On the one hand, I'm highly skeptical that it's actually true. I mean, it's not as though the dealer is the only place you can go for car servicing even now. In fact, going to the dealer is generally considered to be what suckers do. And in terms of complexity, cars are likely peaking out--as we transition to electric, they'll get mechanically simpler. Whether the electronics are serviceable (software lockout, etc.) is an interesting question, but the only way to make independent service operations viable is to mandate that manufacturers make replacement boards and diagnostic manuals available. If we don't, manufacturers will be able to software "lock out" everybody no matter what our sales channels look like.
Even assuming it's the case that we need some mandate to make it happen, there's no reason for car *sales* to go through that channel. You just need to mandate that manufacturers make parts and manuals available to third parties on reasonable terms. The reality is that the dealers don't seem to be making much money on sales anyway these days. It's all through servicing or financing. And in exchange for that, we get a miserable, sleazy buying experience that is widely recognized to be the worst in any industry or market. So it seems clear that whatever we're trying to do, we're doing it wrong.
It seems like the simplest solution would be a ban on manufacturers offering services and parts directly to the public. That way there would be a profitable market for parts and technical manuals, but the manufacturers wouldn't have an incentive to kill off the competition. The service operations would just be customers instead of competition.
Answer: Some middle men add value and some simply extract rents. Distributing and warehousing a wide variety of perishable vegetables for purchase on demand at random times is a valuable and complex service. Having a showroom to poke at cars in person adds value, but basically *everything* else a dealer does is either overpriced compared to alternatives or actively subtracts value from the car buying experience.
...as demand for gas burning cars drops, pinching margins on the gasoline side. At some point in the future, nobody will be buying gas powered cars at all, so dealers won't have anything to compare the electric car margins to at all.
I suppose it's possible that there's a different reality, but it seems to me like we have two major possibilities:
1) Russia, which has a record of being increasingly aggressive about airspace, allowed its aircraft into Turkish airspace despite complaints and warnings and got shot down.
2) Turkey saw some advantage in shooting down Russian aircraft and antagonizing Russia for no reason.
I suppose there's also various forms of incompetence that could be dumped into possibility 3. But really, what does Turkey have to gain from intentionally shooting down Russian aircraft outside its borders? Is there some eleven dimensional chess operation going on here that I'm not seeing?
That is the problem with handing weapons over to "moderates" and hoping they'll be tenacious enough to win a civil war for us, isn't it? At best, it's the, "We'll fight to the end, kill everybody and scorch the Earth," faction against the, "We wish you guys weren't so militant," faction. Pouring crates of weapons in is just going to continue to arm the people who are aggressive and committed enough to seize them.
That's really just a trick for the ultra-rich, though. The vast majority of people, even wealthy people, don't do that and can't do it very easily. For most people, at some point, assets need to come out of the corporation and go into the hands of a person for that person to enjoy them, and those show up on the tax forms. If their income isn't being taxed to begin with, no tricks we do trying to tax corporations is going to help that situation. If anything, the corporations are even better at tax avoidance.
It seems like if we had any sense at all, we'd immediately dump the corporate income tax and replace it with a revenue-neutral increase in the capital gains and dividend taxes. The corporate shareholders ultimately end up paying any dollars that get paid anyway, and humans are much easier to tax than corporations are. A corporation is a shape-shifting non-entity that can "spend a year dead for tax purposes," so trying to change the laws fast enough to get any revenue out of them is a losing battle. All we end up doing is giving them an incentive to do ridiculous things like hold money in foreign accounts and set up subsidiaries all over the world to move revenue around. It's great for the tax lawyers and financial consultants, but it doesn't really get us any real revenue. It's the tax enforcement equivalent of the drug war.
There are different levels of belief, though. Is it just as dogmatic to assert that unicorns don't exist as it is to assert that they do? I'm pretty certain that they don't exist, but if I saw one, I'd be perfectly happy to change my mind. Likewise, I'm an atheist because I don't see any compelling evidence to believe in a god, but if such evidence presented itself, I'd be willing to change my mind. Is that still a dogma or just a general belief that the preponderance of evidence points in one direction?
