Movie buffs might be tempted by Also sprach Zarathrustra. Another possibility might be Elton John's Rocket Man, though I'd prefer the Kate Bush cover version.
At the tackier end of the scale, what about Jamiroquai's Cosmic Girl, because we know why men really want to go into space:)
As far as the reduction-enlargement went, I'm fairly sure that Xerox later acknowledged the story about the R/E machines in a book on the company. And, as you yourself have confirmed, speed difference in the company's product line are artificially created by crippling lower-cost machines.
Do you still give the machines away for cost and make your money back on service contracts?
At least some Xerox photocopiers in the 1980's were sold in two variants: one with reduction and enlargement, and one without. Remember, enlargement and reduction was done optically back in those days, not in software. The only difference between the two models was the control panel and a tiny bit of electronics. The models were otherwise identical! Needless to say, when the machines were traded in and put on the secondhand market, most Xerox dealers "upgradeded" the machines before resale.
Many other photocopier models offering different speeds were identical except for the controller boards, and swapping those over wasn't uncommon either; in fact, at one stage the distributor used to officially sanction it because the manufacturer was screwing them over.
The Kyoto treaty has significant problems, it's true. The main one is that the developing countries were entirely left out. However, John Quiggin's blog links to a study which shows that, given that Kyoto is going to come into force, it's in Australia's economic interest to join up. In any event, the modelling suggests the economic costs to Australia of Kyoto are actually pretty small in the long run, and damn small compared to the potential costs of global warming.
When it came down to it, John Howard made it pretty clear that the reason for going to Iraq was to help "maintain the alliance" - in other words, curry favour with George in the expectation of a few tidbits (the FTA that does SFA for the Australian economy, signs up to the US's insane IP policies, and will hopefully be blocked by the Senate) in the future. Do you really think this is any different?
The USA, OTOH, went to war to make sure it can keep burning its oil.
Yeah, and who was first there in the bunker with Bush the lesser? The Australian government.
As for the solar chimney, it's one of a number of zero-emission energy projects happening around Australia at the moment. This is good. However, they are all tinkering at the margins at this stage, and dirty brown coal remains the primary source of Australia's electricity, and Australia has an outer suburban SUV boom every bit as large as the US's.
There are a large number of technologies that could provide fuel for fission reactors for much longer. These include reprocessing, breeding uranium from thorium, using more fuel-efficient reactors, and Fast Breeder Reactors that turn uranium-238 into fissionable plutonium. The Japanese are also looking into extracting uranium from seawater.
Nuclear reactors have issues, but fuel shortages are an avoidable one.
Right now it's about getting fossil fuel producing countries to even acknowledge that something is wrong. When Australia and the US ratify the Kyoto treaty, then the scientific debate can begin.
That's a mischaracterisation of Australia's opposition. The government has publicly acknowledged that the greenhouse effect is real and something needs to be done; it's just not prepared to do anything about it, and by refusing to sign curries favour with the current US administration, something the government views as more important than climate change. Not to mention, of course, its campaign donors in the mining industry.
If you think that that position - admitting there is a serious problem, but refusing to do anything about it - is particularly indefensible, you'd be right. However, the current Australian government has made quite a lot of indefensible but politically expedient policy decisions. I can't wait till they finally get the boot in the next 12 months or so.
In addition to the other safeguards you describe, the missiles were supposed to be password-protected by this PAL system. They were not. Senior politicians, including the Secretary of Defence at the time, were led to believe that this extra protection existed. It didn't.
And let's be blunt here. A single Minuteman launched at a major world city could kill millions of people. Doesn't it make you even slightly nervous that the military was prepared to discard one of the layers of security in the interests of making it easier to launch them, and lie to their bosses about it?
While I haven't seen the specs on this model, if they're anything like most minijets they'll chew fuel at an astounding rate of knots. While I can't find the stats for the Wren, the smallest model made by this company uses 250 grams (9 ounces) of fuel per minute at full throttle. Even assuming the Wren uses half the fuel, with 8 engines that's 1kg of fuel per minute. That's 1.6 *litres* of fuel per minute, or, if you like, about 140 seconds of flying time for every US gallon of fuel, if I've done my sums right.
