Replies made most of the good points here, but you're onto the correct solution. You need to use an encryption system such that the drive ends up in a state where you are unable to recover it yourself, at least not using any means that the attacker can discover. Of course, that could well mean that you're out the data yourself, and you'll certainly be out your computer hardware until it is so old as to be worthless.
Usually this is best implemented using hardware methods (TPM/etc) - assuming that you can trust the hardware. Hardware designed for key storage usually is tamper-resistant and is designed to permanently wipe its contents upon too many failures/etc.
As others have suggested, you need to avoid attacks on RAM as well.
The problem is that it is easy to save money if you are willing to take on risk, and corporate incentives are usually structure to encourage this.
You collect the bonus for not overrunning your budget now. Your successor might get hammered over the missing backups later, or maybe they'll pass on the blame and everybody will come out fine (but the shareholders, who don't get to make these decisions).
In hindsight the decision to not make backups pays off in every single scenario except the one where there is some huge disaster. 99% of the time there isn't a huge disaster. So, 99% of the time the decision to not make backups pays off. Then the other 1% of the time there is probably an 80% chance that the personal impact on the decision maker will not be sufficient to make planning for the 1% worthwhile.
Why do you need redundancy in the BACKUP drives? They're already redundant. Sure, you do need two sets of drives with the hard drive solution, but you need one set of drives with the tape solution so the cost of one set is effectively zero.
Now, I'll grant you that hard drives are more fragile than tapes, so perhaps some level of redundancy makes sense. I'm not sure you need raid1 for 46TB of data, however, so that you can half half of the drives fail.
Also, putting the carbon dioxide back where it belongs is actually really easy. Just don't cut the trees and put trees back where we cut them. The rest comes naturally.
Yeah, on the timescale on which all that coal formed in the first place.
By then we'll have polluted half the habitable planets on this side of the galaxy.
I'm all for de-externalizing things as much as possible. Tax power based on CO2 production, particulate health problems, the cost to build Yucca Mountain, the cost of maintaining a fleet of carriers and bombing the middle east into the stone age once every 20 years, or whatever you teed to tax it with. Then the market can figure out what is cheapest, while following the regulations appropriate to each type of power.
I think we'll find that some of the "expensive" forms of power generation aren't so expensive when you de-externalize them. I suspect that in the end you'd end up with lots of nuclear in the medium term, but also a much bigger push for technologies that could replace it. The "cheapest" options out there right now probably will end up being the most expensive.
Of course, taxing this stuff correctly will be hard. How many carriers does the US need if it doesn't need to invade the middle east from time to time? What is the true risk of a dam failure and what cost should be assigned to that? Risk is hard to assess - the risk of a 2008 economic meltdown was deemed to be unimaginably low (by people who had incentive to consider it low). No doubt every special interest out there would push hard to have their favorite technology declared "safe" and as such this wouldn't work. Oh, and everybody loves to define the metrics - nobody wants to count dead coal miners, the cost of the US navy, or lung cancer deaths (unless from nuclear accidents), and even regulating CO2 is highly controversial.
NPR had a good series on problems in the healthcare system (unlike most treatments they seemed to take a holistic approach and not just find one issue and make it out to be the single thing that is driving up cost). They had a story about a doctor who didn't prescribe a CT scan for a girl who had a suspected spinal fracture. They ended up getting into a fight with the girl's dad who felt like the doctor was just cutting costs at the risk of the girl's health.
However, the doctor pointed out that he had every reason to do exactly what the father wanted. He would be paid more if he ordered the test. Nobody would dispute the test since doing so would expose them to liability. He was taking on risk of liability in the event he was wrong and she had a fracture. However, the fact was that the CT scan for a girl of her age carried a significant increase in risk of cancer much later in life, and based on his physical exam that risk was much greater than the risk that she might have an undetectable fracture. Of course, if she did get cancer later in life it could never be linked to the one test, and by then the doctor would probably be retired/dead/etc. So, the doctor was sticking his neck out, and taking on lots of personal risk, and declining the opportunity to make money, and wasting his time explaining all of this stuff to a father who would have just said yes no discussion needed to a CT scan, all because he wanted the girl to be healthy. How many doctors would do otherwise?
Antibiotics are a similar situation - the doctor can argue with their patient, and maybe lose them. Or, they can just take 30 seconds to fill out a prescription that they'll never get sued over and everybody is smiling since now when the patent gets better it will be because of what the doctor did. Happy patients lead to more patients, and more repeat business as well. Even if the doctor is employed by something like the NHS they probably have metrics and sending a patient on their way without hassle means more visits per hour. Or, even if they're completely unaccountable who wants to sit and argue with somebody all day?
I'm generally in favor of eliminating the need to get a prescription to get access to drugs, but antibiotics are one area where I'd make an exception. I don't see the role of government as protecting people from themselves. However, antibiotic abuse harms everybody and it is completely legitimate to regulate their use - probably more strictly.
I'd probably require doctors to submit a written justification for every prescription of an antibiotic that is less than 20 years old, and with stricter requirements on anything less than 10 years old (either documented testing that shows resistance to the alternatives, or an assessment which will be reviewed by a board that the patient would suffer irreversible harm if they waited for the results of that testing).
Of course, if you do this the market for new antibiotics is almost worthless (compared to being low-value which it is now). Who will spend a billion dollars working on new antibiotics only to release one and have 30 people take it in 10 years? The solution here is bounties - governments will have to decide how many new antibiotics they want and offer substantial bounties for their discovery (probably hundreds of millions of dollars), and use that money to buy the patent rights (the company has already been paid for their work). The bounties can be adjusted based on the number of candidates that are being submitted vs the number desired.
