As a European, it I amazes me how many slashdotters seem to live in a complete state of denial, even going as far as to claim that "Trade Defcit" is an artificial and meaningless term - it is not! Unlike many other figures like the official GNP (which includes positions like "good will" and and other magic) or the official (hedonic) inflation rate, or the unemployment rate (determined by a telephone poll), the trade deficit is a very solid and easy to understand thing: the value of (real, tangible) goods and servicies going in minus the value of products going out.
If you buy more stuff than you sell, then you have a net flow of dollars out. Whoever receives these dollars can do one of 3 things: (1) use the dollars (directly or indirectly) to buy US products, (2) lend the dollars back to the US (US bonds, treasury bills), expecting to get even more dollars back, or (3) buy stuff in the US (stocks, equity, real estate), again in the hope to make even more dollars from his investment. Since, for one reason or another, the US economy fails to offer enough products anyone outside the US wants to buy (1) (hence the trade deficit), we're taking about debt (2) and sell-out (3).
A "normal" country wouldn't be able to do this for very long; as he would quickly (a) have its currecy devalued (if you have nothing to sell, no one outside needs your currency), (b) run out lenders (most countries dont have the luxury to indept itself abroad in their own currency to begin with) and (c) local assets to sell. Depending on whether you do or do not fire up the printing press you either end up in bancruptcy or hyperinflation.
A super power like the US which is able to strongarm or otherwise convince the producers of some key commodities to trade exclusively in their own currency, can get away with it considerably longer as their are alway products to buy for dollars (most prominently oil) even if those are not made in the USA by US labour. This gives the dollar the liquidity which is necessary so that it even makes sense to lend back dollars to the US (instead of demanding real goods right away).
It is even more helpful, if you can convice some producers of said key commodities (as e.g. Saudi Arabia), to set a good example by putting most of the money into US treasury bills and US assets (petro dollar recycling), as this postpones inflation and the devaluation of the dollar which in turn allows to keep the interest rates in a sane range as long as those bills are kept in the vault and are constantly renewed.
Of course, not even the US can live on credit alone, eventually the lenders will want something "tangible", and if you cannot or don't want to provide goods, at least you can provide them "assets". Of course, there is only so much you can sell, so the trick is to invent more things to sell (e.g. "intellecual property") and to inflate the prices of existing investments. Since the inflation is in assets and not in consumer goods, this is not called inflation but a "boom" (after all, inverstors want prices to rise) before and a "bubble" after people realize that a high price dose not imply a sound and profitable investment (as happend in the dot-com crash and is happening right now in the real estate and mortage sector).
You can, as some in this thread do, consider this a good thing for the US, after all, it basically means converting paper into Mercedes' and BMWs, and they would be correct if this games can be played infinitely - if you happen to be in the group that actually profits (that is, if you work in or close to the finance sector - if you're a production worker, well, too bad - maybe you can get a job as buttler or bodguard by those with better luck). Then it would make sense to deindutrialize your country and neither do you need an efficient education system as the rest of the world is supplying you with consumer goods basically for free.
It won't work infinitely, though. The non-US world sits on an ever increasing pile of trillions of dollars in US-dept and overvalued assets. Ever
As far as I understand, it would have been unfair if they would have denied them the logs because they were available to the AI team as a consequence of the setup as any pair of hands against the human competitors involved exactly the same cards, but in one game the AI and in the other the human opponent would get the winning hand. So the computer, having played each hand from both sides, can reconstruct all hole cards from the logs. It's only fair to give the humans the same possibility (leaving aside the fact that an AI is supposed to profit more from this information as it it usually worse at "filling the blanks" from intution). Doing otherwise would have led to the same rigged competition as in the Kasparov-Deep Blue match where the AI got fed thousends of Kasparov's games yet all training matches of the AI were kept secret.
> > Linux is the fastest growing unix-ish operating system out there.
> If you mean in terms of raw sales, it's OsX, not Linux.
Yeah right, raw sales is surely a good measure for a free OS.
> If you mean in terms of percentage userbase growth, it's QNX, not Linux.
Percentage growth is even more meaningless. A system going from zero to one user has thus infinite grow??
Meaningful measures are e.g. marketshare, total number of users or total number of installed systems.
> Your GPL is costing you tremendously.
No it doesn't. Any GPL project can incorporate BSD code, but not the other way around.
> with GPL, it's illegal for me to refuse even to release one line of source.
Only when you plan on releasing the binaries and that's really the point of the GPL.
> I can't release the object because of my Nintendo NDA, and I can't refuse to release the object because then I'm in violation of the GPL.
It obviously didn't occur to you that it might be the Nintendo NDA which is bogus and broken. NDAs for APIs are just plain stupid, so instead of ranting here against the GPL, simply refuse to develop for a plattform with such a hostile policy. If you don't (be it by choice or by necessity), you reinforce such policies and thus demonstrate the need for the GPL.
> But I can't use them, because GPL is so ridiculously paranoid.
As long as there are companies out there which demand NDAs for APIs, sue customers for reverse engineering or amass software patent porfolios, the question is really: Is it paranoid enough?
> You're looking at hundreds of thousands [of Gs].
No, it's actually two orders of magnitude below that for reasonably sized barrels. For constant acceleration, the formular is a=0.5*v0^2/l. Assuming v0=8000 m/s and a 1 km barrel, you end up with a=32000 m/s2 (about 3300 g). So you end up in the 4-figures. Still a lot, but quite managable and in line with heavy artillery shells. Esp. electronics has no problem to cope with this kind of acceleration (hardend equipment can easily deal with 10 times more).
