The wiring harness will be about as complex for an EV as it is for an ICE.
Why would it be? An electric motor either is getting and drawing power or it isn't. There should be approximately two large power bus wires going to the electric motor, and maybe a couple of wires for an RPM encoder, if they bother putting one in, and depending on where the power switching is located, possibly a couple of control wires. By contrast an ICE design has wires from the computer to dozens of emissions control sensors, speed sensors, temperature sensors, etc., plus wires that allow the computer to control the fuel-air mixture and various aspects of engine timing, control the automatic transmission, etc., plus wires to allow the computer to pressurize the gas tank and sensors to verify that it isn't leaking, plus wires to the relay that powers the fuel pump. So I would expect the wiring harness on an EV to be dramatically simpler than the wiring harness for an ICE-based car. If it isn't, I'd be led to wonder whether the design engineers were getting kickbacks from wire manufacturers....
Re-implementing an API and wholesale lifting a GPL software package are not even vaguely related.
No, they're vaguely related. They both involve taking and using something that you didn't create. Of course, they're related in much the same way that bumming a stick of chewing gum and stealing a car are related....
If the connection is not worth $20k, then why is it magically worth it when someone else is paying for it?
Because the only real alternative to everyone else paying for it is for farmers to let everyone else starve to death. As a society, we need farmers. If we didn't have farmers, every single city dweller would die of starvation. And the low population density that makes it so expensive to provide essential services for farmers is unavoidable, because farming requires large areas of flat dirt without buildings on it. It isn't fair for them to have to pay exorbitant costs solely because they are doing something that our society desperately needs them to do. That's why we have programs like the universal service fund, to ensure that telephone lines are available to groups whose crucial roles in our society prevent them from being able to live in high-density areas. Internet service is fundamentally no different.
Now, could farmers choose to pay tens of thousands of dollars for their Internet connections and then crank up the price of food to compensate? Maybe. But either way, the public as a whole is going to pay for the cost of providing Internet service to the farmers. The advantage of the government paying the ISPs to do it is that rolling out the service universally in an organized fashion tends to be cheaper than doing it piecemeal, and also makes it much harder for the ISPs to say "we'll serve family X, but not family Y."
I addressed that above. Please try again after you have read my comment. You even managed to QUOTE the relevant portion apparently without reading it, but drew some weird conclusions about how it's perfectly safe if your upstream network provider knows everything in one instance but not another. Inconsistent.
I never said it was safe for your upstream network provider to know everything. I said it was moderately safe for the other endpoint to know everything—the content provider, not the service provider. It's at least moderately safe for advertisers to know everything by virtue of having access to info from multiple sites. It isn't particularly safe to allow anybody who has access to a shared public network to know everything about your traffic.
Maybe you should consider implications before coming up with an empty slogan about putting burdens on others just to solve a problem that is not universal and can be solved in other ways.
No, it really can't be solved in other ways. Data on its way to and from the server is either encrypted or it is wide open to anybody who can sniff your traffic. Everyone, at one time or another, uses the Internet in ways that would be detrimental to their safety if the wrong people discovered that usage, such as a potential burglar realizing that you're about to be away from home. The average user is way, way, way too clueless about Internet security to recognize those situations and take steps to protect themselves. Therefore, the burden must fall on everyone who runs a server to protect the general public from their complete lack of understanding by making all content requests secure by default.
And no, caching is not a good reason to break transport security. It is relatively rare for caching to provide big benefits except at large ISPs or college campuses with huge numbers of users. And even then, the benefits mainly come from caching content from big video providers like YouTube or Netflix, both of which have specifically designed their infrastructure to allow ISPs to cache the content on a trusted local server without having to intercept requests, and they make those caching servers available precisely for that reason.
Caching should die already. It was always a cheap hack to work around crappy networks, and it never worked well. HTTPS just adds one more reason to stop doing that nonsense.
The company is legally required to keep that data in secure location. Thus the company's secure location extends to the place where the patient data was found. And accessing it without authorisation is illegal. It's basically similar situation than if you accidentally found out someone's credit card pin code. The person might be careless with communicating his secrets, but still it's still illegal to use the pin code for anything. Same happens with patient data, the secret data might be carelessly handled, but any access to the data is still illegal operation.