I think that the creation of an "agnostic" category implicitly overstates the certainty with which atheists disbelieve. I've know a lot of atheists, but I've never known any who disbelieve in a god as firmly and certainly as the more sincere Christians I have known. It's fairly easy to find a religious person who will say, "I believe this with 100% certainty, and nothing I see will ever change that." I've never met an atheist who took that position.
Which is faulty logic if applied to its conclusion. If we're all guilty then we'll all be locked away in some prison somewhere, and you don't "rule" someone in prison.
No, if we're all guilty then we could be locked away in some prison somewhere, but for the benevolent lenience of our watchers. So long as you don't piss them off. That's the point.
If the law worked the way you suggest (everybody who commits a crime is automatically caught and impartially sentenced according to the law), things would be fine. Bad laws would be taken off the books and we'd all be safe and happy. The problem arises when you can make something illegal and then only charge people some of the time.
What you said is actually a misconception of H1-B visa that is very common for those who do not really know much about the visa. A H1-B holder CAN change his/her employer at any time while holding a visa without the need to let the current employer know.
I'm aware of this, but they still do need to find a new employer who is willing to sponsor an H1-B, which is a hurdle that a US worker doesn't have. It still puts H1-B employees at a competitive disadvantage, and in a tight job market, that advantage can be huge.
Also, auctioning the visa will create another issue later on. If you think that big companies/corporations will not find a way to work around the system, you have too much trust on them.
How would they abuse it, specifically? This is a really straightforward economic question: What would a giant company do to mess up this particular market for everybody else that they don't do with every other market?
Besides, how would small companies (which is the main idea in TFA) compete with bigger companies/corporations for the visa price anyway?
The same way they compete with bigger companies for everything else you buy. With money. Microsoft and Google are huge, but they don't hire all of the employees or buy up all of the computers. They acquire what makes economic sense for them to acquire. Smaller companies do the same.
Another issue with your idea is that it would result in most if not all of the H1-B holders would be in technology.
A couple of things: First, that's not necessarily the case. It's very easy to create "classes" of visas for different industries if that's an important problem to solve. That would provide even clearer information on where the shortages are. Second, if the purpose of the H1-B program is to bring in valuable employees in industries where there is a real shortage, there's not a very good economic argument for bringing them in for low-wage jobs.
TFA is actually talking about how big companies/corporations abuse the visa, NOT about what's wrong with the visa.
There are a bunch of problems with the visa, and they're largely tied to how easy it is to abuse. The key ones are:
1) A lottery system assigns a valuable resource randomly instead of buy actual value. A lucky bonehead can get a visa when an incredibly valuable genius doesn't, which defeats the whole point.
2) The whole job description / prevailing wage system is total nonsense and easily gamed.
3) Whether there's a "shortage" of workers or not is entirely up to debate instead of actual pricing data.
4) Employees on a visa are at a competitive disadvantage because it's harder for them to change jobs, depressing wages.
All of those problems go away automatically (or are at least mitigated, as in [4]) without any additional regulation (and with the removal of a ton of existing red tape) if you just let companies and people buy and sell visa slots. We don't set a up lottery and bureaucracy up to allocate corn or rice. There's really no need to do it with work visas.
I think the overall point is that if you quantize the point in high dimensional space aggressively enough and then hash that value, you're in business. The problem is designing the features such that you can do the quantizing without creating a bunch of collisions. Unfortunately for fingerprinting, that's a tall order. You're limited to metrics that are invariant over the plastic deformation of the fingerprint as you mush it against the sensor. People would be surprised at the number of different ways a typical user can find to smash his finger on a flat surface.
Iris scanning works somewhat in the visible range, but you really want to be working in the near IR range for good results (especially with brown eyes). It's also tough to get enough pixels across the iris with even fairly high res face capture. Definitely not with the selfie camera on the same side as the screen.
But you're right, retinal scanning has been basically dead for a long time.
Along those lines: Something I didn't mention in my original post is that private employees should be allowed to buy the visas themselves as well. If you have your own visa, you can take it with you and never have to worry about whether the company has the extra cash to sponsor you, and you're protected from fluctuations in the market rate for the visa. If not, the company will have to acquire and hold one for you while you work there. That would do at least something to mitigate the "slave labor" situation that currently exists with the H1-B program.