You could actually apply a similar model to other drugs, but it would get expensive (probably cheaper than what we're doing now, but this is completely socializing medicine which is of course much more expensive than having most of the costs be privately borne by patients and their employers). If it were successful enough you'd see the drug patent problem go away without even having to ban them, since patented drugs would be much more expensive than the generics bought with bounties, and companies would still have incentive to do R&D (but not marketing, etc).
That's about as convincing as a CEO claiming that he wasn't responsible for his own company's pollution, because he told his subordinates to follow the law. It isn't enough for somebody in power to tell the people they employ to follow the law - they must do due diligence to ensure they are doing so.
Or, consider a factory that hires a bunch of people for $10/hr and tells them to open/pack boxes/etc. The 5 slowest people each week lose their jobs, but the rule is that you have to wear your safety gear and be careful/etc. Of course, the supervisor doesn't punish you for not wearing your safety gear, but they do watch you to figure out if you're one of the 5 people who will lose their jobs that week. Then a worker chops their hands open with a box cutter and sues, and the company says, well, we did provide safety gloves and tell people to use them.
The company in a position of power has a duty to ensure the law is followed - anything else leads to huge problems. Goods coming from places with these kinds of problems should have 500% tariffs applied.
Agreed. Of course, to replace all the cars in a household with this technology you need to work on charging time.
Stopping every 200 miles on a road trip isn't a big deal if it takes 5-10 minutes to recharge. It would probably be a lot safer than gas so you can just plug the thing into your car and walk into the store and buy something while you wait (leaving gas pumping in this way is dangerous). A replacement battery model could also be fast if you could pull off the necessary standardization and deal with the quality issues.
If you automate/standardize it enough you could make refueling stations on interstates that are just pull-off areas where you pull up to a station, it scans your RFID, swaps your batteries, and in 30 seconds you're hitting the gas and pulling through and back onto the road.
The big win will be greatly reducing oil demand - get rid of the fighting-two-wars-all-the-time military funding and you can probably just give free electricity to everybody.
I work in a large company (not government/CIA) and if we had some kind of break-in with our systems we'd be going through all the sorts of things you suggest. And we care about turning a profit so it isn't like we just spend money for the sake of doing so. The CIO would be ticked, but he'd be brining in those consultants pronto because with something like this you have to do it right.
I'm sure some data would be lower risk than others and might not get as much scrutiny, but just determining what is important and what isn't costs money. You find a server that you think might be at risk, then you look up the contact for that server, then you trigger a 30-email conversation (for one server) as they track down the people who are REALLY responsible and have the authority to speak for it, and then you send them the 14-question survey and put the score on the big list.
Oh, and if any of the data in those systems is subject to regulation (EVERY big company is regulated in some way) you get the lawyer-types asking "what if" questions and you end up assessing more than you probably need to anyway.
As you indicated you can't just roll back production systems a few weeks once you realize the exploit is that old. Does amazon.com just forget about every order they took in the last month that they billed but didn't ship? Do they just reset their inventory to what it was a few weeks ago (an eternity in the just-in-time world)? Do they bill and re-ship orders they already filled? Do they forget about pre-orders and lose business and frustrate customers who wanted their widget on day one at the same time?
And even if you trust that it isn't an inside job you still need the army of consultants. Companies don't staff to just absorb these kinds of distractions - they run their staff at 120% on an average day. You need an army of consultants just to make sure this stuff all gets done reasonably quickly.
Then there is the distraction factor as everybody is busy doing risk assessments and reviewing data when they'd normally be doing something that actually makes the company money (ostensibly the reason they have jobs in the first place).
Yup, a breach like this is simple in the same way that backing up your computers is simple, or buying a hard drive at 0.1 cents/GB is simple. It is all simple until you have a bazillion of them, and your job is on the line.
How? In a word, patronage. It doesn't necessarily have to be government patronage.
Uh, you do realize that developing a drug (successful or not) costs about $10-$100M, and developing a successful drug requires developing about 10 unsuccessful ones first (on average)? You're talking about the better part of a billion dollars. Do you recommend setting up a paypal page for that?
Only other way that might work is to do nothing, and leave it to artists and scientists to get what they can from the first mover advantage and from keeping secrets. That's why we might prefer some kind of patronage, to pay out money in exchange for going public, and because it is very difficult to keep recipes secret.
Considering that just about every civilized country on the planet requires companies to disclose just about every detail of the process in order to even test a drug keeping it secret would be virtually impossible. Or, are you suggesting that we no longer regulate clinical trials to ensure patient safety? Oh, and are you suggesting that doctors should prescribe pills without knowing what is actually in them?
And figuring out what is in a pill isn't all that expensive - once it hits the market everybody will know what is in it. Sure, it will be harder without actually knowing the molecular entity, assuming that we manage to keep that secret, but a few weeks with an LC and an MS is all it will take most of the time. We already know that from there reverse engineering it is pretty cheap - just look at generics in India or whatever.
There are all kinds of prizes, such as the X prize.
Sure, but they tend to be hundreds of thousands of dollars to tens of millions of dollars. You need to expand that by 1-2 orders of magnitude just to come up with a single drug. Also, modern drug development turns out several new drugs per year typically, and the x-prize model is designed to encourage people to come up with some groundbreaking idea once a decade or whatever. If you need a billion dollars for ONE drug, do we really want to limit ourselves to the rate of development we would come up with donations?
They often do not create drugs themselves. They lean more on university research, and the knowledge of native peoples. Then they patent all these drugs that they did not invent, while those who actually originated the drug get nothing. They also externalize much testing, leaving it to the FDA to figure out if a drug is safe. Or they experiment on the public!
Drugs originating from native remedies aren't all that common, though they do happen (I suspect that most of those have already made it to market - at some point you start running out of newly-discovered civilizations). Ones based on university research are very common. However, typically a university tends to come up with a concept, and maybe a compound that inhibits some enzyme in a test tube. Often the original compound either has safety issues or doesn't reach the target tissue in a living person. Sometimes the original compound actually works out, but only after that billion dollars gets spent ($100M for that drug, and then $900M on the other 9 university compounds that didn't work out).