The problem with your logic is that unless the filename makes its contents obvious, there's no way to know what's in a file on an FTP server without downloading it. It clearly makes no sense to prosecute someone for a crime if it isn't possible for them to know that they're committing a crime until they have already finished committing it....
Maybe not thousands that are actually moving, but certainly hundreds if you include all the minor stuff like piston rings, valves, pump motors and splines and gaskets, bearings, various other random seals and gaskets, etc. And add in all the other stupid emissions junk like oxygen sensors, and you probably do have thousands of parts that frequently fail, but I wouldn't want to try to add them all up....
Even better would be to just kill this subsidy program entirely. Median farm income in America is over $80k, about 30% higher than the overall median. Why should poor people be taxed to subsidize other people that are better off?
That covers the farm owners, but what about all the other people who work on the farm?
Besides, you don't seem to comprehend the scale here. Areas defined as "highly rural" have fewer than 7 people per square mile. So at most two or three houses per square mile, and possibly not even one house per square mile. Urban areas have over 1,000 people per square mile. We subsidize services for people who make twice as much money as others because otherwise their Internet connections would cost potentially three orders of magnitude more, and even that's potentially an underestimate. That $200 setup fee suddenly becomes a $20,000 setup fee.
I'd say fix the browsers that can't deal with a transition instead of completely fucking up every form of caching that isn't a deliberate Man In The Middle attack. Local news, weather, entertainment - if there is nothing secret in the communications then why apply the extra overhead?
Because you, as the content provider, have no way to know whether exposing that communication can negatively impact the privacy of the person retrieving it. Even seemingly innocuous communications can reveal things that might raise privacy issues. And when you can amass enough of those innocuous communications, you can often conclude things that qualify as serious privacy invasions.
For example, suppose I'm in California and I look up the weather in Tennessee. This might tell you:
I have family in Tennessee
I'm about to go on vacation to Tennessee and want to know what clothes to bring
Either of these pieces of information is probably harmless by itself, but when combined with other information (e.g. my home address, knowledge of some sort of natural disaster in Tennessee, etc.), either could potentially make me vulnerable to various forms of victimization.
People can still build up a profile on you based on the URLs you ask for without the content.
The companies running the sites can. There's a big difference between Facebook knowing I'm about to go on vacation and the seedy-looking guy across the coffee shop who happens to own a moving truck knowing that I'm about to go on vacation. I don't trust Facebook completely, but I don't trust the random guy with the unmarked, plateless moving truck at all.
All traffic should be end-to-end encrypted, period, unless there is no way to do so. Anything less means that A. you can't trust that the content wasn't altered in transit, and B. you have to assume that everybody near you knows exactly what you are doing. IMO, even if the only benefit were preventing MitM attacks, it would still be worth the minimal overhead to use TLS everywhere. The additional privacy edge cases are just a nice bonus.
If you didn't read the T&Cs for all the apps you've installed, it's entirely possible that you have agreed that, when somebody comes knocking at your door some day, they're welcome to take one of your kidneys. You may wail and threaten, but you can expect a response of "[shrug] you agreed to the terms and conditions when you started using our app; lift your shirt, please".
Not likely. Contract terms that a reasonable person would not expect to see in a contract of that type are almost guaranteed to be found unconscionable.
The really funny thing about having onerous T&Cs that the average person can't understand or reasonably be expected to read is that you run the risk of a court deciding that no contract was actually entered into by the other party. When you're writing contracts that an average person is expected to enter into without consulting a lawyer, giant thirty-page legal documents are, IMO, very risky. This is doubly true when the perceived gain by the individual is small (e.g. access to an app or website), the perceived effort involved in reading the T&Cs is large, and (as a result of that incongruity) more than 99% of your users don't bother to actually read the T&C at all and, at best, rapidly scroll through it just to get to the bottom where you hid the "I agree" button.
IMO, when writing things like this, you should give your lawyers a one-page limit. If they can't express the terms and conditions in one page, then the terms are too complex or too onerous for a typical person to be expected to comply with them anyway, and you need to seriously rethink what you're doing. They'll probably balk and give you back something that's two pages long, but at least it won't be thirty.:-)
A MITM is also in the perfect position for an attack on your device. A single specially crafted media file injected into an unencrypted web site that you're browsing over the coffee shop wifi is sufficient to take over your phone. The WPA-PSK encryption does not protect you from that.