This is why small GA aircraft use propellers, by the way.
I wondered exactly the same thing on my blog recently, and found the Cri-Cri, as well as the BD-5J, which uses a slightly larger turbojet to make a one-person kitplane that can fly up to 500 km/h!
The problem with them is that while the power-to-weight ratio (and thus max speed and altitude) is great, the fuel consumption is terrible, and to get reasonable fuel consumption and range you need to fly such high altitudes you need a pressurised cabin, further adding to the cost and complexity.
Read section 4 of the actual CC-by-sa license. Amongst the restrictions on those who resdistribute the work is that:
You may not distribute, publicly display, publicly perform, or publicly digitally perform the Work with any technological measures that control access or use of the Work in a manner inconsistent with the terms of this License Agreement.
There are a lot of different Creative Commons licenses, some of which are clearly not free in the GNU sense.
However, the by attribution/share-alike license seems like a fairly close analogue to the GPL for non-software content. It lets people freely use, copy, and make derivative works from your content, under the conditions that they acknowledge the source and make the derived work available under the same license. The one thing it doesn't do is insist on the availability of modifiable copy, which is going to be very problematic to define appropriately for a broad variety of media and for many is a moot point.
Anyway, while I can see why the GPL is appropriate for you if you're insisting on "source code", but in my opinion it doesn't make a license for non-software that doesn't have this requirement non-free.
I'm a contributor to the Wikipedia, which is licensed under the FDL. As such, I've seen a lot of debate about the license. In my opinion, while the intention is fine, the specifics of the text make it unsuitable for anything except software documentation. It is too long and complex, with too many over-specific provisions, many of which are designed around the assumption that it will be used for documentation.
It is my belief that if the Wikipedia was restarted from scratch, it would probably use the Creative Commons By-attribution share-alike license, at least for the text, which accomplishes essentially the same thing but is much, much clearer.
The BBC produces a number of channels of radio programming, the most notable of which for an international audience is the BBC World Service, which has some excellent global news coverage. All of this would be extremely useful for a blind person.
As an Aussie, however, my favourite is the Ashes on Test Match Special, where you can learn about all the lovely English ladies who bake the commentators delightful sponge cake for afternoon tea and, incidentally, follow the cricket.
While not up to the standards of the US, the EU contains several members whose militaries are quite sufficient to kick the arse of any invader on home turf, including the US, by the way. Not to mention the fact that the UK and France are nuclear powers.
Now, if you were going to make the more subtle argument that the EU relies on the US's military muscle to secure its oil supplies, you might have a point. But the claim that Europe needs a massive remilitarization to protect against a conventional military threat is complete hogwash.
I don't think your calculation is particularly informative, as the assumption that drag is a constant force is wildly inaccurate.
I understand that at in the automotive world, a very rough approximation used is that drag is proportional to v4. That calculation breaks down as you approach Mach 1, but I dunno what the situation is at supersonic speeds and the very low pressures at these very high altitudes (well beyond the range I thought you could lift a blimp to, to be honest).
Okay. Full disclosure: IAACS (I am a computer scientist).
Full disclosure: So am I (well, my PhD will be in software engineering, but I did a little bit of work on AI-related stuff along the way).
My view is simple - the assumption that faster hardware will automatically give us human-level AI has little evidence to back it up. My arguments for this are as follows:
The Church-Turing thesis says that new computers will be able to do the same things as old ones, just faster. I think there's an extension that says you'll only get, at best, a polynomial-time speedup (limited applications of quantum computers excepted).
Complexity theory means that faster computers help even less than the layman thinks if it turns out that your problem doesn't have a linear-time algorithm. As I understand it, a lot of the neural-net algorithms have training times exponential in the number of neurons, in which case faster computers don't help much at all.
Though we've made lots of progress on lots of specific problems that were once thought to be an instance of the AI problem, we're no closer to building a human-like AI than we were in 1960, despite the fact that the processing and storage available per inflation-adjusted dollar has increased maybe 100,000 fold since then.