I don't dispute that much of the "creative" work happens outside of drug companies. However, the reality is that no matter how great the original ideas were, nothing is getting into the doctor's office until the clinical trials are done, and those cost the better part of a billion dollars for every drug that makes it.
As far as credit for discovery goes - it is pretty typical that if a university or whatever comes up with a compound that they'll patent it and sell the rights to a drug company. Whether the university bothers to pay their scientists well is a different matter. I suspect that compounds discovered internally in drug companies trigger some kind of employee bonuses, but if the company makes $10B over 15
The spartan and functional bit seems to be the cluster of buildings to the south of the desert paradise. Maybe that is where all the ordinary people live?
I suspect that the antenna length isn't nearly as critical for reception, especially if the transmitter is fairly powerful. ELF is one-way. If the sub needs to reply it would go to periscope depth and deploy a satellite antenna. In fact, I suspect that half of the messages that would be sent on ELF amount to "sub 1234, check satellite traffic" - anything else would have to be completely pre-planned (like "sub 1234, go with plan A"). Bitrates on ELF are very low.
From what I've seen on documentaries the US now actually has ELF antennas on large jets - they fly in circles and deploy the antenna from high altitude and it spirals down and it apparently is good enough. That is probably a lot more survivable than a huge antenna in a fixed position. Basically just another step to ensure that we don't miss out on the end of the world should the time come.
As I recall their stated purpose for Android was promoting more access to the web from mobile devices, so that they could then expand the market for their web-based technologies. I'd say that is successful whether they collect license fees for the OS or not.
As long as the Nook has a web browser then it is mission accomplished for Google. That web browser will be pointed at Google-owned sites, and based on the cookies/etc it ends up gathering it increases Google's knowledge of the consumer in a new domain and thus increases their advertising power.
I can't say whether their strategy will work out, but every $50 smartphone that is sold that doesn't run Android is a success for Android by their original goals. They turned smartphones from $500 items owned by the fairly well-to-do or for business use only to something that might sell 3 units to a family of 4.
Using a winch would probably be a lot more practical and less likely to upset the EPA. However, either approach completely misses the point. People are all finite in their capabilities, both physical and mental, and some are more limited than others. As you raise the bar people start getting excluded, and you need to figure out what to do with them.
The average slashdotter is probably still well above the bar, but the average person is probably close to the edge if not below it.
You don't have to make anybody do anything. You just apply a tax of a few hundred percent to their exports if they don't. China is welcome to pollute their streams, but they won't be allowed to sell their products in the US on the cheap if they do.
Countries that are civilized can have free access to US markets, and vice-versa.
I'm not suggesting the US should rule the world - only that US companies shouldn't be able to outsource crimes to countries where they are legal.
No need - just have a tax table with countries on one side and rates on the other. If you're Belgium you get a rate of zero, and if you're China maybe you get a rate of 300%.
There aren't that many countries so no need to "micro-manage." If a country wants to get its rate changed they can bend over backwards to convince everybody that it is justifiable.
Or I guess we can just fall back to the original proposal and tax the living daylights out of imports from anywhere without exception. Then we merely need to micro-manage our own economy.
In what way do you disagree with my usage of it? Most likely if they did pick Apple they'd get something close to the consumer model, but not quite the same. So, it is an "iPhone", but it really isn't, and that seems like an appropriate use of per se.
It might even be just an iPhone minus the EULA and warranty disclaimer paperwork in the box, but that still isn't quite the consumer model.
Per wikipedia a good synonym of "per se" is "without qualifications" - and I don't think they'll be using a consumer phone without qualifications.
I wasn't suggesting that improved communications wouldn't have benefits. I'm just saying that existing consumer devices need many tweaks to be useful in a combat situation.
Just looking at your list you need to ensure: 1. The phone has sufficient communications range to whatever it is talking to. 2. That if you depend on that phone for essential support in an ambush or whatever that whatever it is talking to is itself secure (ie not an unguarded cell tower in the middle of some contested city). 3. That the GPS can't be spoofed (probably not guaranteed on consumer GPS), and that it is ready for selective availability should that happen (definitely not available on consumer GPS). 4. That you can control whether it is radiating. If you have time to fumble with the menus then maybe airplane mode is sufficient, but if not then you need to have something more reliable like a switch. 5. That the device meets any general radiation requirements (a consumer phone is constantly sending chatter, and perhaps reducing this would be desirable for the military).
If you don't address most or all of those concerns, then you're still going to be lugging around a lot of that other gear "just in case." Or, you're going to be in a world of hurt when somebody aims an RPG at a cell tower right before they ambush you and you're now unable to call in support.. The reason those radios are so much bigger than cell phones is that a cell phone just has to reach a tower a mile or two away, and your radio is designed to reach all the way back to base or some other much more secure point. Plus, cell phones will probably never be useful for non-occupation work like capturing the city in the first place.
So, how can a drug industry (to use that example) work without patents? Copying a drug costs relatively little money, but developing one takes years and tens of millions of dollars and usually doesn't work.
The only way I can see that working without patents is if government funds it (completely - not just the original concept). I'm actually a fan of this, but I'd like to get the replacement model working before we scrap the existing system, since drugs are fairly important. Easiest way to do this is have the 1st-world government start funding end-to-end drug development, and holding the patents to themselves (licensing them for free to anybody who wants to make them). Privately-developed drugs would still be patented but would have to compete with cheaper publicly-developed drugs. Then we can see how the new model is working and the old model would pretty-much go away on its own if the public model works (nobody will buy a $5 pill if the government makes a 50 cent one, but if the government botches things up we still at least have the $5 pills we have today). Plus a gradual transition could allow for better use of the expertise we already have - government could outsource development to the existing industry (fee for service with the government holding patent rights). So, as the old model fades away scientists would just end up moving to new departments in the same companies.
