The current state is not a problem and in fact is good. But if the article describes their end-game accurately, this "good period" will be limited and followed by utter evil.
At which point the door will be wide open for competition to come in. At the point the predatory monopoly raises their prices they lose the only thing deterring competition... this is especially true on the Internet, where competitors are just a click away.
Amazon opting to be flat, effectively at 0% margin
No, Amazon is not operating at 0% margin. Amazon is generating healthy profits from its business. But it is choosing to reinvest those profits rather than either pay them out as dividends or pile them up in the bank.
The idea is to endure long enough to starve out competitors until monopoly acheived. At that point, rather obscene profit can be reaped.
Since the basis of your argument is factually incorrect, there's really no reason to go into this, but I want to. Your conclusion "At that point, rather obscene profit can be reaped" is a common assertion of predatory pricing theories... but one that doesn't really seem to actually exist. In the last few decades courts have added a test for predation that requires that there be a realistic prospect of recouping the losses due to predatory pricing in the ensuing period of monopoly rents. After courts adopted that test, no antitrust litigation based on a predation theory has succeeded, because it's extremely hard for anyone to rationally believe that losses will be recouped.
Why? Two reasons.
First, predatory pricing is ruinously expensive. Far more so for the would-be monopolist than its targets. This is because in order for a company to be in position to execute such a play it must already own most of the market, which means that the net effect of the below-cost prices it sets get multiplied by the volume it already does, plus the new share that it acquires from its targets.
Second, assuming the monopolist does manage to drive out all competition, and begins charging obscene prices, it becomes not only possible but very profitable for competitors to enter (or re-enter) the market. Actually, because of this, if the competitors have access to good capital markets it's unlikely that they'll be driven out of business at all, because everyone will recognize the higher prices which are to come and so capital will be available to enable the competitors to survive the price war.
So-called "post-Chicago" theorists have been able to construct some narrow scenarios in which predatory pricing is rational, but they're very narrow and ultimately boil down to circumstances in which the competitors can be successfully bluffed by the would-be monopolist.
Note that all of this is distinct from a different scenario, in which the big player isn't being anti-competitive at all, but is instead is just much more efficient, and therefore has much lower costs. That sort of "monopolist" can maintain healthy profit margins while pricing goods below what the competition can afford. This, however, is not harmful to consumers, in fact it's good for consumers. Of course there's the possibility that once the competition is driven out the monopolist will raise prices while retaining its low costs. Should that ever happen, then that would be an appropriate time for regulators to step in. I'm not aware of any real-world examples, though.
So you favor stealing money from the rest of us (That's where that subsidy money comes from) in order to support your personal decision.
Stealing money from you? That's a rather twisted way of looking at things. You're saying that my money is yours. I disagree, completely. I think 80% of the federal taxes I pay are theft from me, and I'm perfectly happy to reduce that burden in any (legal) way I can. And I still pay plenty.
I'm philosophically opposed to subsidies for EVs (or any other product). If you offer me the chance I'll vote against them. However, if you offer me a legal way to reduce the taxes I pay -- regardless of its basis -- I'll take it.
And, even without the subsidies, I'd still have bought the EV, because it still would have made financial sense!
And I am sure you rant about how evil corporations steal from the people too, don't you?
I don't think I have ever in my life ranted about evil corporations stealing from people.
I should also note that I, personally, don't care that much. I bought a LEAF not because I was anxious to save the planet, but because it's cheaper to own and operate than a gasoline-powered car, at least for my driving patterns.
That's because you're not paying any road taxes on the electricity
True, I'm not. However, even if the electricity I use had additional taxes in the same ratio as gasoline, the EV would still be cheaper.
and electric cars presently represent a very insignificant load on the electrical grid. If there were no road taxes on gasoline and demand for it was low, it would be dirt cheap too.
This argument holds no water, either. It's not like cars are the only users of electricity; there's plenty of other demand. And it's not like grid capacity can't be scaled up. Both gasoline and (most of) the various sources of energy for electricity production are at a dynamic equilibrium at roughly the standard margin; price isn't driven by any hard capacity limits, but by production costs and the cost of capital.