While, in principle, I agree that you could model the human brain on a computer, the fact is we don't yet know that much about how it works - or, more accurately, we have some idea about the individual neuron level, a little bit of an idea at the highest level of observable behaviour, but a huge yawning gap in the middle between those two bodies of research. The biologists have got a whole bunch of new tools to investigate the subject over the past couple of decades, but they don't seem to have made any radical progress so far - or, if they have, they haven't told the rest of us!
It's been suggested, notably by Roger Penrose, that intelligence and consciousness is the result of funky non-computable quantum stuff happening in the brain. Now, there's very little evidence to support that being true, but it's not prudent to rule it out either.
In summary, it's possible that, once we have something to model, more computer power might help us model the important structures of the human brain, or an adaptation thereof. It's quite possible we have enough computer power to do it now. It's also quite possible that we'll never have enough computer power with a conventional computer to do it. And, finally, there's no guarantee that we'll ever figure out what's going on in the brain.
My key point, however, is the claim that super-intelligent AI systems will be the inevitable result of faster computers was wrong when it was first made in the 1960's, was wrong when it was made again in the late 1970's in all manner of popular books, was wrong when Kurzweil made it in the 1990's, and it's still wrong now.
Finally, my impression is that automakers weren't interested in EVs was that batteries, flywheels, etc., can't compete with the awesome energy and power density of gasoline- relative cost and performance of EVs just wasn't competitive, and was unlikely to be anytime soon.
You're right, the parent is wrong.
The electric cars of the early 90's were mostly about a California law requiring manufacturers to produce a certain number of zero emission vehicles to be able to sell their cars in Carlifornia. California wanted this to deal with smog, which is essentially a local pollution phenomenon, and one that EV's solve by shifting the pollution problem from cities to power plants. The manufacturers were never very keen on the idea, as they knew perfectly well that they couldn't build an EV with a competitive range. So, they built a succession of prototypes until the California legislature realized that physics wasn't on their side and until a fuel cell car (or hydrogen cheap enough to burn in a combustion engine) is along, a zero emission vehicle just isn't practical.
I think that once we have fast, large holographic storage, it will only be a short time before we have a strong general purpose AI, and after that the sky is the absolute limit.
Are you familiar with the Church-Turing thesis? Basically, every alternative computing formulation we've been able to come up with computes exactly the same set of functions as conventional computers. Given that, could you please explain to me how any particular hardware breakthrough (modulo a hypercomputer of some kind) is going to give us human-equivalent AI?
Bad acting. Poor plots. Bad writing. A theme tune so excruciating that I want to vomit every time I hear it. And you're worried about "violating canon"?
In Victoria, Australia, one of the power companies is planning to do a similar thing with coal, except they're going to churn out enough of it to supply most of the local market.
If it works, they're going to generate cheap, low-sulfur (and thus low-emission) diesel, run a whopping great electricity plant from the byproducts, and all the CO2 from the generation will be stuffed underground for a very long time. While it's not ideal, it's a heck of a lot better than the current situation (burn the coal straight into the atmosphere and import oil from overseas).
I wasn't trying to belittle the achievements of the X Prize entrants so far, minimize the achievement of whomever finally wins the prize, or claim that the incremental development that the X Prize encourages is a bad idea.
My initial post was in response to the wrong-headed idea that these small companies have achieved the same thing as the Chinese have done with their orbital flight, when the Chinese achievement of putting a person in orbit and bringing them back is much more challenging than a suborbital hop.
Might I point out that Lord Kenneth is a known miscreant who has pulled a number of stunts like this. Keep that in mind when evaluating the worth of his criticisms.
While building an X-Prize class of vehicle is an impressive achievement, it's a long, long way from putting somebody into orbit and retrieving them again. You require a hell of a lot more thrust to put somebody into orbit, the heat shielding requirements for the way down are much tougher, and you've got to be able to maintain life support for at least several hours rather than a few minutes.
