There is more than just Fujitsu supporting PostgreSQL and the reasons there is corporate interest is pure unadulterated self-interest of the best kind.
Postgres is getting really close to the functionality and capabilities of the Big Commercial Enterprise DBMS, close enough that anyone can see that bridging that gap is quite doable. Most of the arguable weaknesses in Postgres are in the more esoteric high-end feature space, as it is already strong and quite feature complete for most routine RDBMS work. And the upcoming new version addresses a great many of those weaknesses. As the article said, this is going to be a major release.
The self-interest part is that it is a HELL OF A LOT CHEAPER for a corporation to pay people to add those last few features and bits that they want to Postgres than to pay an unholy amount of money to buy the required Oracle licenses. The Postgres engine is clean and fundamentally pretty good in an engineering sense, and so enterprise feature tweaks are relatively cheap. It is all about dollars and sense at the end of the day. Purchasing Postgres plus feature development is almost always going to be vastly cheaper than buying Oracle. And unlike Oracle, it is pretty much a one-time fixed cost. It is worth repeating that the engineering strength and scalability of the underlying Postgres platform is the primary reason the market is evolving this way. The gap between MySQL and high-end RDBMS is comparatively much too great for a company to fund closing that gap because a lot of additional arguably unrelated work may be required because of the internals. This increases time to delivery of features, increases the cost of adding high-end features, and increases the risk of problems.
If Oracle suddenly dropped its enterprise licensing costs by a couple of order of magnitude, then it would seriously threaten Postgres development. But since that is unlikely to happen, corporate money will continue to flow into making Postgres a formidable Oracle replacement, which it is already well on its way to being.
Re:Low technology against high technology
on
Japanese Balloon Battle
·
· Score: 2, Informative
The AK-47 has the advantage here. Its bullet has more of a change to travel through cover then the M16 had. Simply put an M16 soldier shooting at an AK47 soldier behind wood cover would be very likely to only hit wood. The AK47 soldier on the other hand would shoot straight through the wood and kill the M16 soldier or at least wound.
Oh hell, mod this guy down. This is factually dead wrong and they apparently know exactly nothing about terminal ballistics and its relation to cartridge design.
The AK47 has lousy penetration and short effective range due to the cartridge selection. The very low sectional density combined with relatively low velocity of the 7.62x39 makes it one of the poorest penetration performers of any military cartridge in common use today. The M16 cartridges that have been in use for the last decade or two (M855) will actually out-penetrate most everything else, particularly through hard materials, including the 7.62x51 (.308) NATO cartridge.
Sectional density * velocity = penetration, with some bonus points for bullet design. The 7.62x39 has neither sectional density nor velocity, the 5.56x45 has moderate sectional density and high velocity.
I can't believe the drivel that is being modded as informative. It is a big steaming pile of urban myth nonsense.
The AK-47 worked much better in real combat, because it almost never jammed, and was quite easy to use, which is good if you don't have time to train soldiers.
There is a huge amount of hearsay and legend surrounding the AK47 and M16, usually propagated by people without the experience or context to have an informed opinion.
For a well-trained soldier, the M16 system is vastly superior to the AK47 in just about every way. And the US has "well-trained" soldiers. In a more general sense, the AR15/M16 system is one of the finest rifles of its class ever fielded and has numerous advantages that aren't immediately obvious unless you've had to do extensive live fire combat training with both. Beside being a very nice rifle to shoot and handle, the M16 has the distinction of being one of the quickest combat rifles out there, while the AK47 is one of the slowest. By "quick" I mean how fast you can you drop some number of multiple random targets at random ranges. The M16 is blindingly quick for fast and furious engagements, the AK47 is not. After using an M16, it feels like you are trying to shoot using something with the handling and aiming characteristics of a cinderblock. In a firefight that is all about you shooting the other guy before he shoots you, this matters a lot.
As for jamming, that is largely mythology. There was a problem in the 1960s where some Army supply person decided to ship ammo that was out-of-spec for the engineering of the rifle because it was cheaper without checking with the designers. That caused a lot of problems. The rifle was quickly redesigned to be more tolerant of out-of-spec ammo. As a result, jamming hasn't been a problem for about 30 years. In the field where the dirt is, rule of thumb is that the typical modern M16 will function flawlessly for about 1500 rounds without cleaning. That is a LOT of ammo, and several times the max load that an infantry soldier normally carries.