For the acceleration issue: Sure, a space gun is not an all-purpose transport, but neither does it have to be. Even today, most of the mass we lauch to space, at the point the spacecraft reaches the first cosmic speed is fuel, esp. when the intended orbit or trajectory is higher than LEO.
If the goal is to establish a permanent space presence and bootstrapping a space economy, then it's safe to assume that at least initially more than 90% of the cargo will be fuel and reaction mass, with oxygen and water being a good part of the rest, until we are able to get this stuff from astroids. I guess you should also be able to lauch solar pannels and comparable equipment, although maybe not in their final configuration.
IMO, what we eventually need is not one but 3 lauch systems: A space gun for fuel and other bulk cargo (which is mass-wise the biggest fraction), a big dumb bootster to dry-lauch larger structures and technical equipment which cannot handle the 1000+ g acceleration, and some sort of man-rated passenger-only shuttle for the crews.
As for the other issues you mention: Sure, the atmospheric losses will be higher than with rockets (IIRC one of the sites mentioned 25%). OTOH, you have no gravitational losses as you are ballistic from the start. Of course, any vessel will require a heat shield. But those can be cheap heavy expendible ablative shields as, unlike a rocket, you don't have to pack many time its weight in fuel to get it up there and you will need a solid casing anyway to survive the lauch.
Building a workable space gun is surely a challenge, but much practical research has already been done (including stuff like g-hardend electronics and navigational equipment) and no major breakthroughs are required. The greatest disadvantage, besides the initial acceleration is probably that it doesn't provide much flexibility for choosing orbits (basically all you can vary is v_0) which makes it less useful for the LEO satellite business which currently dominates the market.
I know it's slightly offtopic, but I always wonder why a highly speculative and fragile concept like the space elevator which is barely theoretically possible is getting so much press, while space guns, which are cheaper, more robust and don't require any new technology, are practically ignored.
In case you're not familiar with the concept: It's basically about accelerating a small vessel (by a light gas gun, a RAM accelerator, electromagnetically or a combination thereof) in a relatively short (about the order of one km) barrel / tunnel to about orbital speed. The vessel itself will only require enough fuel for circularizing its orbit, so unlike conventional boosters, a much bigger part of its mass can be actual payload as the exponential regime of the rocket equation can be mostly avoided.
While the capital costs will be high, a space gun is still dirt cheap compared to a space elevator, and isn't prone to be completely destroyed when hit by lightning, space debris or, for the matter, a shotgun.
Let the patent holders pay for their state granted monopoly they enjoy. When someone files a patent, he should state how much it is worth for him, and then pay a certain percentage of this (say 2%) per year to keep his claim. Whenever someone pays up the sum (adjusted for the remaining period, so 70% after 3 years for a 10-year patent), the patent goes into the public domain. The sum can be adjusted yearly by the holder in a +/- 10% range.
So any patent would come with a price tag and it would cost something to hold them. No more frivolous patents and no more 50000+ portfolios. Only patents that are used as intended i.e. to protect your own product or to sell licences would be worth holding.
So I overestimated the size of the earth slightly. I thought that at about 4-5 times the distance earth-moon, the earth would still be big enough to shade the point.
Thanks for your answer, and also to all others who responded.
Why does this need a sunshield at all? The article says that the telescope should be parked in the 2nd Lagrangian point L2, which is 1.5 Gm from the Earth and should be permanently shaded from sunlight. Isn't the whole point of sending something to L2 that it is not exposed to the sun? Also, how is the energy supply supposed to work? Anyone out there who can shed some light on these questions?
IANAL, but as far as I know, this is the case almost anywhere in Europe. At least here in Austria AFAIK, in a civil court, the loser pays the costs of the trial as well the attorney fees for the other party. All fees are tightly regulated and generally depend on the amount in dispute, so you cannot be ruined by legal fees alone in small cases even when you lose. Also, practices like "success premiums" (where e.g. the plaintif's lawyer would get some share of the amends or settlement sums) are illegal.
... but here in Austria you can order Dell Workstations with Linux (RedHat) preinstalled. Also, about a year ago, I ordered a Dell Precision 380 workstation without a preinstalled OS (It came with a FreeDos partition containing drivers and docs IIRC). YMMV
Linux customers could care less about the pre-installed distrib, but the do care about 100% Linux compatability of the hardware, which is pretty much the same over all distribs (modulo non-free drivers). After all, even for large roll-outs, "installing" a customized system on identical hardware simply means gunzipping a prepared disk image (which can also include the partition table).
Therefore the best way to go about it would probably be to merely install a minimum system with a small footprint (1 GB max) but all hardware drivers installed and configuered in order to demonstrate Linux compatibility and to allow to check the hardware. The distrib should not matter in this case. Then, the customer can install his favorite Linux distrib and opt to keep the minimum installation as a rescue system.
In the case of Dell, this means: Replace the existing FreeDos installation (which you get when you order a Dell w/o OS - at least here in Europe) with a small Linux system, and everyone is happy.
If anything, the demand for CS experts will increase for a simple reason: The free lunch provided by Moores Law, which allowed for sloopy programming, inefficent algorithms and the use of wasteful high level concepts and delevlopment tools is over for several years now. No more can software engineers rely on the fact that the increase in (single core) CPU speed will make their inefficient software perform OK when the product finally hits the market.