Granted that in the short term, a surge in EVs would stress the grid and it would take time for capacity to catch up, and at the same time gasoline prices would drop because we'd have excess production and distribution capacity. But it would re-equilibrate back to about the levels it's at now.
... they'd probably already be a lot more popular.
As it sits, you're looking at putting down an additional $10k on a car *JUST* because it's electric, while the typical price-conscious consumer looks at that money difference and realizes that they can actually just get a whole lot nicer car instead.
I suppose I'm an "atypical" price-conscious consumer, because I looked hard at the costs... and chose to buy electric because it's cheaper. Sure, if you just look at the vehicle price, gasoline-powered vehicles are less expensive, but when you factor in fuel and maintenance costs over a suitable time horizon (I chose 8 years), gasoline isn't so cheap. Even without any subsidies, I found that the break-even point was just shy of 6 years for the less-expensive EVs (LEAF, i-MiEV, etc.), at least for my driving patterns. With the federal and state tax credits available, it was two years.
Running a car on energy from the electric grid is greener than running on gasoline, even if your power comes from coal plants
To put some numbers to this, the EPA says that the average car emits 423 grams of CO2 per mile, and that the average US coal plant emits 1216 lb (551 kg) of CO2 per megawatt-hour produced, which is 551 g per kwh. My Nissan LEAF gets about four miles per kwh. Assuming pure coal power, and ignoring line losses, that means my car causes 138 grams of CO2 to be emitted per mile I drive.
Now, the LEAF is a very small, very efficient car, significantly more efficient than most gasoline-powered cars (mainly for range reasons). So comparing 138 to 423 straight up isn't a fair comparison, but even if you assume a normal car is half as efficient as the LEAF, it's still 276 grams per mile vs 423 grams per mile. Throw in some line losses and the gap closes further... but it's pretty clear that electric vehicles cause less CO2 production than gasoline vehicles, on a per-mile basis, even if all of the electricity comes from coal.
For me it's even better because although Colorado is primarily coal-powered, I mostly charge my car only at the office, and my employer (Google) pays a little extra to buy "green" power, mostly wind and hydro, I think. So my car's carbon footprint is much lower. This highlights another aspect of electric vehicles: if we switch to EVs (where appropriate -- they don't work for everything), it is at least possible to replace coal generation with something cleaner. Wind, hydro, wave, solar, nuclear, geothermal... there are lots of clean ways to generate electricity.
I should also note that I, personally, don't care that much. I bought a LEAF not because I was anxious to save the planet, but because it's cheaper to own and operate than a gasoline-powered car, at least for my driving patterns. The fact that it's cleaner is a pleasant bonus. It's also a lot of fun to drive because electric motors have awesome torque and I love how quiet it is. It's a great little car, and I'm very happy with my decision to buy it (lease it, actually... I think EV tech is changing fast enough right now that there's value in being able to upgrade regularly).
Good point. Actually, I'll bet that's where the $21M figure came from; they figure they lost that much, so WTO has authorized them to earn that much by exploiting US copyrights. Seems fair.
Am I the only person in America who is not a felon, but who believes that convicted felons should be able to have guns AND vote once they've pay their dues (with prison, or whatever), just like regular non-felon folks are able to do?
So sovereignty is pretty much admitting that at the international level the only law that really matters is the law of the jungle.
I'd say it's closer to contract law than the law of the jungle. If you don't make your car payment no one comes to kill you... they just take back your car.
Are you familiar with the phrase "proportionality"?
Yep, and I think if anything the WTO has been too gentle with the US. Our violation of our agreements has cost Antiqua and Barbudos ~$1B per year, and the WTO has only authorized them to make $21M per year from ignoring their agreement to honor our copyrights.
It's not as though the US seriously harmed Antigua by banning online gambling.
said to be recommending the establishment by the Government of Antigua & Barbuda of a statutory body to own, manage and operate the ultimate platform to be created for the monetisation or other exploitation of the suspension of American intellectual property rights authorised
Why does this press release read like an EULA? I mean that is a retarded amount of long words to describe a very simple idea. Why can't they just write it up as "We're bringing back fair use, bitches!"
Because this has nothing to do with fair use. Fair use is restricted to non-commercial, educational, etc., use. This is about unrestricted, anything-goes use.