See that bold part? That's the problem. Unencrypted web sites shouldn't even exist anymore. Even all of my minor personal sites are all encrypted, with the sole exception of one site hosted by DreamHost using CloudFlare, and only then because CloudFlare has not yet made it possible to use TLS in conjunction with hosting partner accounts. [Redacted swearing]
Basically, there are a handful of infrastructure companies that are holdouts. As soon as those few companies get their act together, the only valid reason to use unencrypted HTTP will be for intranet sites provided via multicast DNS (because.local domains aren't eligible for TLS certs), and even then only because it is too much hassle to set up an entire PKI and install internal-network-specific root certs unless you're a big business. At that point, IMO, going to a non-encrypted site ought to pop up a browser warning, with the default being "Cancel". We're basically at that point now.
There are still plenty of smaller colleges and universities out there with sub-$10,000 tuition. Even at minimum wage, you can go to those schools by working just 50-66% of a full-time workload. They might not have names like "Yale", but you'll still get a decent education there.
The idea that a free universal college education will be the saving grace for displaced minimum wage workers is a pipe dream. I know IQ is a much-maligned concept, BUT many current minimum wage workers could not benefit from a college education even if it were handed to them because they lack the intellectual skill to learn advanced concepts.
Although there are differences in people's intellectual abilities, there's every reason to believe that a much bigger difference in success rates stems from parental encouragement, early learning of important skills, etc. We could make a huge dent in that problem just by guaranteeing everyone a couple of years of pre-K education. And the earlier you pick up a concept, the easier it is. Lots of computer programmers start when they're kids. That usually happens because they have ready access to computers, plus the urge to tinker, much of which is learned behavior. All of these sorts of skills and drives can be instilled in kids from a young age, regardless of their IQ. We just have to be willing to make the effort. And if we do that, maybe they won't all end up as genius software architects, but they'll still be able to contribute.
In the short term, we'll still need people to go to trade schools to become skilled labourers - plumbers, mechanics, nursing assistants, etc.
Nursing, maybe, but only because that involves interacting with people.
IMO, mechanics are living on borrowed time. Robots do most of the work building most cars these days. There are only two major reason why humans repair them: 1. Insurance and warranties pay most of the costs, so there's little incentive to reduce those costs, and 2. There's a strong incentive to do the repairs closer to where the cars' owners live. As individual vehicles become less and less popular and borrowed self-driving cars become the norm, the need for fleet mechanics will probably change that equation.
Robot plumbers will take longer, if only because every plumbing job is different, but we'll see them eventually.
In the long-term, there will be new types of menial jobs that don't require a 4-year degree.
In the long term, robots will replace all of the menial jobs. In the medium term, there will be new types of menial jobs, but it will be an ever-shifting landscape.
Judging from all the whining I hear on various forums from young Americans, college education hasn't been a guarantee of a job for a while now, and it definitely stopped being worth the money at least a decade ago for the vast majority of people.
For the most part, it is necessary, but not sufficient. In a lot of ways, it is like the lottery, just with better odds. You can't win if you don't play, but you may not win even if you do play. With the possible exception of parts of the computer industry, if you want a decent job, you have to have some kind of training. For most jobs, that means college. Without college, your opportunities are considerably fewer than they otherwise would be.
As long as that is true, people will want to go to college, because that could make the difference between maybe getting a job and definitely not getting the job.
I think you missed my point, which was that people like me would give people like them something to do, and they could afford to volunteer for people like me because they wouldn't have to work just to put a roof over their heads.
Realistically, they aren't usually open 24x7, but even at 16x7 operation, that's almost 3 employees replaced, because employees work only 5 days per week. So yeah, it's more like paying for itself after four months versus eight months....
It's cheaper to buy a $35,000 machine than to pay humans their current $7.25 per hour. The machines will pay for themselves in a little over two years at the current minimum wage (after factoring in SSI/SSD/Medicare), versus one year at $15 per hour. Either way, they would save money long before the machines' expected lifespan.