You can't change the fact that drugs are expensive to develop, so taxpayers will take a hit. However, taxes are generally more progressive than medical bills so you can at least send the bills to people who can afford them.
I think that snooping is going to be a bit harder than some think unless a government is willing to sniff every connection on the network. Now, if half of the relay nodes end up on EC2 as a result of this article then that is a different story - if Amazon lets them snoop the RAM of these nodes without a warrant then they can probably get the keys to half the network.
My understanding is that most of Tor's weaknesses stem from one of its requirements - providing access to the general internet. It has many competitors that are superior in many ways, but all of them suffer from not providing access to the general internet. I think that this is going to limit their popularity until there is a lot more content, except for niche groups that simply can't post their content on the internet at all.
Yup. You don't even have to tie them to wages per se - just tie them to working conditions, environmental laws, political freedom, access to courts, and such. You could even use the proceeds to help drive improvements in these things overseas as part of foreign aid. I think that much of the cost difference overseas boils down to the fact that if a factory worker gets their arm chewed up by a machine you can hand their family $10 and pay somebody else 50 cents to clean the machine, and if in the process 5000 gallons of toxic waste leaks out you just hose it into the creek. Then if the family does get upset and sues chances are the local government will shut them up (perhaps brutally) since they're in the pocket of the company and they don't like to advertise their conditions.
So, outsource your work to Belgium and vice-versa if that makes sense economically - I'm sure that sort of thing will just balance out. The real problem is jurisdiction shopping to get around basic human rights.
Your argument assumes that most kids are capable of working in a lab, if only they were properly educated. That seems as likely to me as suggesting that if only I had spent more time working on my foul shots in school I'd be in the NBA.
Consider as an example somebody who is mentally retarded and can't tie their own shoelaces, and who has a physical disability and cannot lift ten pounds. Such a person clearly will never be a productive member of society (productive in the sense of being able to produce goods/services that can be sold and used to pay for their basic needs). Now, consider that the only difference between that guy and you is a matter of degree. If I asked you to design a working Mars lander in 15 minutes or to lift a 400 ton concrete block, you would fail, and somewhere in-between we'd find a level where one of any arbitrary pair of persons could perform and the other could not.
Now, I'm sure that if I practiced my foul shots I could probably improve a fair bit. However, I believe that most people are going to operate within a general range of ability. When I was in school I rarely spent time studying either science or math and yet I was easily a top performer in both, compared to many who worked very hard. Now, when I do spend effort I do better still, but the bottom line is that abilities are largely what you are born with, and only to a small degree what you've worked to achieve.
As society progresses and automates, the bar gets set higher and higher. That means that a larger and larger percentage of society will serve no productive purpose. Now, either we adopt fascism and just eliminate the people we don't depend on economically, or we need to change the way society works so that there is incentive to be productive, without the penalty for being non-productive being too great.
Not all industries are as broken as the software industry.
However, suppose for the sake of argument somebody really did come up with a TRULY innovative software concept - one that everybody could actually agree should receive patent protection for a year or two (I think that patent terms should be industry-specific reflecting the pace of development). No matter how clever or innovative or complex the idea is, copying something implemented in software is just a matter of copying bytes. Doing the same in hardware is a little more complicated, but only a little.
Take an industry like Pharmaceuticals. You start with a molecule, and to market it as a drug you need to answer two questions - how safe is it, and how effective is it? Doing this takes about 5-7 years if you're pushing it, and costs about a billion dollars per successful drug (counting costs wasted on drugs that turn out not to be safe or effective). Making the molecule was easy from day one (sure, some money gets spent on making the process cheaper, but it is a pittance compared to what it takes to bribe doctors to do the jobs their patients are already paying them to do). So, patents are used to give companies an incentive to test their drugs, by letting them recoup those costs. Now, I'm not saying the model is completely balanced in this industry, but without making 100% of all development (and not just coming up with the concept) government funded you simply won't get new drugs without it.
I don't see banning patents as a good idea, but I do see reigning them in across the board as being good. I'd set a term per-industry such that the public is happy that progress is getting made, and profits and non-R&D costs within the industry seem reasonable.
There isn't a single phone on the market that runs "Android" - there are dozens of different models that each run some OS that is 99% android and 1% something else (if nothing else device drivers - the open source version of Android can't actually run on any production phone).
Apple is a bit different since they sell a phone, and not an operating system.
When the Army puts this out for bid it won't be to an OS vendor - it will be to a phone vendor (yes, I know Google owns Motorola). Whether or not Android 4 addresses this issue out of the box you can bet that vendors responding to the bid will factor in the need to address this feature if it is in the RFP.
When the Military standardizes it won't be on iOS or Android - it will be on Vendor A model B. I suspect that even if they picked Apple they wouldn't be buying the consumer product per se.
You mean the function that requires browsing through menus in order to trigger? Oh, and you have no real way to know if it is working or not? If your phone's OS crashes then you end up having to pull the battery to reset it, and if you want to use it then you're stuck waiting it to boot up in non-airplane mode before you can turn off the radio.
I think the parent was thinking about a physical switch that could be flipped cutting off power to the relevant modules so that you KNOW they are off.
I'm not sure what the purpose of having smartphones is in general - I suspect that this wouldn't be for front-line use. EMCOM is pretty important if these things are going on the battlefield. However, it isn't like one minute you're transmittng and the next you're not without warning - if you are going zero-emissions then chances are you will be in that state before you ever leave the base. Now, I can see the need to have strong controls to ensure a device is in zero-emission mode when it says it is.