So we aren't allowed to have any laws that might negatively impact the earnings of another nation? I'd prefer to have national sovereignty, thanks.
Sure we are. There's nothing requiring us to honor our international trade agreements. We can break 'em if we like. However, that means that other countries don't have to honor their sides of the bargains either... hence the WTO suspending US copyrights for Antigua and Barbuda. We ignore our obligations and damage them, so they can ignore their obligations to us.
No loss of sovereignty, just a consequence of what essentially boils down to international breach of contract.
The original post on this topic suggested looking for flaws in the math, which implies looking for flaws in the underlying algorithms, which is not something anyone other than a researcher is likely to be able to do.
Does this implementation match its design and goals? Compared to evaluating crypto on its own merits, that is orders of magnitude simpler, very math heavy, and well within reason for someone with calculus experience, and reading material.
Now it sounds like you're talking not about validating the crypto, but instead checking the implementations. That's a reasonable thing to do, I suppose, but with the exception of a tiny handful of public key algorithms, math isn't even relevant, and even in those cases it's not very important. What you would really be doing is looking at abstract descriptions of the algorithms and validating that the implementation does those things.
In practice, the great thing about writing crypto code is that unless it's exactly, 100% correct, it doesn't work at all; it produces garbage (ciphers and hashes, anyway; CPRNGs are a much, much harder problem). This is also the challenging thing about writing crypto code, because if your code is producing garbage you have absolutely no idea where the flaw might be. With most software, as you get closer to a correct implementation, your results get closer to what you want. With crypto code functionality is almost binary.
What that means in the current context is that if the implementation you're examining matches one or two test vectors (term of art for known input/output pairs), then the implementation is correct. That doesn't mean it's secure, though. It could leak information in any number of ways, so what you really need to do when checking an implementation is to ensure that it doesn't leak. Unfortunately, leakage can be extremely subtle, so except for checking the most obvious issues, like timing analysis, we're back in the realm of expert work. Even simple issues like timing can require deep expertise to know how the published algorithm can be modified to make it timing resistant, without compromising it in some other way.
If an input should be cryptographically random, then trust requires that it be tested scientifically. There are ways to do so, which is how the Dual_EC_DRBG prng defect was found.
Running tests for statistical randomness on intermediate and output values, with selected parameters, was how part of the Dual EC DRBG defects were found... but they were found by researchers and the randomness testing was the method used to validate the suspected weakness, after deeper knowledge pointed out the possibility. Could a lay person have made the same discovery? It's possible, but very, very unlikely.
We were talking about trusting no one, not how paranoid I might be.
And my point is that the only way to avoid trusting someone (many people, actually), is to avoid cryptography entirely. The field is too broad and deep for any one person to completely validate the tools they're using. It's a specialist field and the only way you can have any serious knowledge of whether or not the tools are secure is to trust the specialists. Or become one; and even then you'll only deeply understand one piece of the puzzle.
If you don't trust anyone, then the answer is simple: Don't put your secrets on a computer network, and perhaps you shouldn't even put them on a computer. Stick with really old technology which you can fully understand, like pencil and paper.
I never intended to suggest that my solutions are appropriate for your mom.
She should make manual backups periodically, or get you to build something... or buy something. Apple's TimeMachine is really good.
None of that has anything to do with whether or not cloud storage is reliable. It is, like everything else, somewhat reliable. It's probably a bit better than most other options (assuming a good provider), but it's not perfect. Nothing is perfect.
Well... depends on how you pick your storage provider. I use Google Drive, which is backed by the same storage used to store my gmail, and gmail has demonstrated an extremely high level of reliability. Moreover, Google has published technical details of how their distributed, replicated storage systems work, and the architecture is one that makes data loss very unlikely.
Look into how your storage provider implements and manages their system. If they don't say, well, that's a useful bit of information itself.
This is rather unfortunate for him, of course, particularly if he didn't have a backup anywhere else (duh!), but I'm sure we'll get a lot of slashdotters saying "See, this is why I'll never use the cloud!", and that's silly. Now, there are other valid reasons to avoid cloud storage (e.g. privacy and security, assuming you're not encrypting the data), but reliability really isn't one of them. Thumb drives die, get lost or get damaged, hard drives fail... there is no perfectly-reliable storage medium, but I'll posit that a good cloud storage provider has a much lower failure rate than anything you can manage yourself.