But the uranium dumped in the US is usually buried underground, and not close enough to population centers to do any damage.
Errm... no its not. It isn't even a controlled waste. It is simply dumped into the environment. I'm not talking about nuclear power plant waste, which isn't anywhere near thousands of metric tons anyway. I'm talking about radioisotope concentrators like coal ash, which measures substantially above background. And it contains thousands of tons of isotopes that are far nastier than uranium e.g. thorium.
Incidentally, studies have been done on the populations that live near coal slag dumps and similar, which have high concentrations of uranium, thorium, and other radioisotopes. There is a small statistical bump in cancer, most of which is typically attributed to the thorium component.
places where the US have used DU shells are now contaminated with long-lasting radioactivity, and vehicles attacked with these shells show very high levels of radioactivity.... these things are monsters and unnessesary, I've worked with people trying to stop their use, but there seems to be a deaf ear listening.
The deaf ear may have something to do with you being illiterate on the basic facts of the issue.
First, depleted uranium is "depleted", meaning that the active and interesting isotopes have been largely removed. By comparison, you will discover that many other metals and materials routinely used in military vehicles have comparable isotope activity. Just because it has the word "uranium" in it does not make it automagically evil and scary. The radiation is a non-issue in that it amounts to normal background exposure in a lot of places in the world. You don't think uranium is made by magic elves do you? They pull the undepleted stuff out of ground concentrations.
Second, the amount of DU expended in the desert during the war was around 40 tons. By comparison, the amount of undepleted Uranium ash dumped in the USA alone is measured in the thousands of metric tons every year. And that ignores the thousands of tons of thorium ash which is also dumped. That according to the US government. You whine about some tiny amount of radioisotope depleted uranium out in the desert and then continue to ignore the vast quantities of undepleted metal dumped next to population centers. You either have a knee jerk political agenda or some really screwed up priorities.
Third, there is a very good materials science reason to use DU for the purposes it is used, and there isn't much in the way of substitutes that wouldn't be some combination of insanely expensive and far more nasty to the environment.
Quite frankly, if you look at the other metals and materials used in military equipment, and actually read the MSDS on them, you'll discover that a wee bit of Uranium disappears underneath the noise floor. All the selective outrage is absurd.
What I'm interested in, is the quality of the data the government is mining. They are going through a lot of databases that may contain incorrect, false, misleading, or outdated data:
Absolutely true. I worked on a huge commercial/government mining and data integration project in the mid-90s (this isn't exactly something Bush thought of) and the state of the government databases were attrocious. Extremely dirty data, and of the over a thousand databases that were integrated, half the government ones looked like they had been maintained in a spreadsheet or flat file rather than any kind of real DBMS. A lot of the data was so out of spec that we discarded 10% of the records during the integration because it couldn't be unambiguously deciphered. A real data management nightmare; it is a wonder they can find anything at all.
By comparison, commercial databases are immaculate. The government needs to develop some type of professional data management corps. Private industry would think you were mad if you suggested managing data the way the government often does. Lots of one-off hacks by people who know nothing about basic principles of creating a database. This isn't just at the Federal level, the State databases were just as bad.
Back when I was doing work for the government, the open secret was that the quality of the data in the databases was so low that almost nobody used it because it couldn't be trusted. People were paid to maintain the systems, but very little utility was ever realized.
But the US has 10,000 nuculear warheads, enough to 'overkill' the worlds pupulation 12x. For those of you not in the military, this means that if the bodies of the dead were to get up again, we could kill them all 12 more times.
That's a lot of nonsense. There is no comparison to a whopping big asteroid hitting the planet, even as a simple matter of the total energy equivalence. Its like saying that a liter of botulinum toxin can kill the entire race. Technically it might be true (I don't really know), but in practice you can't come remotely close to killing the entire human race even if you had the toxin and nobody stopped you from using it.
If the US used its entire nuclear arsenal with the specific intent to maximize the body count, they still would probably not be able to kill more than one billion at best. Do the math. You have to kill 100,000 people per nuke for every single nuke, and even that will only get you the first billion. And in practice, there are only a relatively small number of targets that will net you that many kills if you nuke them.
Nukes are very destructive, but most people have a conception of their destructive potential that is totally out of proportion to what they are actually capable of. We have enough nukes to scorch a good bit of earth, but not even remotely enough to do a planet kill or even make more than a minor dent in the total human population.