Yes. Determining "market rate" for a worker a company desperately wants to underpay is always the tricky part. Just let companies buy and sell the visas and you'll get an actual market rate right there for you in black and white. No need to write a "market rate" regulation, and then a bunch of additional regulations for how to determine market rate, and then a bunch of additional regulations to put patches around the last set of regulations once companies figure out how to game your definition of "market rate." If you have an actual market, a "market rate" is a real thing that you can just look up.
My proposal is simpler: Just auction off the visas and allow them to be traded on a secondary market. If it's typical to underpay an H1-B holder by $10K per year, that $10K gets built into the price of the visa automatically, removing the incentive. The market price would also be a useful indicator of actual labor shortages (i.e. if companies are paying the equivalent of a $50K per year premium, it's a sign that there really is a shortage of a particular skill set). Having visas with different expiration dates on the market would even provide a skills "yield curve" that you could plot to get in idea of future demand for different skills. It also guarantees that scarce visa slots go to the most valuable workers.
It wouldn't exactly be a visa at that point so much as fungible work authorization token, but I don't think that would be the end of the world. Markets are very good at solving these sorts of problems.
I doubt it's about the 10K being cheaper than the 56K. In my experience, it's more likely that the 10K was in inventory at the factory and the 56K wasn't, so they used it rather than waiting for parts to come in. Bad inventory management combined with not giving a shit can be a real problem. I bet that if you randomly sampled a year's worth of that company's products in production, you'd find all sorts of inconsistent part substitutions.
Yes, intentional abuse, unintentional abuse, simple mistakes, human or machine error, and all manner of things happen in intelligence work. And those errors are such a vanishingly small proportion of what NSA does that it is nearly zero -- and they are still taken seriously.
I have no doubt that the majority of the uses of the data are perfectly legitimate, but it seems to me that "taken seriously" is a bit of an overstatement here. Unless something has changed fairly recently, I think we have good reason to suspect that the generally good behavior of NSA employees has more to do with the fact that most people are decent and honest than with detailed oversight. A couple of disturbing things from that report:
1) While the number of "substantiated" abuses appears to be small, it seems like the cases that were substantiated were caught more out of good luck than through the inevitable grinding gears of ubiquitous oversight. That makes me wonder if those 12 cases were really most of the story or if they were just the fruit that hung low enough to pick with the tools they have. Given the details of the stories, I suspect that we're not even picking low hanging fruit in these investigations. Just harvesting what's on the ground.
2) "Written warnings" to people found to have abused the system sounds pretty thin. Perhaps the story needs more details, but it seems hard to come up with an example of "abuse" that shouldn't lead a loss of a security clearance.
When oversight of people with powerful tools comes up, we always hear a lot of rhetoric about how they're already hamstrung and they're really honorable and it's only just a few bad apples and we'll just have to deal with that. The same song and dance comes out whenever people ask questions about abuse of authority by police. Just a few years ago, nobody with the power to do anything seemed to believe that the police could possibly do nefarious things and then use their authority and general lack of transparency to cover them up. Thanks to ubiquitous cell phone and body cameras, we're starting to realize that people are people, and they'll often do whatever they can get away with.
I'm willing to believe that the NSA's record is much better than that of the average police force, but I'm also inclined to believe that we're still at the very early stages of getting the whole story on abuses.
That is in no way, shape, or form akin to saying, "you have nothing to fear if you have nothing to hide." It is not making an argument that the government "should" have your data. It is saying that the Intelligence Community, in the form of the foreign intelligence agencies, does not want your data -- doesn't want to touch it, doesn't want to see it, doesn't want to read it, whether it's encrypted or not.
So, practically speaking, what does that mean? If we're all in agreement that the intelligence community doesn't want access to my data, but they do want the tools to be able to read the data of foreigners who use the same types of systems I use, some practical problems fall out of that:
1) It means that anybody who isn't a member of the US Intelligence Community who does want my data would likely have access to it through the same channels.