Slashdot Mirror
Replies made most of the good points here, but you're onto the correct solution. You need to use an encryption system such that the drive ends up in a state where you are unable to recover it yourself, at least not using any means that the attacker can discover. Of course, that could well mean that you're out the data yourself, and you'll certainly be out your computer hardware until it is so old as to be worthless.
Usually this is best implemented using hardware methods (TPM/etc) - assuming that you can trust the hardware. Hardware designed for key storage usually is tamper-resistant and is designed to permanently wipe its contents upon too many failures/etc.
As others have suggested, you need to avoid attacks on RAM as well.
The problem is that it is easy to save money if you are willing to take on risk, and corporate incentives are usually structure to encourage this.
You collect the bonus for not overrunning your budget now. Your successor might get hammered over the missing backups later, or maybe they'll pass on the blame and everybody will come out fine (but the shareholders, who don't get to make these decisions).
In hindsight the decision to not make backups pays off in every single scenario except the one where there is some huge disaster. 99% of the time there isn't a huge disaster. So, 99% of the time the decision to not make backups pays off. Then the other 1% of the time there is probably an 80% chance that the personal impact on the decision maker will not be sufficient to make planning for the 1% worthwhile.
Why do you need redundancy in the BACKUP drives? They're already redundant. Sure, you do need two sets of drives with the hard drive solution, but you need one set of drives with the tape solution so the cost of one set is effectively zero.
Now, I'll grant you that hard drives are more fragile than tapes, so perhaps some level of redundancy makes sense. I'm not sure you need raid1 for 46TB of data, however, so that you can half half of the drives fail.
Also, putting the carbon dioxide back where it belongs is actually really easy. Just don't cut the trees and put trees back where we cut them. The rest comes naturally.
Yeah, on the timescale on which all that coal formed in the first place.
By then we'll have polluted half the habitable planets on this side of the galaxy.
Agreed.
I'm all for de-externalizing things as much as possible. Tax power based on CO2 production, particulate health problems, the cost to build Yucca Mountain, the cost of maintaining a fleet of carriers and bombing the middle east into the stone age once every 20 years, or whatever you teed to tax it with. Then the market can figure out what is cheapest, while following the regulations appropriate to each type of power.
I think we'll find that some of the "expensive" forms of power generation aren't so expensive when you de-externalize them. I suspect that in the end you'd end up with lots of nuclear in the medium term, but also a much bigger push for technologies that could replace it. The "cheapest" options out there right now probably will end up being the most expensive.
Of course, taxing this stuff correctly will be hard. How many carriers does the US need if it doesn't need to invade the middle east from time to time? What is the true risk of a dam failure and what cost should be assigned to that? Risk is hard to assess - the risk of a 2008 economic meltdown was deemed to be unimaginably low (by people who had incentive to consider it low). No doubt every special interest out there would push hard to have their favorite technology declared "safe" and as such this wouldn't work. Oh, and everybody loves to define the metrics - nobody wants to count dead coal miners, the cost of the US navy, or lung cancer deaths (unless from nuclear accidents), and even regulating CO2 is highly controversial.
Completely true - well, most of the time.
NPR had a good series on problems in the healthcare system (unlike most treatments they seemed to take a holistic approach and not just find one issue and make it out to be the single thing that is driving up cost). They had a story about a doctor who didn't prescribe a CT scan for a girl who had a suspected spinal fracture. They ended up getting into a fight with the girl's dad who felt like the doctor was just cutting costs at the risk of the girl's health.
However, the doctor pointed out that he had every reason to do exactly what the father wanted. He would be paid more if he ordered the test. Nobody would dispute the test since doing so would expose them to liability. He was taking on risk of liability in the event he was wrong and she had a fracture. However, the fact was that the CT scan for a girl of her age carried a significant increase in risk of cancer much later in life, and based on his physical exam that risk was much greater than the risk that she might have an undetectable fracture. Of course, if she did get cancer later in life it could never be linked to the one test, and by then the doctor would probably be retired/dead/etc. So, the doctor was sticking his neck out, and taking on lots of personal risk, and declining the opportunity to make money, and wasting his time explaining all of this stuff to a father who would have just said yes no discussion needed to a CT scan, all because he wanted the girl to be healthy. How many doctors would do otherwise?
Antibiotics are a similar situation - the doctor can argue with their patient, and maybe lose them. Or, they can just take 30 seconds to fill out a prescription that they'll never get sued over and everybody is smiling since now when the patent gets better it will be because of what the doctor did. Happy patients lead to more patients, and more repeat business as well. Even if the doctor is employed by something like the NHS they probably have metrics and sending a patient on their way without hassle means more visits per hour. Or, even if they're completely unaccountable who wants to sit and argue with somebody all day?
I'm generally in favor of eliminating the need to get a prescription to get access to drugs, but antibiotics are one area where I'd make an exception. I don't see the role of government as protecting people from themselves. However, antibiotic abuse harms everybody and it is completely legitimate to regulate their use - probably more strictly.
I'd probably require doctors to submit a written justification for every prescription of an antibiotic that is less than 20 years old, and with stricter requirements on anything less than 10 years old (either documented testing that shows resistance to the alternatives, or an assessment which will be reviewed by a board that the patient would suffer irreversible harm if they waited for the results of that testing).
Of course, if you do this the market for new antibiotics is almost worthless (compared to being low-value which it is now). Who will spend a billion dollars working on new antibiotics only to release one and have 30 people take it in 10 years? The solution here is bounties - governments will have to decide how many new antibiotics they want and offer substantial bounties for their discovery (probably hundreds of millions of dollars), and use that money to buy the patent rights (the company has already been paid for their work). The bounties can be adjusted based on the number of candidates that are being submitted vs the number desired.