I love the quote from the rover team leader. She not only took responsibility, she and her team managed to fix it! Now there's an idea more people in the federal government (particularly high up in the executive branch) might take some inspiration from...;)
Slashdot Mirror
At the tackier end of the scale, what about Jamiroquai's Cosmic Girl, because we know why men really want to go into space :)
Do you still give the machines away for cost and make your money back on service contracts?
Many other photocopier models offering different speeds were identical except for the controller boards, and swapping those over wasn't uncommon either; in fact, at one stage the distributor used to officially sanction it because the manufacturer was screwing them over.
When it came down to it, John Howard made it pretty clear that the reason for going to Iraq was to help "maintain the alliance" - in other words, curry favour with George in the expectation of a few tidbits (the FTA that does SFA for the Australian economy, signs up to the US's insane IP policies, and will hopefully be blocked by the Senate) in the future. Do you really think this is any different?
Yeah, and who was first there in the bunker with Bush the lesser? The Australian government.
As for the solar chimney, it's one of a number of zero-emission energy projects happening around Australia at the moment. This is good. However, they are all tinkering at the margins at this stage, and dirty brown coal remains the primary source of Australia's electricity, and Australia has an outer suburban SUV boom every bit as large as the US's.
Nuclear reactors have issues, but fuel shortages are an avoidable one.
That's a mischaracterisation of Australia's opposition. The government has publicly acknowledged that the greenhouse effect is real and something needs to be done; it's just not prepared to do anything about it, and by refusing to sign curries favour with the current US administration, something the government views as more important than climate change. Not to mention, of course, its campaign donors in the mining industry.
If you think that that position - admitting there is a serious problem, but refusing to do anything about it - is particularly indefensible, you'd be right. However, the current Australian government has made quite a lot of indefensible but politically expedient policy decisions. I can't wait till they finally get the boot in the next 12 months or so.
And let's be blunt here. A single Minuteman launched at a major world city could kill millions of people. Doesn't it make you even slightly nervous that the military was prepared to discard one of the layers of security in the interests of making it easier to launch them, and lie to their bosses about it?
This is why small GA aircraft use propellers, by the way.
The problem with them is that while the power-to-weight ratio (and thus max speed and altitude) is great, the fuel consumption is terrible, and to get reasonable fuel consumption and range you need to fly such high altitudes you need a pressurised cabin, further adding to the cost and complexity.
Doesn't that deal with most of the problem?
However, the by attribution/share-alike license seems like a fairly close analogue to the GPL for non-software content. It lets people freely use, copy, and make derivative works from your content, under the conditions that they acknowledge the source and make the derived work available under the same license. The one thing it doesn't do is insist on the availability of modifiable copy, which is going to be very problematic to define appropriately for a broad variety of media and for many is a moot point.
Anyway, while I can see why the GPL is appropriate for you if you're insisting on "source code", but in my opinion it doesn't make a license for non-software that doesn't have this requirement non-free.
It is my belief that if the Wikipedia was restarted from scratch, it would probably use the Creative Commons By-attribution share-alike license, at least for the text, which accomplishes essentially the same thing but is much, much clearer.
As an Aussie, however, my favourite is the Ashes on Test Match Special, where you can learn about all the lovely English ladies who bake the commentators delightful sponge cake for afternoon tea and, incidentally, follow the cricket.
While not up to the standards of the US, the EU contains several members whose militaries are quite sufficient to kick the arse of any invader on home turf, including the US, by the way. Not to mention the fact that the UK and France are nuclear powers. Now, if you were going to make the more subtle argument that the EU relies on the US's military muscle to secure its oil supplies, you might have a point. But the claim that Europe needs a massive remilitarization to protect against a conventional military threat is complete hogwash.
I understand that at in the automotive world, a very rough approximation used is that drag is proportional to v4. That calculation breaks down as you approach Mach 1, but I dunno what the situation is at supersonic speeds and the very low pressures at these very high altitudes (well beyond the range I thought you could lift a blimp to, to be honest).
My view is simple - the assumption that faster hardware will automatically give us human-level AI has little evidence to back it up. My arguments for this are as follows:
- The Church-Turing thesis says that new computers will be able to do the same things as old ones, just faster. I think there's an extension that says you'll only get, at best, a polynomial-time speedup (limited applications of quantum computers excepted).