2) I have only the assurances of a group of people who are not particularly transparent that they aren't accessing my data anyway.
Is this the one place where embedded automotive code has no traceability? Nobody knows who checked in the mods, who approved the changes, and whether that approval is traceable back to a defect/requirement/change request? Was I totally misunderstanding how the automotive industry handled its microcontroller firmware? It seems to me "Where did this code come from, who approved it, and what was the justification?" should be just about the easiest questions in the world for this type of engineering shop to answer. Maybe things could get sketchy when the manager who approved it points a finger at unwritten orders from his superior, but until that point, the paper trail should be completely clear. Right?
Your instincts about the article smelling kind of skechy are almost certainly right. Yes, that's what your source says happened, but as you note, they don't back it up with anything. If there's one thing I've learned, it's that anti-GMO activist groups are, if anything, even less trustworthy than international megacorporations when it comes to spinning the truth, omitting important factors, or just making stuff up from whole cloth. They're up there with creationists and anti-vaxers when it comes to needing to follow up on the primary documents for every claim they make.
If an article quotes those activist groups and they phrase something in such a way as to "not exclude" what they want you to think but not to actually come out and say it, it's usually not a real thing. If it was, they'd be pounding the drum and saying it outright and stating the facts clearly. My guess is when you hear meaningless phrases like "Monsanto went after" instead of "Monsanto threatened/filed suit against" what they really mean is that an investigator went to the farmer and asked if they were saving unlicensed GMO seeds, didn't find evidence of a violation, and then closed the case.
From what I've actually been able to verify, actual actions against farmers are extremely rare. Only a handful have actually gone to court, and the cases I've followed up on by reading the court decisions have been obviously one-sided with the farmer obviously intentionally violating the rules. The fact that when they're asked for specific cases, their big figurehead "victim" is usually Percy Schmeiser (side note: This is Monsanto's web site summarizing the situation and they link to the relevant decisions, which should tell us something) is an indicator that there isn't much in the way of real collateral damage here.
I'm generally pretty quick to believe accusations against big corporations because they're very often true. Unfortunately, the anti-GMO lobby has done so much to burn my trust that I'd take a peek outside if they told me the sky was blue. Will Saletan at Slate has a good summary that just scratches the surface of the whole mess here.
Are there corn or soybean compatibility issues I'm not aware of? Because I'm pretty sure Microsoft held on to its monopoly because people who used other software had a hard time inter-operating with the dominant software. Is there something about most farmers growing one type of corn that makes it too difficult for some farmers to grow another type of corn?
Slashdot Mirror
On the one hand, I'm highly skeptical that it's actually true. I mean, it's not as though the dealer is the only place you can go for car servicing even now. In fact, going to the dealer is generally considered to be what suckers do. And in terms of complexity, cars are likely peaking out--as we transition to electric, they'll get mechanically simpler. Whether the electronics are serviceable (software lockout, etc.) is an interesting question, but the only way to make independent service operations viable is to mandate that manufacturers make replacement boards and diagnostic manuals available. If we don't, manufacturers will be able to software "lock out" everybody no matter what our sales channels look like.
Even assuming it's the case that we need some mandate to make it happen, there's no reason for car *sales* to go through that channel. You just need to mandate that manufacturers make parts and manuals available to third parties on reasonable terms. The reality is that the dealers don't seem to be making much money on sales anyway these days. It's all through servicing or financing. And in exchange for that, we get a miserable, sleazy buying experience that is widely recognized to be the worst in any industry or market. So it seems clear that whatever we're trying to do, we're doing it wrong.
It seems like the simplest solution would be a ban on manufacturers offering services and parts directly to the public. That way there would be a profitable market for parts and technical manuals, but the manufacturers wouldn't have an incentive to kill off the competition. The service operations would just be customers instead of competition.
Answer: Some middle men add value and some simply extract rents. Distributing and warehousing a wide variety of perishable vegetables for purchase on demand at random times is a valuable and complex service. Having a showroom to poke at cars in person adds value, but basically *everything* else a dealer does is either overpriced compared to alternatives or actively subtracts value from the car buying experience.