Now as the main increase in CPU power comes from multiple cores rather than more GHz, and multicore CPUs are the norm even on Stadard PCs, further improvments have to be done the hard way: by more efficient code and parallelization. And to quote Linus Torvalds:
And anybody who tells you that distributed algorithms are "simpler" is just so full of sh*t that it's not even funny. So no need to get your Taxi licence just yet...
I would fancy that, evaporisation of water at the body surface in the vacuum of space would also substantially contribute to net heat loss, at least until the skin is completely dried up. (OTOH, those Himalaya climbers wear no pressure suits, either...)
You would have to lose only about 1kg of Water per hour to match the 640W of radiation loss you come up with.
Free will is not about desires, it's about decisions. Desires are a matter or perception: It's entirely possible to have desires without the possibility to act upon them. Free will is a different beast: While obviously you need individuality and perception to exercise free will, the rules of the physical universe also need to provide a backdoor for the influence to enter. As you corretly pointed out, determinism would deny the possibility of such a backdoor and would consequently render us passive observers.
> Free will couldn't be described as a random event, it is a consciously determined event.
This is not contradictory. It can BE consciously determined, and still APPEAR as randomness in the context of physical theory. Mental phenomena cannot be subjected to the scientific method as introspection lacks the necessary intersubjetivity and reproducability. Consequently, they would fall victim to Occams razor in any physical description of the world. This does not mean that they don't exist, just that they cannont be incorporated into the physical theory: The physics would be the same, whether consciousness exists or not.
With free will, the influence goes the other direction, so in a physical theory, it would appear as effects without deterministic cause - the very definition of randomness. This does not mean that ANY observed wavefunction collapses is attributable to the exercise of free will, very much the same as not anything that happens in the universe is consciously percepted. For as far as we know, for humans, the "interface area" - i.e. the biological substrate - can be confined to a few square inches in the skull.
Depending on how the "interface" works on the quantum level - and we know basically nothing about this, the probabiliy distribution can or cannot deviate from what quantum theory predicts. If the mechanism is subtle enough, there would be no detectable differences, just all e.g. virtually all kinds of limited small effects, given in large enought numbers will accumulate into a gaussian distribution.
> Randomness is not a prerequisite for free will, but is rather incompatible with it
Randomness is a mathematical artefact which happens to lend itself nicely to formal treatment. Mathematically (and physically), there is no difference between "true" randomness and some external effect which happens to come up with the same or a sufficiently similar probabiliy spectrum. (This is not to be mixed up with hidden variables, as mental effects are bound to individual consciousnesses and not some physical enities.)
It's a misguided thought to think that neurobiology can help anything to settle the question of free will. Mental experiences (in this case the desire to abuse children) require a biological substrate (in this case involving a tumor) - this is not exactly a new thought. We always knew that vision (a mental experience) requires eyes (a biological substrate). Neuroscience will tell us that it also requires a few other things like nerves and certain structures in the brain - nice to know, but nothing qualitatively new. Drugs (a physical substance) can dampen, amplify or create desires (a mental phenomenon) - to know how the mechanisms involved in addiction work in detail is of practical value, but yields no philosophical insight.
If, beyond the very convincing, however necessarily subjective evidence given by introspection, we were to look for scientific evidence of free will, we should rather turn to physics: As a physical phenomenon, free will would show up as an effect without a cause WITHIN THE SYSTEM, i.e. the intersubjectiv, physically observable universe. Or, with other words, as a random event. The existence of genuine randomness (e.g. in radioactive decay, but basically in any form of quantum measurement) in the observable universe is pretty much a settled fact in the physical community since the thirties of the previous century. Alas, philosophy (and psychology, for the matter) is, as usual, about a century behind, and still trapped in Newtons mechanistic and deterministic worldview.
Don't get me wrong - of course, the existence of randomness does not PROVE the existence of free will - it's only a necessary requirement (in a less strict sense - for all practical purposes, so to say - deterministic chaos or simply intractability would also suffice). But here, Occam's razor kicks in: Perception (such as the fundamental perception of my own existence as a single individual) is an immaterial phenomenon (albeit with a physical substrate). Introspection shows me to have free will, likewise an immaterial phenomenon. The known rules of the intersubjective universe, as established by physics, allow for observable phenomena without a deterministic cause (quantum measurement), so they are compatible with the idea of free will. The actual existence of free will is the simplest explaination which accounts for all of the above. The concept of free will is no more absurd than the idea of individual perception, just the direction of the influence is not from the physical "outside" to the mental "inside", but the other way around (with the additional benefit that it could therefore be disproved if we found our observable universe to be deterministic, after all).
Of course, there are people who deny both, but firstly I doubt that their mechanistic explaination of how the bunch of atoms that they think they are manages to develop the "illusions" of consciousness, individuality, perception and willful behaviour is much simpler. And secondly, with mental phenomena, the illusion IS just the same as the real thing.
Assuming a limited amount of storage for your adversary is a much more optimistic assumption than the mathematical assumptions underlying conventional encryption schemes and not comparable to the unconditional security provided by quantum crypto. It basically relies on the fact that the rate of both, the entropy source as well as the communication channel times your maximum communication latency is lower than the capacity of any feasable storage device. I doubt that this conditions can be reliably met, at least in the realm of fiber-bound communication (certainly not, if you want it to be cheaper than current quantum schemes).