The problem is that the VOIP providers like Vonage want to use the network but don't want to be responsible for it, and then bitch about the SLAs and agreements made between the parties that actually do run the networks. If they want specific terms and conditions from the network providers, they can pay for them like everyone. And the VOIP customers have to understand as well that if they choose an ISP that has a poor SLA for VOIP, they'll have poor VOIP service.
That said, I find it generally unlikely that ISPs would do any type of overt targeted network shaping. They make their money by moving packets, and for more and more contracts these days, the more packets you move the more money you make.
The benefit of ISPs getting into the VOIP, streaming, and other services where network properties matter is that those are exactly the kind of people who can optimize their networks to give the customer the best experience. ISPs want to displace Vonage because Vonage isn't their customer, but they have to deal with all the network issues generated by customers that use Vonage. It is cheaper to offer an optimized solution designed and tested to work beautifully on your network for free or nearly free than to support the problems caused when people use whatever random VOIP software suits their fancy.
Not all networks are created equal, and this really starts to become apparent when using QoS sensitive services. It is cheaper and generally gives better results for the ISP to integrate those services vertically, which ultimately will be a win for the customer.
Wasting a 100x100 mile area is what the enviros will also complain about because of the disruption to the local ecology. There is no group harder to please than they are.
The US is sitting on the Saudi Arabia of geothermal sources (the Great Basin rift zone out in the western desert i.e. Nevada), and the environmentalists already complain about how geothermal power farms are unsightly and interfering with the local ecosystems. Mind you, geothermal is pretty clean power, there is enough of it in the Great Basin alone to meet the energy needs of the US many times over, and these power stations are often situated in barren locations where no one ever goes. And geothermal is pretty sparse, as you need about 20 acres for every pipe you put in the ground to keep the bore holes far enough apart.
So yeah, paving 10,000 square miles of desert to put in algae vats would seriously annoy the environmentalists. By comparison, geothermal leaves most of the environment intact and they already complain about that.
Is this a full inertial platform or just really the accellerometer thing that I mentioned (which should be good enough for a bomb that just glides down). A full INS kit would be probably overkill.
The JDAMs use a real Ring Laser Gyroscope for their INS package (it would be hard to explain the cost otherwise).
As for whether or not this is overkill, remember that this is the primary guidance system for these munitions and a fairly high degree of precision is needed. Crude accelerometers won't give the resolution required to get arcsecond accuracy from the guidance package and allow fine course correction from GPS signals.
Inertial guidance on a bomb? They may have some simple accellerometers, but I doubt whether they have full inertial navigation (with the laser gyroscopes), it is still rather expensive for a regular HE bomb.
The current generation of JDAMs is primarily an inertial guidance weapon that accepts GPS corrections, and is capable of being quite precise in the absence of GPS. The very first generation ten years ago may have been primarily GPS guided.
As for price, the JDAM units are something like 15-20% of the cost of missiles that use a thermal imaging guidance package, so the cheapness is relative.
GPS jammers are nearly useless. They are only powerful enough to cover a small area, so their only use is to protect a stationary target from attack by GPS guided bombs. Unfortunately, as demonstrated in the Iraq war last year, they don't even do that effectively. All six of the Russian-made GPS jammers fielded by Iraq were destroyed in short order, some of them by GPS guided missiles!
That the US uses GPS guided bombs is a common misconception. The US uses GPS corrected guidance systems. The so-called "GPS guided bombs" actually use an inertial guidance system (not jammable), with a very jam resistant GPS antenna on the back-end to make minor corrections to the inertial guidance package. The accuracy difference is the difference between making the bomb land in your bathtub and land in your house. Even if you did jam the GPS effectively, you still don't want to hang around for the bomb to hit.
The majority of espionage conducted against the US is by our friends, largely from Europe. UK, France, and Germany being the major active players from Europe as I seem to remember. While it doesn't get wide press, the US catches (and then deports) several hundred European spies every year. How spies are treated depends on what country they are from.
I remember over a decade ago when I actually worked in a business in which we were espionage aware, that the number one espionage problem in the US was the French (followed by the Chinese, and then a laundry list of European countries -- including the UK), the French being primarily interested in stealing US weapons technology and listening in on business deals they were competing with. Which was primarily a business move; along with the Russians and the US, the French are one of the world's major arms exporting countries and they have to compete with US designed weapons on the open market.