...as demand for gas burning cars drops, pinching margins on the gasoline side. At some point in the future, nobody will be buying gas powered cars at all, so dealers won't have anything to compare the electric car margins to at all.
I suppose it's possible that there's a different reality, but it seems to me like we have two major possibilities:
1) Russia, which has a record of being increasingly aggressive about airspace, allowed its aircraft into Turkish airspace despite complaints and warnings and got shot down.
2) Turkey saw some advantage in shooting down Russian aircraft and antagonizing Russia for no reason.
I suppose there's also various forms of incompetence that could be dumped into possibility 3. But really, what does Turkey have to gain from intentionally shooting down Russian aircraft outside its borders? Is there some eleven dimensional chess operation going on here that I'm not seeing?
That is the problem with handing weapons over to "moderates" and hoping they'll be tenacious enough to win a civil war for us, isn't it? At best, it's the, "We'll fight to the end, kill everybody and scorch the Earth," faction against the, "We wish you guys weren't so militant," faction. Pouring crates of weapons in is just going to continue to arm the people who are aggressive and committed enough to seize them.
That's really just a trick for the ultra-rich, though. The vast majority of people, even wealthy people, don't do that and can't do it very easily. For most people, at some point, assets need to come out of the corporation and go into the hands of a person for that person to enjoy them, and those show up on the tax forms. If their income isn't being taxed to begin with, no tricks we do trying to tax corporations is going to help that situation. If anything, the corporations are even better at tax avoidance.
Thanks. Somebody notices it and comments maybe once a year. I should hand out prizes to the people who figure it out.
It seems like if we had any sense at all, we'd immediately dump the corporate income tax and replace it with a revenue-neutral increase in the capital gains and dividend taxes. The corporate shareholders ultimately end up paying any dollars that get paid anyway, and humans are much easier to tax than corporations are. A corporation is a shape-shifting non-entity that can "spend a year dead for tax purposes," so trying to change the laws fast enough to get any revenue out of them is a losing battle. All we end up doing is giving them an incentive to do ridiculous things like hold money in foreign accounts and set up subsidiaries all over the world to move revenue around. It's great for the tax lawyers and financial consultants, but it doesn't really get us any real revenue. It's the tax enforcement equivalent of the drug war.
Frustrated NSA Now Forced To Rely On Mass Surveillance Programs That Haven’t Come To Light Yet
There are different levels of belief, though. Is it just as dogmatic to assert that unicorns don't exist as it is to assert that they do? I'm pretty certain that they don't exist, but if I saw one, I'd be perfectly happy to change my mind. Likewise, I'm an atheist because I don't see any compelling evidence to believe in a god, but if such evidence presented itself, I'd be willing to change my mind. Is that still a dogma or just a general belief that the preponderance of evidence points in one direction?
I think that the creation of an "agnostic" category implicitly overstates the certainty with which atheists disbelieve. I've know a lot of atheists, but I've never known any who disbelieve in a god as firmly and certainly as the more sincere Christians I have known. It's fairly easy to find a religious person who will say, "I believe this with 100% certainty, and nothing I see will ever change that." I've never met an atheist who took that position.
No, if we're all guilty then we could be locked away in some prison somewhere, but for the benevolent lenience of our watchers. So long as you don't piss them off. That's the point.
If the law worked the way you suggest (everybody who commits a crime is automatically caught and impartially sentenced according to the law), things would be fine. Bad laws would be taken off the books and we'd all be safe and happy. The problem arises when you can make something illegal and then only charge people some of the time.
I'm aware of this, but they still do need to find a new employer who is willing to sponsor an H1-B, which is a hurdle that a US worker doesn't have. It still puts H1-B employees at a competitive disadvantage, and in a tight job market, that advantage can be huge.
How would they abuse it, specifically? This is a really straightforward economic question: What would a giant company do to mess up this particular market for everybody else that they don't do with every other market?
The same way they compete with bigger companies for everything else you buy. With money. Microsoft and Google are huge, but they don't hire all of the employees or buy up all of the computers. They acquire what makes economic sense for them to acquire. Smaller companies do the same.