As for the novelty of quantum crypto: Of course, there is not much practical experience, but this is the same with any new classical encryption algorithm. The problem is also easily avoided: Simply sandwich the quantum crypto between two layers of conventional cryptography, all with their own authentication and key-setup. Nobody is suggesting to completly replace classical crypto by quantum, after all. - It's just another brick in the wall of your overall security system (and the rest of the wall, esp. the human portion, is usually the expensive part).
I work at a research company which designs, among other things, quantum crypto hardware. One important aspect of quantum crypto is that, once the channel is authenticated, the communication (more more precisely: the establishment of the key) cannot be wiretapped and stored away for later decryption. It's basically a method to strech a pre-shared secret (necessary for authentication) almost arbitraily, which can then be used as one-time-pad to encrypt the subsequent classical communication, rendering it unconditionally secure.
This makes sense if you need your secrets to last for a very long time in the face of a determined and resourceful attacker, outlasting potential breakthroughs in mathematics or quantum computing. Of course, only few customers actually need this level of security. However, if the secrets you are protecting are worth several billions, the cost for quantum hardware can be negligable, esp. when compared with all the other security measures you will have to take to make up a secure overall system.
Sea levels are actually an excellent example of slow adaption at work: Coast land is created and eroded/flooded all the time. This usually does not pose much of a problem, as the change is much slower than the typical lifetime of a building. Every city is in a constant process of being rebuilt. If the ocean level rises, houses on lower ground will tend to be replaced by buildings on higher ground.
If some land is going to become unhabitable during the next decades, it's not that people will flee in panic: They will simply tend to settle in different areas as they always do - and on much shorter timescales - for economic reasons. Even if the ocean level will rise 10m during the next 100 years, the resulting migration - while huge in total - will still be negligable when compared to the ongoing migration caused by unempolyment or poverty or such mundane reasons as couples moving together.
The same thing will happen when fossil fuels supply slowly declines: It's no problem to get along with half the current energy consumption or less without having to significantly lower your standard of living. This is especially true for the US, as this would merely mean that an American citizen will have to make do with what an average European currently gets along quite well (and often with a higher standard of living).
Similar picture with the food supply, even though it is doubtful that global warming will have a significant negative net-effect on the global agricultural output (the regional effects might as well cancel out for the most part): There are huge reserves in terms of overproduction, unused land, inefficient distribution and processing (e.g. one calorie of meat equals seven calories of crops), which could easily feed twice the current world population NOW, if we decide to go about it, again with only moderate changes to our current lifestyle (which might, in fact, be for the better in many Western countries).
The bottom line of all this is: yes, changes will come along, like they always have. However, instead of worring how to PREVENT change (which might be futile or at least very hard), it seems more wise to me to think how we can ADAPT to the change in due course (with the added benefit that this will also help with non-anthropogenic changes). And of course, it's already happening. At the end of the day, many measures proposed by the "safe the climate" crowd happen to be the very same measures will will come along anyway, albait less hysterically and with less fanfare, as people adapt to the decreasing supply in fossil fuels.
What we do NOT need, is a fatalistic doomsday-attitude or extreme govermental intervention to pressure developments which will happen in due curse, anyway. A resonable Oil-tax (which we already have in Europe), some relocation of long-term infrastructure spending away from regions in danger of being flooded (because they are under sealevel now) and other similar moderate measures are no doubt reasonable to preempt some of the economic pressure, but the real change will be brought about by higher oil and energy prices all by itself.
So climate changes? Big deal. It always has and always will, locally and gobally, with or without human intervention. It does not matter a bit whether the next climate change will be caused by human activity or not.
The current peak in CO2 emissions will decline all by itself when the coal and hydrocarbon deposits slowly run out. As those reserves are, in historic timescales, basically fixed, so is the total amount of CO2 that will eventually be released back into the biosphere, and it does not really matter if this happens in 200, 500 or 1000 years. Will will have to deal with the effects eventually, both the economic (the end of the fossil fuel era) as well as any climatic ones (in addition to any climatic changes which come about for unrelated reasons).
And guess what? That's exactly what we will do! If the oil supply stopped overnight, it would be the end of the world as we know it. If global sealevels rose five or ten meters within a week, it would be a global catastrophy. If the same things happen in the course of a decade, it will be a huge crisis, but civilisation will survive. If it happens over the course of a century (in line with the most pessimistic scenarios), mankind will face huge changes, of course, but to the average person, IMO, those changes will be LESS noticable then the major conflicts and revolutions of the 20th century. Think about how life has changed for the average European or American in the last hundred of years.
Adopting to slowly changing circumstances is something that we, as humans, are really good at. It's basically our second nature. We are so good at dealing with these slow revolutions, that most of us don't notice them in their everyday life.
If you want more scientists, pay them more! Once an average scientist makes more than your typical lawyer, doctor or business executive, the social status of science will increase accordingly and a higher proportion of the most talented will pursuit a scientific career.
W/$ is only a viable metric for short term installations when lifetime considerations do not enter the picture. A precise assessment would also include projected interest rates and the degradation curve of the cell as well as costs for maintenance (cleaning, etc.) and disposal/recycling (toxic materials, etc.).
Slashdot Mirror
As a European, it I amazes me how many slashdotters seem to live in a complete state of denial, even going as far as to claim that "Trade Defcit" is an artificial and meaningless term - it is not! Unlike many other figures like the official GNP (which includes positions like "good will" and and other magic) or the official (hedonic) inflation rate, or the unemployment rate (determined by a telephone poll), the trade deficit is a very solid and easy to understand thing: the value of (real, tangible) goods and servicies going in minus the value of products going out.