The solution, as always, is backups. Any one storage medium may fail, but the odds of several of them failing simultaneously is very low. Personally, my most important files live on a RAID-6 array with a hot spare on my home file server, and on my laptop's SSD, on my workstation's HD, and on Google Drive. There is a fair amount of low-priority stuff which lives only on Google Drive. It gets automatically synced to multiple machines, but that wouldn't help if someone else got access to my account and deleted my files (of course, I use two-factor auth). It's still better than what I'd do without a cloud service, which is that I'd have those files only on my laptop.
Hmm... It occurs to me that it'd be trivial to write a small script that uses rdiff-backup to copy the contents of my Drive folder to another folder, then run that in a cron job. Then I'd have automatic, persistent synchronization to multiple devices. I think I'll do that right now:-)
Bottom line: This is a sad story, but not a reason to avoid cloud storage. It is a reason to recommend backups. Especially completely automated, effortless backups.
If anyone is going to collect data it is going to be us! After all we are the only ones who can properly monetize it.
uProxy doesn't send data to Google. There's also a huge difference between data users send to Google as part of the deal by which they use its services and connection-level eavesdropping.
Who supported privacy measures before Snowden's revelations?
Google, for one. Google was the first major service to enable SSL for basically all of its services. Google also pushed back hard against Chinese censorship. They caved for a while, but ultimately just took their business and left mainland China because they refused to censor. Granted that Google has made some mistakes (Wifi over-capture by streetview cars, Safari DNT workaround), but they've always tried to support user privacy.
Slashdot Mirror
The current state is not a problem and in fact is good. But if the article describes their end-game accurately, this "good period" will be limited and followed by utter evil.
At which point the door will be wide open for competition to come in. At the point the predatory monopoly raises their prices they lose the only thing deterring competition... this is especially true on the Internet, where competitors are just a click away.
Amazon opting to be flat, effectively at 0% margin
No, Amazon is not operating at 0% margin. Amazon is generating healthy profits from its business. But it is choosing to reinvest those profits rather than either pay them out as dividends or pile them up in the bank.
The idea is to endure long enough to starve out competitors until monopoly acheived. At that point, rather obscene profit can be reaped.
Since the basis of your argument is factually incorrect, there's really no reason to go into this, but I want to. Your conclusion "At that point, rather obscene profit can be reaped" is a common assertion of predatory pricing theories... but one that doesn't really seem to actually exist. In the last few decades courts have added a test for predation that requires that there be a realistic prospect of recouping the losses due to predatory pricing in the ensuing period of monopoly rents. After courts adopted that test, no antitrust litigation based on a predation theory has succeeded, because it's extremely hard for anyone to rationally believe that losses will be recouped.
Why? Two reasons.
First, predatory pricing is ruinously expensive. Far more so for the would-be monopolist than its targets. This is because in order for a company to be in position to execute such a play it must already own most of the market, which means that the net effect of the below-cost prices it sets get multiplied by the volume it already does, plus the new share that it acquires from its targets.
Second, assuming the monopolist does manage to drive out all competition, and begins charging obscene prices, it becomes not only possible but very profitable for competitors to enter (or re-enter) the market. Actually, because of this, if the competitors have access to good capital markets it's unlikely that they'll be driven out of business at all, because everyone will recognize the higher prices which are to come and so capital will be available to enable the competitors to survive the price war.
So-called "post-Chicago" theorists have been able to construct some narrow scenarios in which predatory pricing is rational, but they're very narrow and ultimately boil down to circumstances in which the competitors can be successfully bluffed by the would-be monopolist.
Note that all of this is distinct from a different scenario, in which the big player isn't being anti-competitive at all, but is instead is just much more efficient, and therefore has much lower costs. That sort of "monopolist" can maintain healthy profit margins while pricing goods below what the competition can afford. This, however, is not harmful to consumers, in fact it's good for consumers. Of course there's the possibility that once the competition is driven out the monopolist will raise prices while retaining its low costs. Should that ever happen, then that would be an appropriate time for regulators to step in. I'm not aware of any real-world examples, though.