A couple of things: First, that's not necessarily the case. It's very easy to create "classes" of visas for different industries if that's an important problem to solve. That would provide even clearer information on where the shortages are. Second, if the purpose of the H1-B program is to bring in valuable employees in industries where there is a real shortage, there's not a very good economic argument for bringing them in for low-wage jobs.
There are a bunch of problems with the visa, and they're largely tied to how easy it is to abuse. The key ones are:
1) A lottery system assigns a valuable resource randomly instead of buy actual value. A lucky bonehead can get a visa when an incredibly valuable genius doesn't, which defeats the whole point.
2) The whole job description / prevailing wage system is total nonsense and easily gamed.
3) Whether there's a "shortage" of workers or not is entirely up to debate instead of actual pricing data.
4) Employees on a visa are at a competitive disadvantage because it's harder for them to change jobs, depressing wages.
All of those problems go away automatically (or are at least mitigated, as in [4]) without any additional regulation (and with the removal of a ton of existing red tape) if you just let companies and people buy and sell visa slots. We don't set a up lottery and bureaucracy up to allocate corn or rice. There's really no need to do it with work visas.
I think the overall point is that if you quantize the point in high dimensional space aggressively enough and then hash that value, you're in business. The problem is designing the features such that you can do the quantizing without creating a bunch of collisions. Unfortunately for fingerprinting, that's a tall order. You're limited to metrics that are invariant over the plastic deformation of the fingerprint as you mush it against the sensor. People would be surprised at the number of different ways a typical user can find to smash his finger on a flat surface.
Iris scanning works somewhat in the visible range, but you really want to be working in the near IR range for good results (especially with brown eyes). It's also tough to get enough pixels across the iris with even fairly high res face capture. Definitely not with the selfie camera on the same side as the screen.
But you're right, retinal scanning has been basically dead for a long time.
Along those lines: Something I didn't mention in my original post is that private employees should be allowed to buy the visas themselves as well. If you have your own visa, you can take it with you and never have to worry about whether the company has the extra cash to sponsor you, and you're protected from fluctuations in the market rate for the visa. If not, the company will have to acquire and hold one for you while you work there. That would do at least something to mitigate the "slave labor" situation that currently exists with the H1-B program.
Yes. Determining "market rate" for a worker a company desperately wants to underpay is always the tricky part. Just let companies buy and sell the visas and you'll get an actual market rate right there for you in black and white. No need to write a "market rate" regulation, and then a bunch of additional regulations for how to determine market rate, and then a bunch of additional regulations to put patches around the last set of regulations once companies figure out how to game your definition of "market rate." If you have an actual market, a "market rate" is a real thing that you can just look up.
My proposal is simpler: Just auction off the visas and allow them to be traded on a secondary market. If it's typical to underpay an H1-B holder by $10K per year, that $10K gets built into the price of the visa automatically, removing the incentive. The market price would also be a useful indicator of actual labor shortages (i.e. if companies are paying the equivalent of a $50K per year premium, it's a sign that there really is a shortage of a particular skill set). Having visas with different expiration dates on the market would even provide a skills "yield curve" that you could plot to get in idea of future demand for different skills. It also guarantees that scarce visa slots go to the most valuable workers.
It wouldn't exactly be a visa at that point so much as fungible work authorization token, but I don't think that would be the end of the world. Markets are very good at solving these sorts of problems.
I doubt it's about the 10K being cheaper than the 56K. In my experience, it's more likely that the 10K was in inventory at the factory and the 56K wasn't, so they used it rather than waiting for parts to come in. Bad inventory management combined with not giving a shit can be a real problem. I bet that if you randomly sampled a year's worth of that company's products in production, you'd find all sorts of inconsistent part substitutions.
I haven't seen any that are smaller, cheaper, or more reliable than a resistor.
Let me guess: Mankiw?
At least it's a good textbook. I didn't sell mine back, and not just because old editions are completely worthless.