If you buy more stuff than you sell, then you have a net flow of dollars out. Whoever receives these dollars can do one of 3 things: (1) use the dollars (directly or indirectly) to buy US products, (2) lend the dollars back to the US (US bonds, treasury bills), expecting to get even more dollars back, or (3) buy stuff in the US (stocks, equity, real estate), again in the hope to make even more dollars from his investment. Since, for one reason or another, the US economy fails to offer enough products anyone outside the US wants to buy (1) (hence the trade deficit), we're taking about debt (2) and sell-out (3).
A "normal" country wouldn't be able to do this for very long; as he would quickly (a) have its currecy devalued (if you have nothing to sell, no one outside needs your currency), (b) run out lenders (most countries dont have the luxury to indept itself abroad in their own currency to begin with) and (c) local assets to sell. Depending on whether you do or do not fire up the printing press you either end up in bancruptcy or hyperinflation.
A super power like the US which is able to strongarm or otherwise convince the producers of some key commodities to trade exclusively in their own currency, can get away with it considerably longer as their are alway products to buy for dollars (most prominently oil) even if those are not made in the USA by US labour. This gives the dollar the liquidity which is necessary so that it even makes sense to lend back dollars to the US (instead of demanding real goods right away).
It is even more helpful, if you can convice some producers of said key commodities (as e.g. Saudi Arabia), to set a good example by putting most of the money into US treasury bills and US assets (petro dollar recycling), as this postpones inflation and the devaluation of the dollar which in turn allows to keep the interest rates in a sane range as long as those bills are kept in the vault and are constantly renewed.
Of course, not even the US can live on credit alone, eventually the lenders will want something "tangible", and if you cannot or don't want to provide goods, at least you can provide them "assets". Of course, there is only so much you can sell, so the trick is to invent more things to sell (e.g. "intellecual property") and to inflate the prices of existing investments. Since the inflation is in assets and not in consumer goods, this is not called inflation but a "boom" (after all, inverstors want prices to rise) before and a "bubble" after people realize that a high price dose not imply a sound and profitable investment (as happend in the dot-com crash and is happening right now in the real estate and mortage sector).
You can, as some in this thread do, consider this a good thing for the US, after all, it basically means converting paper into Mercedes' and BMWs, and they would be correct if this games can be played infinitely - if you happen to be in the group that actually profits (that is, if you work in or close to the finance sector - if you're a production worker, well, too bad - maybe you can get a job as buttler or bodguard by those with better luck). Then it would make sense to deindutrialize your country and neither do you need an efficient education system as the rest of the world is supplying you with consumer goods basically for free.
It won't work infinitely, though. The non-US world sits on an ever increasing pile of trillions of dollars in US-dept and overvalued assets. Ever
As far as I understand, it would have been unfair if they would have denied them the logs because they were available to the AI team as a consequence of the setup as any pair of hands against the human competitors involved exactly the same cards, but in one game the AI and in the other the human opponent would get the winning hand. So the computer, having played each hand from both sides, can reconstruct all hole cards from the logs. It's only fair to give the humans the same possibility (leaving aside the fact that an AI is supposed to profit more from this information as it it usually worse at "filling the blanks" from intution). Doing otherwise would have led to the same rigged competition as in the Kasparov-Deep Blue match where the AI got fed thousends of Kasparov's games yet all training matches of the AI were kept secret.
ignatius
Any videoplayer with "democratic" in the title isn't.
ignatius
> > Linux is the fastest growing unix-ish operating system out there.
> If you mean in terms of raw sales, it's OsX, not Linux.
Yeah right, raw sales is surely a good measure for a free OS.
> If you mean in terms of percentage userbase growth, it's QNX, not Linux.
Percentage growth is even more meaningless. A system going from zero to one user has thus infinite grow??
Meaningful measures are e.g. marketshare, total number of users or total number of installed systems.
> Your GPL is costing you tremendously.
No it doesn't. Any GPL project can incorporate BSD code, but not the other way around.
> with GPL, it's illegal for me to refuse even to release one line of source.
Only when you plan on releasing the binaries and that's really the point of the GPL.
> I can't release the object because of my Nintendo NDA, and I can't refuse to release the object because then I'm in violation of the GPL.
It obviously didn't occur to you that it might be the Nintendo NDA which is bogus and broken. NDAs for APIs are just plain stupid, so instead of ranting here against the GPL, simply refuse to develop for a plattform with such a hostile policy. If you don't (be it by choice or by necessity), you reinforce such policies and thus demonstrate the need for the GPL.
> But I can't use them, because GPL is so ridiculously paranoid.
As long as there are companies out there which demand NDAs for APIs, sue customers for reverse engineering or amass software patent porfolios, the question is really: Is it paranoid enough?
> You're looking at hundreds of thousands [of Gs].
No, it's actually two orders of magnitude below that for reasonably sized barrels. For constant acceleration, the formular is a=0.5*v0^2/l. Assuming v0=8000 m/s and a 1 km barrel, you end up with a=32000 m/s2 (about 3300 g). So you end up in the 4-figures. Still a lot, but quite managable and in line with heavy artillery shells. Esp. electronics has no problem to cope with this kind of acceleration (hardend equipment can easily deal with 10 times more).