Everyone spies on everyone, and for varying reasons. The French actually used to have one of the most aggressive intelligence services on the globe, disproportionate to their size and geopolitical importance, which some people find surprising. I don't know if it as large today, though. But this is nothing new, and all the governments understand that this goes on. As long as it doesn't get out of hand, it is tolerated between countries that are nominally friendly.
As a point of reference, I could really use about a terabyte of RAM (not that I have it) for research purposes right now and I can easily seeing having a ton of use for a petabyte of RAM because we will eventually outgrow our current needs by a couple orders of magnitude. Access time is a matter of tractability for many algorithm spaces.
Of course, at this point the real problem starts to become memory latency and bandwidth.
"I can imagine this kind of technology being really applicable in situations where large databases are in use -- but potentially, slightly cheaper then just keeping the entire database in ram."
One of the major problems with databases is how much RAM you can physically put in the machine. You can put terabytes of "almost RAM" on your disk subsystem for a server that maxes out at 16Gb of physical motherboard RAM. If your working set is large such that there is no way you'll be able to practically fit it in RAM on the motherboard, this can serve as an intermediate solution for extending the system RAM rather than bottlenecking on platters.
It is primarily useful as a pseudo RAM extender, for applications that simply won't fit in the amount of RAM most servers can hold but which need RAM-like access speeds.
"There are a number of phenomeona (unitary experience and the flow of time to name but two) that are completely ignored by computational models of consciousness."
I don't think we are talking about the same thing here. I'm not aware of any computational model of consciousness, only that there are general computational models from which very thorough definitions of "consciousness" will fall out, and that includes addressing issues like the above.
The quantum theory is a red herring; quantum mechanics (from a computational standpoint) only modify time complexity, whereas definitions of "consciousness" generally fall out of the math as a function of space complexity. To put it another way, even if the mind was a quantum computer of some type, the mathematics from which my assertion is made would still hold, making the nature of the computer a distinction without a difference. This has actually been flogged pretty thoroughly in other forums.
"After all, consciousness does a lot of things that a purely logical calculation would have a hard time coming to. Like having a favorite color, falling in love, or disliking lima beans."
My point about the definition of "computation" you were using is that it is pedestrian, and when stated like that it appears to exclude a great many things which it does not exclude under most rigorous definitions. Having a favorite color, falling in love, and disliking lima beans can all be reduced to finite algorithmic information (at least in theory); the fact that we can describe them and perceive them supports this.
The problem with your above comment is that it is only kinda correct assuming vanilla axiomatic logic systems, a theoretical subset, which it appears you seem to think is the only kind. A lot of people do, because it is the one most people are taught. The human brain is pretty clearly uses a non-axiomatic (read: inductive) computational model. Note that "non-axiomatic" is descriptive -- these systems are still formally describable. Non-axiomatic reasoning systems and computational models simply don't have an absolute frame of reference for represented patterns, which gives them most of their interesting properties.
I would agree that consciousness is hard to describe in strict axiomatic systems, but I think most people working on core theory will tell you that the brain and all the properties we attribute to it are quite expressible in non-axiomatic systems.
"This assumes that consciousness is based solely on computation. Not proven yet."
Ignoring the strong mathematical and theoretical evidence that it is based on computation, let me ask you: What do you think "computation" is?
It is fairly apparent that you are using a definition of "computation" that is sufficiently narrow as to be essentially invalid for theoretical or mathematical purposes. I don't think that word means what you think it means.
Whether something is algorithmically finite is something that can be tested theoretically to extremely high probability without knowing the exact nature of the thing being tested. Consciousness tests as such in any measurable way you care to define/describe it.
Note that "algorithmically finite" is not the same thing as "finite state". Some classes of infinite state system are algorithmically finite. The expression of certain computational properties in a system (an analog of "the laws of thermodynamics" being one) are very strong indicators of an algorithmically finite system and would be a valid characterization even if the universe has a countably infinite state space.
In fact, it looks very much like it is describing something that was already known to be true with more rigor and detail than is provided here. This may have been a "new" idea to the author of the paper, but it isn't original.
Slashdot Mirror
Postgres is getting really close to the functionality and capabilities of the Big Commercial Enterprise DBMS, close enough that anyone can see that bridging that gap is quite doable. Most of the arguable weaknesses in Postgres are in the more esoteric high-end feature space, as it is already strong and quite feature complete for most routine RDBMS work. And the upcoming new version addresses a great many of those weaknesses. As the article said, this is going to be a major release.