You could actually apply a similar model to other drugs, but it would get expensive (probably cheaper than what we're doing now, but this is completely socializing medicine which is of course much more expensive than having most of the costs be privately borne by patients and their employers). If it were successful enough you'd see the drug patent problem go away without even having to ban them, since patented drugs would be much more expensive than the generics bought with bounties, and companies would still have incentive to do R&D (but not marketing, etc).
That's about as convincing as a CEO claiming that he wasn't responsible for his own company's pollution, because he told his subordinates to follow the law. It isn't enough for somebody in power to tell the people they employ to follow the law - they must do due diligence to ensure they are doing so.
Or, consider a factory that hires a bunch of people for $10/hr and tells them to open/pack boxes/etc. The 5 slowest people each week lose their jobs, but the rule is that you have to wear your safety gear and be careful/etc. Of course, the supervisor doesn't punish you for not wearing your safety gear, but they do watch you to figure out if you're one of the 5 people who will lose their jobs that week. Then a worker chops their hands open with a box cutter and sues, and the company says, well, we did provide safety gloves and tell people to use them.
The company in a position of power has a duty to ensure the law is followed - anything else leads to huge problems. Goods coming from places with these kinds of problems should have 500% tariffs applied.
Agreed. Of course, to replace all the cars in a household with this technology you need to work on charging time.
Stopping every 200 miles on a road trip isn't a big deal if it takes 5-10 minutes to recharge. It would probably be a lot safer than gas so you can just plug the thing into your car and walk into the store and buy something while you wait (leaving gas pumping in this way is dangerous). A replacement battery model could also be fast if you could pull off the necessary standardization and deal with the quality issues.
If you automate/standardize it enough you could make refueling stations on interstates that are just pull-off areas where you pull up to a station, it scans your RFID, swaps your batteries, and in 30 seconds you're hitting the gas and pulling through and back onto the road.
The big win will be greatly reducing oil demand - get rid of the fighting-two-wars-all-the-time military funding and you can probably just give free electricity to everybody.
I work in a large company (not government/CIA) and if we had some kind of break-in with our systems we'd be going through all the sorts of things you suggest. And we care about turning a profit so it isn't like we just spend money for the sake of doing so. The CIO would be ticked, but he'd be brining in those consultants pronto because with something like this you have to do it right.
I'm sure some data would be lower risk than others and might not get as much scrutiny, but just determining what is important and what isn't costs money. You find a server that you think might be at risk, then you look up the contact for that server, then you trigger a 30-email conversation (for one server) as they track down the people who are REALLY responsible and have the authority to speak for it, and then you send them the 14-question survey and put the score on the big list.
Oh, and if any of the data in those systems is subject to regulation (EVERY big company is regulated in some way) you get the lawyer-types asking "what if" questions and you end up assessing more than you probably need to anyway.
As you indicated you can't just roll back production systems a few weeks once you realize the exploit is that old. Does amazon.com just forget about every order they took in the last month that they billed but didn't ship? Do they just reset their inventory to what it was a few weeks ago (an eternity in the just-in-time world)? Do they bill and re-ship orders they already filled? Do they forget about pre-orders and lose business and frustrate customers who wanted their widget on day one at the same time?
And even if you trust that it isn't an inside job you still need the army of consultants. Companies don't staff to just absorb these kinds of distractions - they run their staff at 120% on an average day. You need an army of consultants just to make sure this stuff all gets done reasonably quickly.
Then there is the distraction factor as everybody is busy doing risk assessments and reviewing data when they'd normally be doing something that actually makes the company money (ostensibly the reason they have jobs in the first place).
Yup, a breach like this is simple in the same way that backing up your computers is simple, or buying a hard drive at 0.1 cents/GB is simple. It is all simple until you have a bazillion of them, and your job is on the line.
How? In a word, patronage. It doesn't necessarily have to be government patronage.
Uh, you do realize that developing a drug (successful or not) costs about $10-$100M, and developing a successful drug requires developing about 10 unsuccessful ones first (on average)? You're talking about the better part of a billion dollars. Do you recommend setting up a paypal page for that?
Only other way that might work is to do nothing, and leave it to artists and scientists to get what they can from the first mover advantage and from keeping secrets. That's why we might prefer some kind of patronage, to pay out money in exchange for going public, and because it is very difficult to keep recipes secret.
Considering that just about every civilized country on the planet requires companies to disclose just about every detail of the process in order to even test a drug keeping it secret would be virtually impossible. Or, are you suggesting that we no longer regulate clinical trials to ensure patient safety? Oh, and are you suggesting that doctors should prescribe pills without knowing what is actually in them?
And figuring out what is in a pill isn't all that expensive - once it hits the market everybody will know what is in it. Sure, it will be harder without actually knowing the molecular entity, assuming that we manage to keep that secret, but a few weeks with an LC and an MS is all it will take most of the time. We already know that from there reverse engineering it is pretty cheap - just look at generics in India or whatever.
There are all kinds of prizes, such as the X prize.
Sure, but they tend to be hundreds of thousands of dollars to tens of millions of dollars. You need to expand that by 1-2 orders of magnitude just to come up with a single drug. Also, modern drug development turns out several new drugs per year typically, and the x-prize model is designed to encourage people to come up with some groundbreaking idea once a decade or whatever. If you need a billion dollars for ONE drug, do we really want to limit ourselves to the rate of development we would come up with donations?
They often do not create drugs themselves. They lean more on university research, and the knowledge of native peoples. Then they patent all these drugs that they did not invent, while those who actually originated the drug get nothing. They also externalize much testing, leaving it to the FDA to figure out if a drug is safe. Or they experiment on the public!