It will always eventually become cheaper to automate menial work than to pay humans to do it. In a hundred years, nobody will be complaining about the lack of fast food jobs, just like nobody is complaining now about the lack of cotton picking jobs. The real problem is finding ways to pay for universal college education so that all the people who are no longer able to work their way through college can still afford to get the education that they'll require to get them to a point where they can contribute usefully to society in a post-menial-labor job market.
Speak for yourself. If I had enough money to retire right now, I'd have more than enough cool personal projects to last me the rest of my life.:-)
But seriously, you're right that most people need some kind of work to have a purpose. However, that doesn't necessarily mean that it has to be corporate work for a company trying to make a profit. Right now, there are lots of nonprofits that just can't work because of the costs involved. I have at least a couple of them in mind right now that I'd love to start, but lack the tens of millions of dollars of seed money to build facilities and hire people. (I'm not going to hire people unless I know I can keep them employed for more than a few weeks, and that's not something I can do with my own personal cash supply.)
However, if I knew that there were millions of unemployed people on the government's payroll who would be willing to volunteer, that would change the equation significantly. Suddenly, I could bring in volunteers for the nonprofit to do the work, and I wouldn't have to worry about not being able to pay them, because they would be guaranteed enough of an income to pay the bills. I might even be able to raise enough money to supplement that income a bit, knowing that if things got tight, we could cut our expenses down to basically zero without our staff starving.
I predict that a basic income would create a new revival for nonprofits, providing a wellspring of staff willing to volunteer.
Wireless is insecure. It's that simple. Do not trust anything that transmits your data without a physical wire because, no matter what protocol, passwords or encryption are used it will always, without a shadow of a doubt, be broken.
If you know where the keyboard is located physically and have a sensitive enough array of directional antennas, you could theoretically detect the voltage spikes from each individual keyswitch closing. And you could probably do some sort of advanced Van Eck phreaking to sniff the USB data lines as well.
Besides, physical security is a minimum requirement for electronic security. By the time you let someone plant unknown hardware inside your building, you've already lost, because they could just as easily replace your USB keyboard with a modified version that retransmits the signal or stores the keystrokes.
Most of us have known for almost a decade that many of Microsoft's wireless mice and keyboards use an insecure protocol. So although this is a clever piece of hardware, it's really sad if anybody is still using vulnerable hardware.
This is just another reason why every time I review a wireless keyboard or mouse or trackball or trackpad, if it isn't Bluetooth, that's usually the first complaint in my review. We have standards for a reason, and those standards are at least moderately robust against this type of attack. Unfortunately, too many keyboard/mouse manufacturers try to cut corners by using whatever cheap custom hardware they've been using for a decade, and they wonder why they get lousy range (not to mention lousy security).
Assuming everyone uses up to the 250 GB cap, that means there are only about 3 million cell phone users in the United States. No, wait... I think I did something wrong....:-D
But seriously, 10 billion gigabytes / 190 million users / 12 months = ~4.4 gigabytes per month average. So why do metered cell phone plans measured in megabytes even still exist?
See, I don't get all the hate towards JavaScript. From what I've seen, it's a perfectly usable language except for the lack of true classes, and even that flaw can be fairly easily worked around if you're a C programmer who did OO back before it was cool (function pointers).
What makes client-side JavaScript a nightmare is not the language, but rather the DOM, which is kind of a pain in any language. The biggest difference is that in other languages, most developers never have to deal with the DOM....
Slashdot Mirror
Why would it be? An electric motor either is getting and drawing power or it isn't. There should be approximately two large power bus wires going to the electric motor, and maybe a couple of wires for an RPM encoder, if they bother putting one in, and depending on where the power switching is located, possibly a couple of control wires. By contrast an ICE design has wires from the computer to dozens of emissions control sensors, speed sensors, temperature sensors, etc., plus wires that allow the computer to control the fuel-air mixture and various aspects of engine timing, control the automatic transmission, etc., plus wires to allow the computer to pressurize the gas tank and sensors to verify that it isn't leaking, plus wires to the relay that powers the fuel pump. So I would expect the wiring harness on an EV to be dramatically simpler than the wiring harness for an ICE-based car. If it isn't, I'd be led to wonder whether the design engineers were getting kickbacks from wire manufacturers....