For the acceleration issue: Sure, a space gun is not an all-purpose transport, but neither does it have to be. Even today, most of the mass we lauch to space, at the point the spacecraft reaches the first cosmic speed is fuel, esp. when the intended orbit or trajectory is higher than LEO.
If the goal is to establish a permanent space presence and bootstrapping a space economy, then it's safe to assume that at least initially more than 90% of the cargo will be fuel and reaction mass, with oxygen and water being a good part of the rest, until we are able to get this stuff from astroids. I guess you should also be able to lauch solar pannels and comparable equipment, although maybe not in their final configuration.
IMO, what we eventually need is not one but 3 lauch systems: A space gun for fuel and other bulk cargo (which is mass-wise the biggest fraction), a big dumb bootster to dry-lauch larger structures and technical equipment which cannot handle the 1000+ g acceleration, and some sort of man-rated passenger-only shuttle for the crews.
As for the other issues you mention: Sure, the atmospheric losses will be higher than with rockets (IIRC one of the sites mentioned 25%). OTOH, you have no gravitational losses as you are ballistic from the start. Of course, any vessel will require a heat shield. But those can be cheap heavy expendible ablative shields as, unlike a rocket, you don't have to pack many time its weight in fuel to get it up there and you will need a solid casing anyway to survive the lauch.
Building a workable space gun is surely a challenge, but much practical research has already been done (including stuff like g-hardend electronics and navigational equipment) and no major breakthroughs are required. The greatest disadvantage, besides the initial acceleration is probably that it doesn't provide much flexibility for choosing orbits (basically all you can vary is v_0) which makes it less useful for the LEO satellite business which currently dominates the market.
I know it's slightly offtopic, but I always wonder why a highly speculative and fragile concept like the space elevator which is barely theoretically possible is getting so much press, while space guns, which are cheaper, more robust and don't require any new technology, are practically ignored.
h tm
In case you're not familiar with the concept: It's basically about accelerating a small vessel (by a light gas gun, a RAM accelerator, electromagnetically or a combination thereof) in a relatively short (about the order of one km) barrel / tunnel to about orbital speed. The vessel itself will only require enough fuel for circularizing its orbit, so unlike conventional boosters, a much bigger part of its mass can be actual payload as the exponential regime of the rocket equation can be mostly avoided.
While the capital costs will be high, a space gun is still dirt cheap compared to a space elevator, and isn't prone to be completely destroyed when hit by lightning, space debris or, for the matter, a shotgun.
http://en.wikipedia.org/wiki/Space_gun
http://www.fas.org/news/iraq/1998/05/980500-bull.
http://www.astronautix.com/lvs/julncher.htm
Let the patent holders pay for their state granted monopoly they enjoy. When someone files a patent, he should state how much it is worth for him, and then pay a certain percentage of this (say 2%) per year to keep his claim. Whenever someone pays up the sum (adjusted for the remaining period, so 70% after 3 years for a 10-year patent), the patent goes into the public domain. The sum can be adjusted yearly by the holder in a +/- 10% range.
So any patent would come with a price tag and it would cost something to hold them. No more frivolous patents and no more 50000+ portfolios. Only patents that are used as intended i.e. to protect your own product or to sell licences would be worth holding.
So I overestimated the size of the earth slightly. I thought that at about 4-5 times the distance earth-moon, the earth would still be big enough to shade the point.
Thanks for your answer, and also to all others who responded.
ignatius
Why does this need a sunshield at all? The article says that the telescope should be parked in the 2nd Lagrangian point L2, which is 1.5 Gm from the Earth and should be permanently shaded from sunlight. Isn't the whole point of sending something to L2 that it is not exposed to the sun? Also, how is the energy supply supposed to work? Anyone out there who can shed some light on these questions?
ignatius
IANAL, but as far as I know, this is the case almost anywhere in Europe. At least here in Austria AFAIK, in a civil court, the loser pays the costs of the trial as well the attorney fees for the other party. All fees are tightly regulated and generally depend on the amount in dispute, so you cannot be ruined by legal fees alone in small cases even when you lose. Also, practices like "success premiums" (where e.g. the plaintif's lawyer would get some share of the amends or settlement sums) are illegal.
... but here in Austria you can order Dell Workstations with Linux (RedHat) preinstalled. Also, about a year ago, I ordered a Dell Precision 380 workstation without a preinstalled OS (It came with a FreeDos partition containing drivers and docs IIRC). YMMV
thesis - antithesis - synthesis: the porn in the browser history and his girl friend are in fact the same thing.
Linux customers could care less about the pre-installed distrib, but the do care about 100% Linux compatability of the hardware, which is pretty much the same over all distribs (modulo non-free drivers). After all, even for large roll-outs, "installing" a customized system on identical hardware simply means gunzipping a prepared disk image (which can also include the partition table).
Therefore the best way to go about it would probably be to merely install a minimum system with a small footprint (1 GB max) but all hardware drivers installed and configuered in order to demonstrate Linux compatibility and to allow to check the hardware. The distrib should not matter in this case. Then, the customer can install his favorite Linux distrib and opt to keep the minimum installation as a rescue system.
In the case of Dell, this means: Replace the existing FreeDos installation (which you get when you order a Dell w/o OS - at least here in Europe) with a small Linux system, and everyone is happy.