The self-interest part is that it is a HELL OF A LOT CHEAPER for a corporation to pay people to add those last few features and bits that they want to Postgres than to pay an unholy amount of money to buy the required Oracle licenses. The Postgres engine is clean and fundamentally pretty good in an engineering sense, and so enterprise feature tweaks are relatively cheap. It is all about dollars and sense at the end of the day. Purchasing Postgres plus feature development is almost always going to be vastly cheaper than buying Oracle. And unlike Oracle, it is pretty much a one-time fixed cost. It is worth repeating that the engineering strength and scalability of the underlying Postgres platform is the primary reason the market is evolving this way. The gap between MySQL and high-end RDBMS is comparatively much too great for a company to fund closing that gap because a lot of additional arguably unrelated work may be required because of the internals. This increases time to delivery of features, increases the cost of adding high-end features, and increases the risk of problems.
If Oracle suddenly dropped its enterprise licensing costs by a couple of order of magnitude, then it would seriously threaten Postgres development. But since that is unlikely to happen, corporate money will continue to flow into making Postgres a formidable Oracle replacement, which it is already well on its way to being.
Oh hell, mod this guy down. This is factually dead wrong and they apparently know exactly nothing about terminal ballistics and its relation to cartridge design.
The AK47 has lousy penetration and short effective range due to the cartridge selection. The very low sectional density combined with relatively low velocity of the 7.62x39 makes it one of the poorest penetration performers of any military cartridge in common use today. The M16 cartridges that have been in use for the last decade or two (M855) will actually out-penetrate most everything else, particularly through hard materials, including the 7.62x51 (.308) NATO cartridge.
Sectional density * velocity = penetration, with some bonus points for bullet design. The 7.62x39 has neither sectional density nor velocity, the 5.56x45 has moderate sectional density and high velocity.
I can't believe the drivel that is being modded as informative. It is a big steaming pile of urban myth nonsense.
There is a huge amount of hearsay and legend surrounding the AK47 and M16, usually propagated by people without the experience or context to have an informed opinion.
For a well-trained soldier, the M16 system is vastly superior to the AK47 in just about every way. And the US has "well-trained" soldiers. In a more general sense, the AR15/M16 system is one of the finest rifles of its class ever fielded and has numerous advantages that aren't immediately obvious unless you've had to do extensive live fire combat training with both. Beside being a very nice rifle to shoot and handle, the M16 has the distinction of being one of the quickest combat rifles out there, while the AK47 is one of the slowest. By "quick" I mean how fast you can you drop some number of multiple random targets at random ranges. The M16 is blindingly quick for fast and furious engagements, the AK47 is not. After using an M16, it feels like you are trying to shoot using something with the handling and aiming characteristics of a cinderblock. In a firefight that is all about you shooting the other guy before he shoots you, this matters a lot.
As for jamming, that is largely mythology. There was a problem in the 1960s where some Army supply person decided to ship ammo that was out-of-spec for the engineering of the rifle because it was cheaper without checking with the designers. That caused a lot of problems. The rifle was quickly redesigned to be more tolerant of out-of-spec ammo. As a result, jamming hasn't been a problem for about 30 years. In the field where the dirt is, rule of thumb is that the typical modern M16 will function flawlessly for about 1500 rounds without cleaning. That is a LOT of ammo, and several times the max load that an infantry soldier normally carries.
Errm... no its not. It isn't even a controlled waste. It is simply dumped into the environment. I'm not talking about nuclear power plant waste, which isn't anywhere near thousands of metric tons anyway. I'm talking about radioisotope concentrators like coal ash, which measures substantially above background. And it contains thousands of tons of isotopes that are far nastier than uranium e.g. thorium.
Incidentally, studies have been done on the populations that live near coal slag dumps and similar, which have high concentrations of uranium, thorium, and other radioisotopes. There is a small statistical bump in cancer, most of which is typically attributed to the thorium component.
The deaf ear may have something to do with you being illiterate on the basic facts of the issue.
First, depleted uranium is "depleted", meaning that the active and interesting isotopes have been largely removed. By comparison, you will discover that many other metals and materials routinely used in military vehicles have comparable isotope activity. Just because it has the word "uranium" in it does not make it automagically evil and scary. The radiation is a non-issue in that it amounts to normal background exposure in a lot of places in the world. You don't think uranium is made by magic elves do you? They pull the undepleted stuff out of ground concentrations.