I have no doubt that the majority of the uses of the data are perfectly legitimate, but it seems to me that "taken seriously" is a bit of an overstatement here. Unless something has changed fairly recently, I think we have good reason to suspect that the generally good behavior of NSA employees has more to do with the fact that most people are decent and honest than with detailed oversight. A couple of disturbing things from that report:
1) While the number of "substantiated" abuses appears to be small, it seems like the cases that were substantiated were caught more out of good luck than through the inevitable grinding gears of ubiquitous oversight. That makes me wonder if those 12 cases were really most of the story or if they were just the fruit that hung low enough to pick with the tools they have. Given the details of the stories, I suspect that we're not even picking low hanging fruit in these investigations. Just harvesting what's on the ground.
2) "Written warnings" to people found to have abused the system sounds pretty thin. Perhaps the story needs more details, but it seems hard to come up with an example of "abuse" that shouldn't lead a loss of a security clearance.
When oversight of people with powerful tools comes up, we always hear a lot of rhetoric about how they're already hamstrung and they're really honorable and it's only just a few bad apples and we'll just have to deal with that. The same song and dance comes out whenever people ask questions about abuse of authority by police. Just a few years ago, nobody with the power to do anything seemed to believe that the police could possibly do nefarious things and then use their authority and general lack of transparency to cover them up. Thanks to ubiquitous cell phone and body cameras, we're starting to realize that people are people, and they'll often do whatever they can get away with.
I'm willing to believe that the NSA's record is much better than that of the average police force, but I'm also inclined to believe that we're still at the very early stages of getting the whole story on abuses.
So, practically speaking, what does that mean? If we're all in agreement that the intelligence community doesn't want access to my data, but they do want the tools to be able to read the data of foreigners who use the same types of systems I use, some practical problems fall out of that:
1) It means that anybody who isn't a member of the US Intelligence Community who does want my data would likely have access to it through the same channels.
2) I have only the assurances of a group of people who are not particularly transparent that they aren't accessing my data anyway.
Is this the one place where embedded automotive code has no traceability? Nobody knows who checked in the mods, who approved the changes, and whether that approval is traceable back to a defect/requirement/change request? Was I totally misunderstanding how the automotive industry handled its microcontroller firmware? It seems to me "Where did this code come from, who approved it, and what was the justification?" should be just about the easiest questions in the world for this type of engineering shop to answer. Maybe things could get sketchy when the manager who approved it points a finger at unwritten orders from his superior, but until that point, the paper trail should be completely clear. Right?
Right?
Your instincts about the article smelling kind of skechy are almost certainly right. Yes, that's what your source says happened, but as you note, they don't back it up with anything. If there's one thing I've learned, it's that anti-GMO activist groups are, if anything, even less trustworthy than international megacorporations when it comes to spinning the truth, omitting important factors, or just making stuff up from whole cloth. They're up there with creationists and anti-vaxers when it comes to needing to follow up on the primary documents for every claim they make.
If an article quotes those activist groups and they phrase something in such a way as to "not exclude" what they want you to think but not to actually come out and say it, it's usually not a real thing. If it was, they'd be pounding the drum and saying it outright and stating the facts clearly. My guess is when you hear meaningless phrases like "Monsanto went after" instead of "Monsanto threatened/filed suit against" what they really mean is that an investigator went to the farmer and asked if they were saving unlicensed GMO seeds, didn't find evidence of a violation, and then closed the case.
From what I've actually been able to verify, actual actions against farmers are extremely rare. Only a handful have actually gone to court, and the cases I've followed up on by reading the court decisions have been obviously one-sided with the farmer obviously intentionally violating the rules. The fact that when they're asked for specific cases, their big figurehead "victim" is usually Percy Schmeiser (side note: This is Monsanto's web site summarizing the situation and they link to the relevant decisions, which should tell us something) is an indicator that there isn't much in the way of real collateral damage here.
I'm generally pretty quick to believe accusations against big corporations because they're very often true. Unfortunately, the anti-GMO lobby has done so much to burn my trust that I'd take a peek outside if they told me the sky was blue. Will Saletan at Slate has a good summary that just scratches the surface of the whole mess here.
Are there corn or soybean compatibility issues I'm not aware of? Because I'm pretty sure Microsoft held on to its monopoly because people who used other software had a hard time inter-operating with the dominant software. Is there something about most farmers growing one type of corn that makes it too difficult for some farmers to grow another type of corn?