No, they're vaguely related. They both involve taking and using something that you didn't create. Of course, they're related in much the same way that bumming a stick of chewing gum and stealing a car are related....
Because the only real alternative to everyone else paying for it is for farmers to let everyone else starve to death. As a society, we need farmers. If we didn't have farmers, every single city dweller would die of starvation. And the low population density that makes it so expensive to provide essential services for farmers is unavoidable, because farming requires large areas of flat dirt without buildings on it. It isn't fair for them to have to pay exorbitant costs solely because they are doing something that our society desperately needs them to do. That's why we have programs like the universal service fund, to ensure that telephone lines are available to groups whose crucial roles in our society prevent them from being able to live in high-density areas. Internet service is fundamentally no different.
Now, could farmers choose to pay tens of thousands of dollars for their Internet connections and then crank up the price of food to compensate? Maybe. But either way, the public as a whole is going to pay for the cost of providing Internet service to the farmers. The advantage of the government paying the ISPs to do it is that rolling out the service universally in an organized fashion tends to be cheaper than doing it piecemeal, and also makes it much harder for the ISPs to say "we'll serve family X, but not family Y."
I never said it was safe for your upstream network provider to know everything. I said it was moderately safe for the other endpoint to know everything—the content provider, not the service provider. It's at least moderately safe for advertisers to know everything by virtue of having access to info from multiple sites. It isn't particularly safe to allow anybody who has access to a shared public network to know everything about your traffic.
No, it really can't be solved in other ways. Data on its way to and from the server is either encrypted or it is wide open to anybody who can sniff your traffic. Everyone, at one time or another, uses the Internet in ways that would be detrimental to their safety if the wrong people discovered that usage, such as a potential burglar realizing that you're about to be away from home. The average user is way, way, way too clueless about Internet security to recognize those situations and take steps to protect themselves. Therefore, the burden must fall on everyone who runs a server to protect the general public from their complete lack of understanding by making all content requests secure by default.
And no, caching is not a good reason to break transport security. It is relatively rare for caching to provide big benefits except at large ISPs or college campuses with huge numbers of users. And even then, the benefits mainly come from caching content from big video providers like YouTube or Netflix, both of which have specifically designed their infrastructure to allow ISPs to cache the content on a trusted local server without having to intercept requests, and they make those caching servers available precisely for that reason.
Caching should die already. It was always a cheap hack to work around crappy networks, and it never worked well. HTTPS just adds one more reason to stop doing that nonsense.
The problem with your logic is that unless the filename makes its contents obvious, there's no way to know what's in a file on an FTP server without downloading it. It clearly makes no sense to prosecute someone for a crime if it isn't possible for them to know that they're committing a crime until they have already finished committing it....
Maybe not thousands that are actually moving, but certainly hundreds if you include all the minor stuff like piston rings, valves, pump motors and splines and gaskets, bearings, various other random seals and gaskets, etc. And add in all the other stupid emissions junk like oxygen sensors, and you probably do have thousands of parts that frequently fail, but I wouldn't want to try to add them all up....
That covers the farm owners, but what about all the other people who work on the farm?
Besides, you don't seem to comprehend the scale here. Areas defined as "highly rural" have fewer than 7 people per square mile. So at most two or three houses per square mile, and possibly not even one house per square mile. Urban areas have over 1,000 people per square mile. We subsidize services for people who make twice as much money as others because otherwise their Internet connections would cost potentially three orders of magnitude more, and even that's potentially an underestimate. That $200 setup fee suddenly becomes a $20,000 setup fee.
Because you, as the content provider, have no way to know whether exposing that communication can negatively impact the privacy of the person retrieving it. Even seemingly innocuous communications can reveal things that might raise privacy issues. And when you can amass enough of those innocuous communications, you can often conclude things that qualify as serious privacy invasions.
For example, suppose I'm in California and I look up the weather in Tennessee. This might tell you:
Either of these pieces of information is probably harmless by itself, but when combined with other information (e.g. my home address, knowledge of some sort of natural disaster in Tennessee, etc.), either could potentially make me vulnerable to various forms of victimization.