- Complexity theory means that faster computers help even less than the layman thinks if it turns out that your problem doesn't have a linear-time algorithm. As I understand it, a lot of the neural-net algorithms have training times exponential in the number of neurons, in which case faster computers don't help much at all.
- Though we've made lots of progress on lots of specific problems that were once thought to be an instance of the AI problem, we're no closer to building a human-like AI than we were in 1960, despite the fact that the processing and storage available per inflation-adjusted dollar has increased maybe 100,000 fold since then.
- While, in principle, I agree that you could model the human brain on a computer, the fact is we don't yet know that much about how it works - or, more accurately, we have some idea about the individual neuron level, a little bit of an idea at the highest level of observable behaviour, but a huge yawning gap in the middle between those two bodies of research. The biologists have got a whole bunch of new tools to investigate the subject over the past couple of decades, but they don't seem to have made any radical progress so far - or, if they have, they haven't told the rest of us!
- It's been suggested, notably by Roger Penrose, that intelligence and consciousness is the result of funky non-computable quantum stuff happening in the brain. Now, there's very little evidence to support that being true, but it's not prudent to rule it out either.
In summary, it's possible that, once we have something to model, more computer power might help us model the important structures of the human brain, or an adaptation thereof. It's quite possible we have enough computer power to do it now. It's also quite possible that we'll never have enough computer power with a conventional computer to do it. And, finally, there's no guarantee that we'll ever figure out what's going on in the brain.My key point, however, is the claim that super-intelligent AI systems will be the inevitable result of faster computers was wrong when it was first made in the 1960's, was wrong when it was made again in the late 1970's in all manner of popular books, was wrong when Kurzweil made it in the 1990's, and it's still wrong now.
The electric cars of the early 90's were mostly about a California law requiring manufacturers to produce a certain number of zero emission vehicles to be able to sell their cars in Carlifornia. California wanted this to deal with smog, which is essentially a local pollution phenomenon, and one that EV's solve by shifting the pollution problem from cities to power plants. The manufacturers were never very keen on the idea, as they knew perfectly well that they couldn't build an EV with a competitive range. So, they built a succession of prototypes until the California legislature realized that physics wasn't on their side and until a fuel cell car (or hydrogen cheap enough to burn in a combustion engine) is along, a zero emission vehicle just isn't practical.
Are you familiar with the Church-Turing thesis? Basically, every alternative computing formulation we've been able to come up with computes exactly the same set of functions as conventional computers. Given that, could you please explain to me how any particular hardware breakthrough (modulo a hypercomputer of some kind) is going to give us human-equivalent AI?
Bad acting. Poor plots. Bad writing. A theme tune so excruciating that I want to vomit every time I hear it. And you're worried about "violating canon"?
In Victoria, Australia, one of the power companies is planning to do a similar thing with coal, except they're going to churn out enough of it to supply most of the local market. If it works, they're going to generate cheap, low-sulfur (and thus low-emission) diesel, run a whopping great electricity plant from the byproducts, and all the CO2 from the generation will be stuffed underground for a very long time. While it's not ideal, it's a heck of a lot better than the current situation (burn the coal straight into the atmosphere and import oil from overseas).
My initial post was in response to the wrong-headed idea that these small companies have achieved the same thing as the Chinese have done with their orbital flight, when the Chinese achievement of putting a person in orbit and bringing them back is much more challenging than a suborbital hop.
Might I point out that Lord Kenneth is a known miscreant who has pulled a number of stunts like this. Keep that in mind when evaluating the worth of his criticisms.
While building an X-Prize class of vehicle is an impressive achievement, it's a long, long way from putting somebody into orbit and retrieving them again. You require a hell of a lot more thrust to put somebody into orbit, the heat shielding requirements for the way down are much tougher, and you've got to be able to maintain life support for at least several hours rather than a few minutes.
I love the quote from the rover team leader. She not only took responsibility, she and her team managed to fix it! Now there's an idea more people in the federal government (particularly high up in the executive branch) might take some inspiration from... ;)