Drugs originating from native remedies aren't all that common, though they do happen (I suspect that most of those have already made it to market - at some point you start running out of newly-discovered civilizations). Ones based on university research are very common. However, typically a university tends to come up with a concept, and maybe a compound that inhibits some enzyme in a test tube. Often the original compound either has safety issues or doesn't reach the target tissue in a living person. Sometimes the original compound actually works out, but only after that billion dollars gets spent ($100M for that drug, and then $900M on the other 9 university compounds that didn't work out).
I don't dispute that much of the "creative" work happens outside of drug companies. However, the reality is that no matter how great the original ideas were, nothing is getting into the doctor's office until the clinical trials are done, and those cost the better part of a billion dollars for every drug that makes it.
As far as credit for discovery goes - it is pretty typical that if a university or whatever comes up with a compound that they'll patent it and sell the rights to a drug company. Whether the university bothers to pay their scientists well is a different matter. I suspect that compounds discovered internally in drug companies trigger some kind of employee bonuses, but if the company makes $10B over 15
The spartan and functional bit seems to be the cluster of buildings to the south of the desert paradise. Maybe that is where all the ordinary people live?
I suspect that the antenna length isn't nearly as critical for reception, especially if the transmitter is fairly powerful. ELF is one-way. If the sub needs to reply it would go to periscope depth and deploy a satellite antenna. In fact, I suspect that half of the messages that would be sent on ELF amount to "sub 1234, check satellite traffic" - anything else would have to be completely pre-planned (like "sub 1234, go with plan A"). Bitrates on ELF are very low.
From what I've seen on documentaries the US now actually has ELF antennas on large jets - they fly in circles and deploy the antenna from high altitude and it spirals down and it apparently is good enough. That is probably a lot more survivable than a huge antenna in a fixed position. Basically just another step to ensure that we don't miss out on the end of the world should the time come.
As I recall their stated purpose for Android was promoting more access to the web from mobile devices, so that they could then expand the market for their web-based technologies. I'd say that is successful whether they collect license fees for the OS or not.
As long as the Nook has a web browser then it is mission accomplished for Google. That web browser will be pointed at Google-owned sites, and based on the cookies/etc it ends up gathering it increases Google's knowledge of the consumer in a new domain and thus increases their advertising power.
I can't say whether their strategy will work out, but every $50 smartphone that is sold that doesn't run Android is a success for Android by their original goals. They turned smartphones from $500 items owned by the fairly well-to-do or for business use only to something that might sell 3 units to a family of 4.
Using a winch would probably be a lot more practical and less likely to upset the EPA. However, either approach completely misses the point. People are all finite in their capabilities, both physical and mental, and some are more limited than others. As you raise the bar people start getting excluded, and you need to figure out what to do with them.
The average slashdotter is probably still well above the bar, but the average person is probably close to the edge if not below it.
You don't have to make anybody do anything. You just apply a tax of a few hundred percent to their exports if they don't. China is welcome to pollute their streams, but they won't be allowed to sell their products in the US on the cheap if they do.
Countries that are civilized can have free access to US markets, and vice-versa.
I'm not suggesting the US should rule the world - only that US companies shouldn't be able to outsource crimes to countries where they are legal.
No need - just have a tax table with countries on one side and rates on the other. If you're Belgium you get a rate of zero, and if you're China maybe you get a rate of 300%.
There aren't that many countries so no need to "micro-manage." If a country wants to get its rate changed they can bend over backwards to convince everybody that it is justifiable.
Or I guess we can just fall back to the original proposal and tax the living daylights out of imports from anywhere without exception. Then we merely need to micro-manage our own economy.
In what way do you disagree with my usage of it? Most likely if they did pick Apple they'd get something close to the consumer model, but not quite the same. So, it is an "iPhone", but it really isn't, and that seems like an appropriate use of per se.
It might even be just an iPhone minus the EULA and warranty disclaimer paperwork in the box, but that still isn't quite the consumer model.
Per wikipedia a good synonym of "per se" is "without qualifications" - and I don't think they'll be using a consumer phone without qualifications.
I wasn't suggesting that improved communications wouldn't have benefits. I'm just saying that existing consumer devices need many tweaks to be useful in a combat situation.
Just looking at your list you need to ensure:
1. The phone has sufficient communications range to whatever it is talking to.
2. That if you depend on that phone for essential support in an ambush or whatever that whatever it is talking to is itself secure (ie not an unguarded cell tower in the middle of some contested city).
3. That the GPS can't be spoofed (probably not guaranteed on consumer GPS), and that it is ready for selective availability should that happen (definitely not available on consumer GPS).
4. That you can control whether it is radiating. If you have time to fumble with the menus then maybe airplane mode is sufficient, but if not then you need to have something more reliable like a switch.
5. That the device meets any general radiation requirements (a consumer phone is constantly sending chatter, and perhaps reducing this would be desirable for the military).
If you don't address most or all of those concerns, then you're still going to be lugging around a lot of that other gear "just in case." Or, you're going to be in a world of hurt when somebody aims an RPG at a cell tower right before they ambush you and you're now unable to call in support.. The reason those radios are so much bigger than cell phones is that a cell phone just has to reach a tower a mile or two away, and your radio is designed to reach all the way back to base or some other much more secure point. Plus, cell phones will probably never be useful for non-occupation work like capturing the city in the first place.
So, how can a drug industry (to use that example) work without patents? Copying a drug costs relatively little money, but developing one takes years and tens of millions of dollars and usually doesn't work.
The only way I can see that working without patents is if government funds it (completely - not just the original concept). I'm actually a fan of this, but I'd like to get the replacement model working before we scrap the existing system, since drugs are fairly important. Easiest way to do this is have the 1st-world government start funding end-to-end drug development, and holding the patents to themselves (licensing them for free to anybody who wants to make them). Privately-developed drugs would still be patented but would have to compete with cheaper publicly-developed drugs. Then we can see how the new model is working and the old model would pretty-much go away on its own if the public model works (nobody will buy a $5 pill if the government makes a 50 cent one, but if the government botches things up we still at least have the $5 pills we have today). Plus a gradual transition could allow for better use of the expertise we already have - government could outsource development to the existing industry (fee for service with the government holding patent rights). So, as the old model fades away scientists would just end up moving to new departments in the same companies.