The companies running the sites can. There's a big difference between Facebook knowing I'm about to go on vacation and the seedy-looking guy across the coffee shop who happens to own a moving truck knowing that I'm about to go on vacation. I don't trust Facebook completely, but I don't trust the random guy with the unmarked, plateless moving truck at all.
All traffic should be end-to-end encrypted, period, unless there is no way to do so. Anything less means that A. you can't trust that the content wasn't altered in transit, and B. you have to assume that everybody near you knows exactly what you are doing. IMO, even if the only benefit were preventing MitM attacks, it would still be worth the minimal overhead to use TLS everywhere. The additional privacy edge cases are just a nice bonus.
Not likely. Contract terms that a reasonable person would not expect to see in a contract of that type are almost guaranteed to be found unconscionable.
The really funny thing about having onerous T&Cs that the average person can't understand or reasonably be expected to read is that you run the risk of a court deciding that no contract was actually entered into by the other party. When you're writing contracts that an average person is expected to enter into without consulting a lawyer, giant thirty-page legal documents are, IMO, very risky. This is doubly true when the perceived gain by the individual is small (e.g. access to an app or website), the perceived effort involved in reading the T&Cs is large, and (as a result of that incongruity) more than 99% of your users don't bother to actually read the T&C at all and, at best, rapidly scroll through it just to get to the bottom where you hid the "I agree" button.
IMO, when writing things like this, you should give your lawyers a one-page limit. If they can't express the terms and conditions in one page, then the terms are too complex or too onerous for a typical person to be expected to comply with them anyway, and you need to seriously rethink what you're doing. They'll probably balk and give you back something that's two pages long, but at least it won't be thirty. :-)
See that bold part? That's the problem. Unencrypted web sites shouldn't even exist anymore. Even all of my minor personal sites are all encrypted, with the sole exception of one site hosted by DreamHost using CloudFlare, and only then because CloudFlare has not yet made it possible to use TLS in conjunction with hosting partner accounts. [Redacted swearing]
Basically, there are a handful of infrastructure companies that are holdouts. As soon as those few companies get their act together, the only valid reason to use unencrypted HTTP will be for intranet sites provided via multicast DNS (because .local domains aren't eligible for TLS certs), and even then only because it is too much hassle to set up an entire PKI and install internal-network-specific root certs unless you're a big business. At that point, IMO, going to a non-encrypted site ought to pop up a browser warning, with the default being "Cancel". We're basically at that point now.
You've been using Java too much. I think you mean JMOL.
:-D
There are still plenty of smaller colleges and universities out there with sub-$10,000 tuition. Even at minimum wage, you can go to those schools by working just 50-66% of a full-time workload. They might not have names like "Yale", but you'll still get a decent education there.
That was the reason for the ":-D".
Although there are differences in people's intellectual abilities, there's every reason to believe that a much bigger difference in success rates stems from parental encouragement, early learning of important skills, etc. We could make a huge dent in that problem just by guaranteeing everyone a couple of years of pre-K education. And the earlier you pick up a concept, the easier it is. Lots of computer programmers start when they're kids. That usually happens because they have ready access to computers, plus the urge to tinker, much of which is learned behavior. All of these sorts of skills and drives can be instilled in kids from a young age, regardless of their IQ. We just have to be willing to make the effort. And if we do that, maybe they won't all end up as genius software architects, but they'll still be able to contribute.
Nursing, maybe, but only because that involves interacting with people.
IMO, mechanics are living on borrowed time. Robots do most of the work building most cars these days. There are only two major reason why humans repair them: 1. Insurance and warranties pay most of the costs, so there's little incentive to reduce those costs, and 2. There's a strong incentive to do the repairs closer to where the cars' owners live. As individual vehicles become less and less popular and borrowed self-driving cars become the norm, the need for fleet mechanics will probably change that equation.
Robot plumbers will take longer, if only because every plumbing job is different, but we'll see them eventually.
In the long term, robots will replace all of the menial jobs. In the medium term, there will be new types of menial jobs, but it will be an ever-shifting landscape.