So you favor stealing money from the rest of us (That's where that subsidy money comes from) in order to support your personal decision.
Stealing money from you? That's a rather twisted way of looking at things. You're saying that my money is yours. I disagree, completely. I think 80% of the federal taxes I pay are theft from me, and I'm perfectly happy to reduce that burden in any (legal) way I can. And I still pay plenty.
I'm philosophically opposed to subsidies for EVs (or any other product). If you offer me the chance I'll vote against them. However, if you offer me a legal way to reduce the taxes I pay -- regardless of its basis -- I'll take it.
And, even without the subsidies, I'd still have bought the EV, because it still would have made financial sense!
And I am sure you rant about how evil corporations steal from the people too, don't you?
I don't think I have ever in my life ranted about evil corporations stealing from people.
I should also note that I, personally, don't care that much. I bought a LEAF not because I was anxious to save the planet, but because it's cheaper to own and operate than a gasoline-powered car, at least for my driving patterns.
That's because you're not paying any road taxes on the electricity
True, I'm not. However, even if the electricity I use had additional taxes in the same ratio as gasoline, the EV would still be cheaper.
and electric cars presently represent a very insignificant load on the electrical grid. If there were no road taxes on gasoline and demand for it was low, it would be dirt cheap too.
This argument holds no water, either. It's not like cars are the only users of electricity; there's plenty of other demand. And it's not like grid capacity can't be scaled up. Both gasoline and (most of) the various sources of energy for electricity production are at a dynamic equilibrium at roughly the standard margin; price isn't driven by any hard capacity limits, but by production costs and the cost of capital.
Granted that in the short term, a surge in EVs would stress the grid and it would take time for capacity to catch up, and at the same time gasoline prices would drop because we'd have excess production and distribution capacity. But it would re-equilibrate back to about the levels it's at now.
As it sits, you're looking at putting down an additional $10k on a car *JUST* because it's electric, while the typical price-conscious consumer looks at that money difference and realizes that they can actually just get a whole lot nicer car instead.
I suppose I'm an "atypical" price-conscious consumer, because I looked hard at the costs... and chose to buy electric because it's cheaper. Sure, if you just look at the vehicle price, gasoline-powered vehicles are less expensive, but when you factor in fuel and maintenance costs over a suitable time horizon (I chose 8 years), gasoline isn't so cheap. Even without any subsidies, I found that the break-even point was just shy of 6 years for the less-expensive EVs (LEAF, i-MiEV, etc.), at least for my driving patterns. With the federal and state tax credits available, it was two years.
Running a car on energy from the electric grid is greener than running on gasoline, even if your power comes from coal plants
To put some numbers to this, the EPA says that the average car emits 423 grams of CO2 per mile, and that the average US coal plant emits 1216 lb (551 kg) of CO2 per megawatt-hour produced, which is 551 g per kwh. My Nissan LEAF gets about four miles per kwh. Assuming pure coal power, and ignoring line losses, that means my car causes 138 grams of CO2 to be emitted per mile I drive.
Now, the LEAF is a very small, very efficient car, significantly more efficient than most gasoline-powered cars (mainly for range reasons). So comparing 138 to 423 straight up isn't a fair comparison, but even if you assume a normal car is half as efficient as the LEAF, it's still 276 grams per mile vs 423 grams per mile. Throw in some line losses and the gap closes further... but it's pretty clear that electric vehicles cause less CO2 production than gasoline vehicles, on a per-mile basis, even if all of the electricity comes from coal.
For me it's even better because although Colorado is primarily coal-powered, I mostly charge my car only at the office, and my employer (Google) pays a little extra to buy "green" power, mostly wind and hydro, I think. So my car's carbon footprint is much lower. This highlights another aspect of electric vehicles: if we switch to EVs (where appropriate -- they don't work for everything), it is at least possible to replace coal generation with something cleaner. Wind, hydro, wave, solar, nuclear, geothermal... there are lots of clean ways to generate electricity.