Now as the main increase in CPU power comes from multiple cores rather than more GHz, and multicore CPUs are the norm even on Stadard PCs, further improvments have to be done the hard way: by more efficient code and parallelization. And to quote Linus Torvalds: And anybody who tells you that distributed algorithms are "simpler" is just so full of sh*t that it's not even funny. So no need to get your Taxi licence just yet
I would fancy that, evaporisation of water at the body surface in the vacuum of space would also substantially contribute to net heat loss, at least until the skin is completely dried up. (OTOH, those Himalaya climbers wear no pressure suits, either ...)
You would have to lose only about 1kg of Water per hour to match the 640W of radiation loss you come up with.
Free will is not about desires, it's about decisions. Desires are a matter or perception: It's entirely possible to have desires without the possibility to act upon them. Free will is a different beast: While obviously you need individuality and perception to exercise free will, the rules of the physical universe also need to provide a backdoor for the influence to enter. As you corretly pointed out, determinism would deny the possibility of such a backdoor and would consequently render us passive observers.
> Free will couldn't be described as a random event, it is a consciously determined event.
This is not contradictory. It can BE consciously determined, and still APPEAR as randomness in the context of physical theory. Mental phenomena cannot be subjected to the scientific method as introspection lacks the necessary intersubjetivity and reproducability. Consequently, they would fall victim to Occams razor in any physical description of the world. This does not mean that they don't exist, just that they cannont be incorporated into the physical theory: The physics would be the same, whether consciousness exists or not.
With free will, the influence goes the other direction, so in a physical theory, it would appear as effects without deterministic cause - the very definition of randomness. This does not mean that ANY observed wavefunction collapses is attributable to the exercise of free will, very much the same as not anything that happens in the universe is consciously percepted. For as far as we know, for humans, the "interface area" - i.e. the biological substrate - can be confined to a few square inches in the skull.
Depending on how the "interface" works on the quantum level - and we know basically nothing about this, the probabiliy distribution can or cannot deviate from what quantum theory predicts. If the mechanism is subtle enough, there would be no detectable differences, just all e.g. virtually all kinds of limited small effects, given in large enought numbers will accumulate into a gaussian distribution.
> Randomness is not a prerequisite for free will, but is rather incompatible with it
Randomness is a mathematical artefact which happens to lend itself nicely to formal treatment. Mathematically (and physically), there is no difference between "true" randomness and some external effect which happens to come up with the same or a sufficiently similar probabiliy spectrum. (This is not to be mixed up with hidden variables, as mental effects are bound to individual consciousnesses and not some physical enities.)
It's a misguided thought to think that neurobiology can help anything to settle the question of free will. Mental experiences (in this case the desire to abuse children) require a biological substrate (in this case involving a tumor) - this is not exactly a new thought. We always knew that vision (a mental experience) requires eyes (a biological substrate). Neuroscience will tell us that it also requires a few other things like nerves and certain structures in the brain - nice to know, but nothing qualitatively new. Drugs (a physical substance) can dampen, amplify or create desires (a mental phenomenon) - to know how the mechanisms involved in addiction work in detail is of practical value, but yields no philosophical insight.
If, beyond the very convincing, however necessarily subjective evidence given by introspection, we were to look for scientific evidence of free will, we should rather turn to physics: As a physical phenomenon, free will would show up as an effect without a cause WITHIN THE SYSTEM, i.e. the intersubjectiv, physically observable universe. Or, with other words, as a random event. The existence of genuine randomness (e.g. in radioactive decay, but basically in any form of quantum measurement) in the observable universe is pretty much a settled fact in the physical community since the thirties of the previous century. Alas, philosophy (and psychology, for the matter) is, as usual, about a century behind, and still trapped in Newtons mechanistic and deterministic worldview.
Don't get me wrong - of course, the existence of randomness does not PROVE the existence of free will - it's only a necessary requirement (in a less strict sense - for all practical purposes, so to say - deterministic chaos or simply intractability would also suffice). But here, Occam's razor kicks in: Perception (such as the fundamental perception of my own existence as a single individual) is an immaterial phenomenon (albeit with a physical substrate). Introspection shows me to have free will, likewise an immaterial phenomenon. The known rules of the intersubjective universe, as established by physics, allow for observable phenomena without a deterministic cause (quantum measurement), so they are compatible with the idea of free will. The actual existence of free will is the simplest explaination which accounts for all of the above. The concept of free will is no more absurd than the idea of individual perception, just the direction of the influence is not from the physical "outside" to the mental "inside", but the other way around (with the additional benefit that it could therefore be disproved if we found our observable universe to be deterministic, after all).
Of course, there are people who deny both, but firstly I doubt that their mechanistic explaination of how the bunch of atoms that they think they are manages to develop the "illusions" of consciousness, individuality, perception and willful behaviour is much simpler. And secondly, with mental phenomena, the illusion IS just the same as the real thing.
from http://en.wikipedia.org/wiki/Hyper-encryption:
"hyper-encryption can be proved to be information-theoretically secure, providing the storage bound cannot be surpassed."
Assuming a limited amount of storage for your adversary is a much more optimistic assumption than the mathematical assumptions underlying conventional encryption schemes and not comparable to the unconditional security provided by quantum crypto. It basically relies on the fact that the rate of both, the entropy source as well as the communication channel times your maximum communication latency is lower than the capacity of any feasable storage device. I doubt that this conditions can be reliably met, at least in the realm of fiber-bound communication (certainly not, if you want it to be cheaper than current quantum schemes).