For the most part, it is necessary, but not sufficient. In a lot of ways, it is like the lottery, just with better odds. You can't win if you don't play, but you may not win even if you do play. With the possible exception of parts of the computer industry, if you want a decent job, you have to have some kind of training. For most jobs, that means college. Without college, your opportunities are considerably fewer than they otherwise would be.
As long as that is true, people will want to go to college, because that could make the difference between maybe getting a job and definitely not getting the job.
I think you missed my point, which was that people like me would give people like them something to do, and they could afford to volunteer for people like me because they wouldn't have to work just to put a roof over their heads.
Realistically, they aren't usually open 24x7, but even at 16x7 operation, that's almost 3 employees replaced, because employees work only 5 days per week. So yeah, it's more like paying for itself after four months versus eight months....
It's cheaper to buy a $35,000 machine than to pay humans their current $7.25 per hour. The machines will pay for themselves in a little over two years at the current minimum wage (after factoring in SSI/SSD/Medicare), versus one year at $15 per hour. Either way, they would save money long before the machines' expected lifespan.
It will always eventually become cheaper to automate menial work than to pay humans to do it. In a hundred years, nobody will be complaining about the lack of fast food jobs, just like nobody is complaining now about the lack of cotton picking jobs. The real problem is finding ways to pay for universal college education so that all the people who are no longer able to work their way through college can still afford to get the education that they'll require to get them to a point where they can contribute usefully to society in a post-menial-labor job market.
Speak for yourself. If I had enough money to retire right now, I'd have more than enough cool personal projects to last me the rest of my life. :-)
But seriously, you're right that most people need some kind of work to have a purpose. However, that doesn't necessarily mean that it has to be corporate work for a company trying to make a profit. Right now, there are lots of nonprofits that just can't work because of the costs involved. I have at least a couple of them in mind right now that I'd love to start, but lack the tens of millions of dollars of seed money to build facilities and hire people. (I'm not going to hire people unless I know I can keep them employed for more than a few weeks, and that's not something I can do with my own personal cash supply.)
However, if I knew that there were millions of unemployed people on the government's payroll who would be willing to volunteer, that would change the equation significantly. Suddenly, I could bring in volunteers for the nonprofit to do the work, and I wouldn't have to worry about not being able to pay them, because they would be guaranteed enough of an income to pay the bills. I might even be able to raise enough money to supplement that income a bit, knowing that if things got tight, we could cut our expenses down to basically zero without our staff starving.
I predict that a basic income would create a new revival for nonprofits, providing a wellspring of staff willing to volunteer.
Maybe this is what Deep Thought's answer was about....
If you know where the keyboard is located physically and have a sensitive enough array of directional antennas, you could theoretically detect the voltage spikes from each individual keyswitch closing. And you could probably do some sort of advanced Van Eck phreaking to sniff the USB data lines as well.
Besides, physical security is a minimum requirement for electronic security. By the time you let someone plant unknown hardware inside your building, you've already lost, because they could just as easily replace your USB keyboard with a modified version that retransmits the signal or stores the keystrokes.
Most of us have known for almost a decade that many of Microsoft's wireless mice and keyboards use an insecure protocol. So although this is a clever piece of hardware, it's really sad if anybody is still using vulnerable hardware.
This is just another reason why every time I review a wireless keyboard or mouse or trackball or trackpad, if it isn't Bluetooth, that's usually the first complaint in my review. We have standards for a reason, and those standards are at least moderately robust against this type of attack. Unfortunately, too many keyboard/mouse manufacturers try to cut corners by using whatever cheap custom hardware they've been using for a decade, and they wonder why they get lousy range (not to mention lousy security).
Assuming everyone uses up to the 250 GB cap, that means there are only about 3 million cell phone users in the United States. No, wait... I think I did something wrong.... :-D
But seriously, 10 billion gigabytes / 190 million users / 12 months = ~4.4 gigabytes per month average. So why do metered cell phone plans measured in megabytes even still exist?
See, I don't get all the hate towards JavaScript. From what I've seen, it's a perfectly usable language except for the lack of true classes, and even that flaw can be fairly easily worked around if you're a C programmer who did OO back before it was cool (function pointers).
What makes client-side JavaScript a nightmare is not the language, but rather the DOM, which is kind of a pain in any language. The biggest difference is that in other languages, most developers never have to deal with the DOM....