You can't change the fact that drugs are expensive to develop, so taxpayers will take a hit. However, taxes are generally more progressive than medical bills so you can at least send the bills to people who can afford them.
I think that snooping is going to be a bit harder than some think unless a government is willing to sniff every connection on the network. Now, if half of the relay nodes end up on EC2 as a result of this article then that is a different story - if Amazon lets them snoop the RAM of these nodes without a warrant then they can probably get the keys to half the network.
My understanding is that most of Tor's weaknesses stem from one of its requirements - providing access to the general internet. It has many competitors that are superior in many ways, but all of them suffer from not providing access to the general internet. I think that this is going to limit their popularity until there is a lot more content, except for niche groups that simply can't post their content on the internet at all.
Yup. You don't even have to tie them to wages per se - just tie them to working conditions, environmental laws, political freedom, access to courts, and such. You could even use the proceeds to help drive improvements in these things overseas as part of foreign aid. I think that much of the cost difference overseas boils down to the fact that if a factory worker gets their arm chewed up by a machine you can hand their family $10 and pay somebody else 50 cents to clean the machine, and if in the process 5000 gallons of toxic waste leaks out you just hose it into the creek. Then if the family does get upset and sues chances are the local government will shut them up (perhaps brutally) since they're in the pocket of the company and they don't like to advertise their conditions.
So, outsource your work to Belgium and vice-versa if that makes sense economically - I'm sure that sort of thing will just balance out. The real problem is jurisdiction shopping to get around basic human rights.
Your argument assumes that most kids are capable of working in a lab, if only they were properly educated. That seems as likely to me as suggesting that if only I had spent more time working on my foul shots in school I'd be in the NBA.
Consider as an example somebody who is mentally retarded and can't tie their own shoelaces, and who has a physical disability and cannot lift ten pounds. Such a person clearly will never be a productive member of society (productive in the sense of being able to produce goods/services that can be sold and used to pay for their basic needs). Now, consider that the only difference between that guy and you is a matter of degree. If I asked you to design a working Mars lander in 15 minutes or to lift a 400 ton concrete block, you would fail, and somewhere in-between we'd find a level where one of any arbitrary pair of persons could perform and the other could not.
Now, I'm sure that if I practiced my foul shots I could probably improve a fair bit. However, I believe that most people are going to operate within a general range of ability. When I was in school I rarely spent time studying either science or math and yet I was easily a top performer in both, compared to many who worked very hard. Now, when I do spend effort I do better still, but the bottom line is that abilities are largely what you are born with, and only to a small degree what you've worked to achieve.
As society progresses and automates, the bar gets set higher and higher. That means that a larger and larger percentage of society will serve no productive purpose. Now, either we adopt fascism and just eliminate the people we don't depend on economically, or we need to change the way society works so that there is incentive to be productive, without the penalty for being non-productive being too great.
Not all industries are as broken as the software industry.
However, suppose for the sake of argument somebody really did come up with a TRULY innovative software concept - one that everybody could actually agree should receive patent protection for a year or two (I think that patent terms should be industry-specific reflecting the pace of development). No matter how clever or innovative or complex the idea is, copying something implemented in software is just a matter of copying bytes. Doing the same in hardware is a little more complicated, but only a little.
Take an industry like Pharmaceuticals. You start with a molecule, and to market it as a drug you need to answer two questions - how safe is it, and how effective is it? Doing this takes about 5-7 years if you're pushing it, and costs about a billion dollars per successful drug (counting costs wasted on drugs that turn out not to be safe or effective). Making the molecule was easy from day one (sure, some money gets spent on making the process cheaper, but it is a pittance compared to what it takes to bribe doctors to do the jobs their patients are already paying them to do). So, patents are used to give companies an incentive to test their drugs, by letting them recoup those costs. Now, I'm not saying the model is completely balanced in this industry, but without making 100% of all development (and not just coming up with the concept) government funded you simply won't get new drugs without it.
I don't see banning patents as a good idea, but I do see reigning them in across the board as being good. I'd set a term per-industry such that the public is happy that progress is getting made, and profits and non-R&D costs within the industry seem reasonable.
There isn't a single phone on the market that runs "Android" - there are dozens of different models that each run some OS that is 99% android and 1% something else (if nothing else device drivers - the open source version of Android can't actually run on any production phone).
Apple is a bit different since they sell a phone, and not an operating system.
When the Army puts this out for bid it won't be to an OS vendor - it will be to a phone vendor (yes, I know Google owns Motorola). Whether or not Android 4 addresses this issue out of the box you can bet that vendors responding to the bid will factor in the need to address this feature if it is in the RFP.
When the Military standardizes it won't be on iOS or Android - it will be on Vendor A model B. I suspect that even if they picked Apple they wouldn't be buying the consumer product per se.
You mean the function that requires browsing through menus in order to trigger? Oh, and you have no real way to know if it is working or not? If your phone's OS crashes then you end up having to pull the battery to reset it, and if you want to use it then you're stuck waiting it to boot up in non-airplane mode before you can turn off the radio.
I think the parent was thinking about a physical switch that could be flipped cutting off power to the relevant modules so that you KNOW they are off.
I'm not sure what the purpose of having smartphones is in general - I suspect that this wouldn't be for front-line use. EMCOM is pretty important if these things are going on the battlefield. However, it isn't like one minute you're transmittng and the next you're not without warning - if you are going zero-emissions then chances are you will be in that state before you ever leave the base. Now, I can see the need to have strong controls to ensure a device is in zero-emission mode when it says it is.