As for the novelty of quantum crypto: Of course, there is not much practical experience, but this is the same with any new classical encryption algorithm. The problem is also easily avoided: Simply sandwich the quantum crypto between two layers of conventional cryptography, all with their own authentication and key-setup. Nobody is suggesting to completly replace classical crypto by quantum, after all. - It's just another brick in the wall of your overall security system (and the rest of the wall, esp. the human portion, is usually the expensive part).
I work at a research company which designs, among other things, quantum crypto hardware. One important aspect of quantum crypto is that, once the channel is authenticated, the communication (more more precisely: the establishment of the key) cannot be wiretapped and stored away for later decryption. It's basically a method to strech a pre-shared secret (necessary for authentication) almost arbitraily, which can then be used as one-time-pad to encrypt the subsequent classical communication, rendering it unconditionally secure.
This makes sense if you need your secrets to last for a very long time in the face of a determined and resourceful attacker, outlasting potential breakthroughs in mathematics or quantum computing. Of course, only few customers actually need this level of security. However, if the secrets you are protecting are worth several billions, the cost for quantum hardware can be negligable, esp. when compared with all the other security measures you will have to take to make up a secure overall system.
Sea levels are actually an excellent example of slow adaption at work: Coast land is created and eroded/flooded all the time. This usually does not pose much of a problem, as the change is much slower than the typical lifetime of a building. Every city is in a constant process of being rebuilt. If the ocean level rises, houses on lower ground will tend to be replaced by buildings on higher ground.
If some land is going to become unhabitable during the next decades, it's not that people will flee in panic: They will simply tend to settle in different areas as they always do - and on much shorter timescales - for economic reasons. Even if the ocean level will rise 10m during the next 100 years, the resulting migration - while huge in total - will still be negligable when compared to the ongoing migration caused by unempolyment or poverty or such mundane reasons as couples moving together.
The same thing will happen when fossil fuels supply slowly declines: It's no problem to get along with half the current energy consumption or less without having to significantly lower your standard of living. This is especially true for the US, as this would merely mean that an American citizen will have to make do with what an average European currently gets along quite well (and often with a higher standard of living).
Similar picture with the food supply, even though it is doubtful that global warming will have a significant negative net-effect on the global agricultural output (the regional effects might as well cancel out for the most part): There are huge reserves in terms of overproduction, unused land, inefficient distribution and processing (e.g. one calorie of meat equals seven calories of crops), which could easily feed twice the current world population NOW, if we decide to go about it, again with only moderate changes to our current lifestyle (which might, in fact, be for the better in many Western countries).
The bottom line of all this is: yes, changes will come along, like they always have. However, instead of worring how to PREVENT change (which might be futile or at least very hard), it seems more wise to me to think how we can ADAPT to the change in due course (with the added benefit that this will also help with non-anthropogenic changes). And of course, it's already happening. At the end of the day, many measures proposed by the "safe the climate" crowd happen to be the very same measures will will come along anyway, albait less hysterically and with less fanfare, as people adapt to the decreasing supply in fossil fuels.
What we do NOT need, is a fatalistic doomsday-attitude or extreme govermental intervention to pressure developments which will happen in due curse, anyway. A resonable Oil-tax (which we already have in Europe), some relocation of long-term infrastructure spending away from regions in danger of being flooded (because they are under sealevel now) and other similar moderate measures are no doubt reasonable to preempt some of the economic pressure, but the real change will be brought about by higher oil and energy prices all by itself.
So climate changes? Big deal. It always has and always will, locally and gobally, with or without human intervention. It does not matter a bit whether the next climate change will be caused by human activity or not.
The current peak in CO2 emissions will decline all by itself when the coal and hydrocarbon deposits slowly run out. As those reserves are, in historic timescales, basically fixed, so is the total amount of CO2 that will eventually be released back into the biosphere, and it does not really matter if this happens in 200, 500 or 1000 years. Will will have to deal with the effects eventually, both the economic (the end of the fossil fuel era) as well as any climatic ones (in addition to any climatic changes which come about for unrelated reasons).
And guess what? That's exactly what we will do! If the oil supply stopped overnight, it would be the end of the world as we know it. If global sealevels rose five or ten meters within a week, it would be a global catastrophy. If the same things happen in the course of a decade, it will be a huge crisis, but civilisation will survive. If it happens over the course of a century (in line with the most pessimistic scenarios), mankind will face huge changes, of course, but to the average person, IMO, those changes will be LESS noticable then the major conflicts and revolutions of the 20th century. Think about how life has changed for the average European or American in the last hundred of years.
Adopting to slowly changing circumstances is something that we, as humans, are really good at. It's basically our second nature. We are so good at dealing with these slow revolutions, that most of us don't notice them in their everyday life.
If you want more scientists, pay them more! Once an average scientist makes more than your typical lawyer, doctor or business executive, the social status of science will increase accordingly and a higher proportion of the most talented will pursuit a scientific career.
n ce
A good writeup on the situation (hooked on the topic of women in science) can be found here: http://philip.greenspun.com/careers/women-in-scie
The relevant efficiency metric depends completely on the application, e.g.
space, aircraft: W/kg
compact portable devices, vehicles, boats: W/m^2
rural housing, large scale grid power: kWh/$ (over lifetime)
W/$ is only a viable metric for short term installations when lifetime considerations do not enter the picture. A precise assessment would also include projected interest rates and the degradation curve of the cell as well as costs for maintenance (cleaning, etc.) and disposal/recycling (toxic materials, etc.).