Second, the amount of DU expended in the desert during the war was around 40 tons. By comparison, the amount of undepleted Uranium ash dumped in the USA alone is measured in the thousands of metric tons every year. And that ignores the thousands of tons of thorium ash which is also dumped. That according to the US government. You whine about some tiny amount of radioisotope depleted uranium out in the desert and then continue to ignore the vast quantities of undepleted metal dumped next to population centers. You either have a knee jerk political agenda or some really screwed up priorities.
Third, there is a very good materials science reason to use DU for the purposes it is used, and there isn't much in the way of substitutes that wouldn't be some combination of insanely expensive and far more nasty to the environment.
Quite frankly, if you look at the other metals and materials used in military equipment, and actually read the MSDS on them, you'll discover that a wee bit of Uranium disappears underneath the noise floor. All the selective outrage is absurd.
Absolutely true. I worked on a huge commercial/government mining and data integration project in the mid-90s (this isn't exactly something Bush thought of) and the state of the government databases were attrocious. Extremely dirty data, and of the over a thousand databases that were integrated, half the government ones looked like they had been maintained in a spreadsheet or flat file rather than any kind of real DBMS. A lot of the data was so out of spec that we discarded 10% of the records during the integration because it couldn't be unambiguously deciphered. A real data management nightmare; it is a wonder they can find anything at all.
By comparison, commercial databases are immaculate. The government needs to develop some type of professional data management corps. Private industry would think you were mad if you suggested managing data the way the government often does. Lots of one-off hacks by people who know nothing about basic principles of creating a database. This isn't just at the Federal level, the State databases were just as bad.
Back when I was doing work for the government, the open secret was that the quality of the data in the databases was so low that almost nobody used it because it couldn't be trusted. People were paid to maintain the systems, but very little utility was ever realized.
That's a lot of nonsense. There is no comparison to a whopping big asteroid hitting the planet, even as a simple matter of the total energy equivalence. Its like saying that a liter of botulinum toxin can kill the entire race. Technically it might be true (I don't really know), but in practice you can't come remotely close to killing the entire human race even if you had the toxin and nobody stopped you from using it.
If the US used its entire nuclear arsenal with the specific intent to maximize the body count, they still would probably not be able to kill more than one billion at best. Do the math. You have to kill 100,000 people per nuke for every single nuke, and even that will only get you the first billion. And in practice, there are only a relatively small number of targets that will net you that many kills if you nuke them.
Nukes are very destructive, but most people have a conception of their destructive potential that is totally out of proportion to what they are actually capable of. We have enough nukes to scorch a good bit of earth, but not even remotely enough to do a planet kill or even make more than a minor dent in the total human population.
That said, I find it generally unlikely that ISPs would do any type of overt targeted network shaping. They make their money by moving packets, and for more and more contracts these days, the more packets you move the more money you make.
The benefit of ISPs getting into the VOIP, streaming, and other services where network properties matter is that those are exactly the kind of people who can optimize their networks to give the customer the best experience. ISPs want to displace Vonage because Vonage isn't their customer, but they have to deal with all the network issues generated by customers that use Vonage. It is cheaper to offer an optimized solution designed and tested to work beautifully on your network for free or nearly free than to support the problems caused when people use whatever random VOIP software suits their fancy.
Not all networks are created equal, and this really starts to become apparent when using QoS sensitive services. It is cheaper and generally gives better results for the ISP to integrate those services vertically, which ultimately will be a win for the customer.
The US is sitting on the Saudi Arabia of geothermal sources (the Great Basin rift zone out in the western desert i.e. Nevada), and the environmentalists already complain about how geothermal power farms are unsightly and interfering with the local ecosystems. Mind you, geothermal is pretty clean power, there is enough of it in the Great Basin alone to meet the energy needs of the US many times over, and these power stations are often situated in barren locations where no one ever goes. And geothermal is pretty sparse, as you need about 20 acres for every pipe you put in the ground to keep the bore holes far enough apart.
So yeah, paving 10,000 square miles of desert to put in algae vats would seriously annoy the environmentalists. By comparison, geothermal leaves most of the environment intact and they already complain about that.
The JDAMs use a real Ring Laser Gyroscope for their INS package (it would be hard to explain the cost otherwise).
As for whether or not this is overkill, remember that this is the primary guidance system for these munitions and a fairly high degree of precision is needed. Crude accelerometers won't give the resolution required to get arcsecond accuracy from the guidance package and allow fine course correction from GPS signals.