I should also note that I, personally, don't care that much. I bought a LEAF not because I was anxious to save the planet, but because it's cheaper to own and operate than a gasoline-powered car, at least for my driving patterns. The fact that it's cleaner is a pleasant bonus. It's also a lot of fun to drive because electric motors have awesome torque and I love how quiet it is. It's a great little car, and I'm very happy with my decision to buy it (lease it, actually... I think EV tech is changing fast enough right now that there's value in being able to upgrade regularly).
Good point. Actually, I'll bet that's where the $21M figure came from; they figure they lost that much, so WTO has authorized them to earn that much by exploiting US copyrights. Seems fair.
Am I the only person in America who is not a felon, but who believes that convicted felons should be able to have guns AND vote once they've pay their dues (with prison, or whatever), just like regular non-felon folks are able to do?
+1
So sovereignty is pretty much admitting that at the international level the only law that really matters is the law of the jungle.
I'd say it's closer to contract law than the law of the jungle. If you don't make your car payment no one comes to kill you... they just take back your car.
Are you familiar with the phrase "proportionality"?
Yep, and I think if anything the WTO has been too gentle with the US. Our violation of our agreements has cost Antiqua and Barbudos ~$1B per year, and the WTO has only authorized them to make $21M per year from ignoring their agreement to honor our copyrights.
It's not as though the US seriously harmed Antigua by banning online gambling.
You clearly didn't RTFA.
said to be recommending the establishment by the Government of Antigua & Barbuda of a statutory body to own, manage and operate the ultimate platform to be created for the monetisation or other exploitation of the suspension of American intellectual property rights authorised
Why does this press release read like an EULA? I mean that is a retarded amount of long words to describe a very simple idea. Why can't they just write it up as "We're bringing back fair use, bitches!"
Because this has nothing to do with fair use. Fair use is restricted to non-commercial, educational, etc., use. This is about unrestricted, anything-goes use.
So we aren't allowed to have any laws that might negatively impact the earnings of another nation? I'd prefer to have national sovereignty, thanks.
Sure we are. There's nothing requiring us to honor our international trade agreements. We can break 'em if we like. However, that means that other countries don't have to honor their sides of the bargains either... hence the WTO suspending US copyrights for Antigua and Barbuda. We ignore our obligations and damage them, so they can ignore their obligations to us.
No loss of sovereignty, just a consequence of what essentially boils down to international breach of contract.
The original post on this topic suggested looking for flaws in the math, which implies looking for flaws in the underlying algorithms, which is not something anyone other than a researcher is likely to be able to do.
Does this implementation match its design and goals? Compared to evaluating crypto on its own merits, that is orders of magnitude simpler, very math heavy, and well within reason for someone with calculus experience, and reading material.
Now it sounds like you're talking not about validating the crypto, but instead checking the implementations. That's a reasonable thing to do, I suppose, but with the exception of a tiny handful of public key algorithms, math isn't even relevant, and even in those cases it's not very important. What you would really be doing is looking at abstract descriptions of the algorithms and validating that the implementation does those things.
In practice, the great thing about writing crypto code is that unless it's exactly, 100% correct, it doesn't work at all; it produces garbage (ciphers and hashes, anyway; CPRNGs are a much, much harder problem). This is also the challenging thing about writing crypto code, because if your code is producing garbage you have absolutely no idea where the flaw might be. With most software, as you get closer to a correct implementation, your results get closer to what you want. With crypto code functionality is almost binary.
What that means in the current context is that if the implementation you're examining matches one or two test vectors (term of art for known input/output pairs), then the implementation is correct. That doesn't mean it's secure, though. It could leak information in any number of ways, so what you really need to do when checking an implementation is to ensure that it doesn't leak. Unfortunately, leakage can be extremely subtle, so except for checking the most obvious issues, like timing analysis, we're back in the realm of expert work. Even simple issues like timing can require deep expertise to know how the published algorithm can be modified to make it timing resistant, without compromising it in some other way.
If an input should be cryptographically random, then trust requires that it be tested scientifically. There are ways to do so, which is how the Dual_EC_DRBG prng defect was found.
Running tests for statistical randomness on intermediate and output values, with selected parameters, was how part of the Dual EC DRBG defects were found... but they were found by researchers and the randomness testing was the method used to validate the suspected weakness, after deeper knowledge pointed out the possibility. Could a lay person have made the same discovery? It's possible, but very, very unlikely.
