. The lower bound is the safety-factor-1 number, which with your values is 95 billion kg, or 95 million metric tons. That's well within plastics manufacturing capacity of today, though a lot more than the current annual market for spectra itself
The problem with this sort of number is it takes tens of millions of trips to lift itself up, so the bootstrapping technique that is normally assumed just doesn't help. So you could build it with spectra in principle, but what would be the point? Having an elevator with a mass ratio down in the thousands rather than the millions seems to be essential, and that means materials with at least 20 GPa strength. We're obviously not there yet.
None of this is hidden. Laine worked with Brad Edwards on "HighLift Systems" (google it for some background) which did a NASA-sponsored study of space elevator engineering with nanotubes, but the basic engineering isn't that different for other materials, you just have to taper the ribbon more aggressively. You could get a copy of Liftport's book - of course I'm slightly biased since I wrote one of the chapters. It's a mix of fiction (some really good, some not so good) and essays on the basic engineering challenges. "Liftport: Opening Space to Everyone", you can find it in their "store" or Amazon, etc.
Just a minor correction...
The 7.303 billion tons is carbon, not CO2. CO2 has those extra 2 oxygen atoms that makes it about 3.67 times heavier, so 7.303 billion tons of carbon is 26.8 billion tons of CO2. Changes your numbers a bit.
Amdahl's law still applies - look at Gustafson's graphs. The assumption is that the serial part stays constant while the parallel part grows with the number of processors (because you are handling a larger number of problems). Therefore the serial fraction shrinks, and the speedup available is larger. If you tried to run the large problem on a single CPU it really would take that much longer. But that's only in this sort of case where the serial fraction is independent of problem size. Seems like a pretty special case to me, but in any case, Amdahl's law is as valid as ever here.
I've worked with parallel software for years - there are lots of ways to do it, lots of good programming tools around even a couple of decades back (my stuff ranged from custom message passing in C to using "Connection-Machine Fortran"; now it's java threads) but the fundamental problem was stated long ago by Gene Amdahl - if half the things you need to do are simply not parallelizable, then it doesn't matter how much you parallelize everything else, you'll never go more than twice as fast as using a single thread.
Now there's been lots of work on eliminating those single-threaded bits in our algorithms, but every new software problem needs to be analyzed anew. It's just another example of the no-silver-bullet problem of software engineering...
stop/go traffic lets the Prius run purely off the battery. If the top speed is not too high, the ground is flat, and you don't have much in the way of accessories, you can go through a lot of traffic without the gasoline engine turning on at all - effectively very high mpg, though it gradually drains the battery. The braking system does more than "capture heat" - it runs the electric motor in reverse, doing just the opposite of what the motor did to get the car moving. Not quite a perpetual motion machine, but far far better efficiency-wise than anything you can do by burning gasoline.
There have been a lot of responses here, but I don't think they quite answer your question. The fundamental problem is that human causation of global warming is market failure on a massive scale. It profoundly violates the free-markets-solve-everything government-is-the-problem individual-greed-is-all-we-need concepts that are the bedrock foundation of modern conservative belief.
When your fundamental beliefs are being so profoundly proved inadequate, denying scientific facts becomes essential.
You'll notice that, once most people accept the science, they cease being conservatives in this American "libertarian" sense. Because regulation and government action to change the basic market rules is the only way we're going to solve this one. The die-hards are going to be harder to change, but they'll be coming. Bush was at 40%, then 35%. Now it's under 30%. There's hope...
The truth is, we've sent far more and better spacecraft to Mars in the last few decades than to the Moon. The only things the US has sent to the Moon since 1972 have been Clementine, a DoD low-cost project that didn't have anywhere near a good enough camera, and Lunar Prospector, another low-budget item that had no camera at all. Galileo swung by briefly, but not enough to take close-range pictures. Europe has sent SMART-1, again decidedly low-budget: it took over a year to get there and was mainly for testing other things besides photography.
But that's the Moon for you - the inner city of the solar system that everybody says they care about but nobody does anything.
It was pretty cool seeing the teams trying to climb the tether. I only saw a couple make it to the top (200 ft), but several got part way. I don't believe anybody beat the 1-minute time limit to meet the goal.
One interesting thing is that, having to power the climbers from beamed power, they had to make them as light as possible, relative to the area of solar panels trying to capture energy. So these were pretty flimsy looking devices, and you could see wind causing trouble. Stripped bolts and computer glitches also caused their share of failures...
It was also nice to see all those young teams of excited people trying to do this - mostly undergraduate engineering students, but there were even some high school students participating.
And having John Carmack hanging out chatting with the crowd while his crew was trying to get his "hover" craft back in shape was fun. They had jumbotron displays for their challenge attempts, but you could also see it just hovering there a hundred feet up (not too close to the crowd, but quite visible). Of course the crashes had a bit of a car-wreck interest too... The most successful things seemed to be some straightforward high powered rocket launches. But there was a big enthusiastic crowd, and lots of sideshows. Definitely worth a trip to the El Paso area if they do this again!
As all the discussion about cheats indicates, "telepathy" is a word for some "magical" form of communication between people; given that we have lots of real-life means of communciation between people, and more and better ones coming out every year, it's almost certain that within a few decades humans will be communicating with one another via means that are essentially indistinguishable from classic telepathy.
That doesn't mean it was likely to have evolved naturally though. There does seem to be a whiff of real "irreducible complexity" in an iPod...
Losses in electric transmission are typically less than 10% from power plant to outlet. Steam-turbine power plants convert the chemical energy of coal or oil to heat (by burning) and then to electricity through an engine cycle that is limited to 35% efficiency or so, but combined-cycle gas turbines can be 60% efficient by making use of mechanical as well as thermal output from the chemical fuel. Batteries typically return over 90% of their charging energy for use. Electric motors are also very efficient, though you'll get some losses with the drivetrain (as you do with an internal combustion engine as well). And regenerative braking is very natural with electric vehicles, so the losses from braking may go away almost completely. So even with coal steam turbine generators as the source you're still getting well over 25% from the initial chemical energy to the energy supplied to move your car (minus drivetrain losses). With internal combustion engines you're lucky to get 20%.