The current generation of JDAMs is primarily an inertial guidance weapon that accepts GPS corrections, and is capable of being quite precise in the absence of GPS. The very first generation ten years ago may have been primarily GPS guided.
As for price, the JDAM units are something like 15-20% of the cost of missiles that use a thermal imaging guidance package, so the cheapness is relative.
That the US uses GPS guided bombs is a common misconception. The US uses GPS corrected guidance systems. The so-called "GPS guided bombs" actually use an inertial guidance system (not jammable), with a very jam resistant GPS antenna on the back-end to make minor corrections to the inertial guidance package. The accuracy difference is the difference between making the bomb land in your bathtub and land in your house. Even if you did jam the GPS effectively, you still don't want to hang around for the bomb to hit.
I remember over a decade ago when I actually worked in a business in which we were espionage aware, that the number one espionage problem in the US was the French (followed by the Chinese, and then a laundry list of European countries -- including the UK), the French being primarily interested in stealing US weapons technology and listening in on business deals they were competing with. Which was primarily a business move; along with the Russians and the US, the French are one of the world's major arms exporting countries and they have to compete with US designed weapons on the open market.
Everyone spies on everyone, and for varying reasons. The French actually used to have one of the most aggressive intelligence services on the globe, disproportionate to their size and geopolitical importance, which some people find surprising. I don't know if it as large today, though. But this is nothing new, and all the governments understand that this goes on. As long as it doesn't get out of hand, it is tolerated between countries that are nominally friendly.
Of course, at this point the real problem starts to become memory latency and bandwidth.
One of the major problems with databases is how much RAM you can physically put in the machine. You can put terabytes of "almost RAM" on your disk subsystem for a server that maxes out at 16Gb of physical motherboard RAM. If your working set is large such that there is no way you'll be able to practically fit it in RAM on the motherboard, this can serve as an intermediate solution for extending the system RAM rather than bottlenecking on platters.
It is primarily useful as a pseudo RAM extender, for applications that simply won't fit in the amount of RAM most servers can hold but which need RAM-like access speeds.
I don't think we are talking about the same thing here. I'm not aware of any computational model of consciousness, only that there are general computational models from which very thorough definitions of "consciousness" will fall out, and that includes addressing issues like the above.
The quantum theory is a red herring; quantum mechanics (from a computational standpoint) only modify time complexity, whereas definitions of "consciousness" generally fall out of the math as a function of space complexity. To put it another way, even if the mind was a quantum computer of some type, the mathematics from which my assertion is made would still hold, making the nature of the computer a distinction without a difference. This has actually been flogged pretty thoroughly in other forums.
My point about the definition of "computation" you were using is that it is pedestrian, and when stated like that it appears to exclude a great many things which it does not exclude under most rigorous definitions. Having a favorite color, falling in love, and disliking lima beans can all be reduced to finite algorithmic information (at least in theory); the fact that we can describe them and perceive them supports this.
The problem with your above comment is that it is only kinda correct assuming vanilla axiomatic logic systems, a theoretical subset, which it appears you seem to think is the only kind. A lot of people do, because it is the one most people are taught. The human brain is pretty clearly uses a non-axiomatic (read: inductive) computational model. Note that "non-axiomatic" is descriptive -- these systems are still formally describable. Non-axiomatic reasoning systems and computational models simply don't have an absolute frame of reference for represented patterns, which gives them most of their interesting properties.
I would agree that consciousness is hard to describe in strict axiomatic systems, but I think most people working on core theory will tell you that the brain and all the properties we attribute to it are quite expressible in non-axiomatic systems.
Ignoring the strong mathematical and theoretical evidence that it is based on computation, let me ask you: What do you think "computation" is?
It is fairly apparent that you are using a definition of "computation" that is sufficiently narrow as to be essentially invalid for theoretical or mathematical purposes. I don't think that word means what you think it means.
Note that "algorithmically finite" is not the same thing as "finite state". Some classes of infinite state system are algorithmically finite. The expression of certain computational properties in a system (an analog of "the laws of thermodynamics" being one) are very strong indicators of an algorithmically finite system and would be a valid characterization even if the universe has a countably infinite state space.
In fact, it looks very much like it is describing something that was already known to be true with more rigor and detail than is provided here. This may have been a "new" idea to the author of the paper, but it isn't original.