We were talking about trusting no one, not how paranoid I might be.
And my point is that the only way to avoid trusting someone (many people, actually), is to avoid cryptography entirely. The field is too broad and deep for any one person to completely validate the tools they're using. It's a specialist field and the only way you can have any serious knowledge of whether or not the tools are secure is to trust the specialists. Or become one; and even then you'll only deeply understand one piece of the puzzle.
If you don't trust anyone, then the answer is simple: Don't put your secrets on a computer network, and perhaps you shouldn't even put them on a computer. Stick with really old technology which you can fully understand, like pencil and paper.
Do you mean this book, by Tom Vanderbilt: https://play.google.com/store/books/details/Tom_Vanderbilt_Traffic?id=O4lsPQttQ5IC
You don't see the flaw their? really?
Did you miss the last sentence?
I never intended to suggest that my solutions are appropriate for your mom.
She should make manual backups periodically, or get you to build something... or buy something. Apple's TimeMachine is really good.
None of that has anything to do with whether or not cloud storage is reliable. It is, like everything else, somewhat reliable. It's probably a bit better than most other options (assuming a good provider), but it's not perfect. Nothing is perfect.
The problem is: how do you know it's good?
Well... depends on how you pick your storage provider. I use Google Drive, which is backed by the same storage used to store my gmail, and gmail has demonstrated an extremely high level of reliability. Moreover, Google has published technical details of how their distributed, replicated storage systems work, and the architecture is one that makes data loss very unlikely.
Look into how your storage provider implements and manages their system. If they don't say, well, that's a useful bit of information itself.
I mentioned that privacy/security are potential concerns with cloud storage. That's a separate issue from the one I was discussing.
This is rather unfortunate for him, of course, particularly if he didn't have a backup anywhere else (duh!), but I'm sure we'll get a lot of slashdotters saying "See, this is why I'll never use the cloud!", and that's silly. Now, there are other valid reasons to avoid cloud storage (e.g. privacy and security, assuming you're not encrypting the data), but reliability really isn't one of them. Thumb drives die, get lost or get damaged, hard drives fail... there is no perfectly-reliable storage medium, but I'll posit that a good cloud storage provider has a much lower failure rate than anything you can manage yourself.
The solution, as always, is backups. Any one storage medium may fail, but the odds of several of them failing simultaneously is very low. Personally, my most important files live on a RAID-6 array with a hot spare on my home file server, and on my laptop's SSD, on my workstation's HD, and on Google Drive. There is a fair amount of low-priority stuff which lives only on Google Drive. It gets automatically synced to multiple machines, but that wouldn't help if someone else got access to my account and deleted my files (of course, I use two-factor auth). It's still better than what I'd do without a cloud service, which is that I'd have those files only on my laptop.
Hmm... It occurs to me that it'd be trivial to write a small script that uses rdiff-backup to copy the contents of my Drive folder to another folder, then run that in a cron job. Then I'd have automatic, persistent synchronization to multiple devices. I think I'll do that right now :-)
Bottom line: This is a sad story, but not a reason to avoid cloud storage. It is a reason to recommend backups. Especially completely automated, effortless backups.
See the previous /. articles.
If anyone is going to collect data it is going to be us! After all we are the only ones who can properly monetize it.
uProxy doesn't send data to Google. There's also a huge difference between data users send to Google as part of the deal by which they use its services and connection-level eavesdropping.
Who supported privacy measures before Snowden's revelations?
Google, for one. Google was the first major service to enable SSL for basically all of its services. Google also pushed back hard against Chinese censorship. They caved for a while, but ultimately just took their business and left mainland China because they refused to censor. Granted that Google has made some mistakes (Wifi over-capture by streetview cars, Safari DNT workaround), but they've always tried to support user privacy.
The TOR issues are mostly Javascript
No, the TOR issue is that the NSA runs (or has compromised) a large percentage of the exit nodes.
They (Google) wrote the software, right?
Nope. RTFS.
you've got that backwards... tablet's ends before the user's
See why UI design is so tricky?