He doesn't have to worry as much about getting robbed.
Unless...someone steals his laptop.
Actually, if the wallet is deterministic and encrypted with a passphrase—the default for the Electrum client, among others—and he either memorized the seed or recorded it somewhere safe, then even that wouldn't be a problem. Of course, he'd still be out the cost of replacing the laptop, but he wouldn't lose his bitcoins.
Physical theft of the device is less of a threat these days than malware. Malicious software running on your device has a better chance of being able to access your private keys.
Why would they need to store the models? Document it thoroughly, make sure it works the way the application says it does, and then archive the digital data. The submitter can keep the prototype.
Then you must have one of the new Nexus devices with inductive charging, and I can only assume you already know what to do with it and whether it's working properly.
That means a 100kg person weighs 700g more near the North Pole
You are confused by the units they are using. This is actually the kilogram-force [wikipedia.org], a non-SI unit of weight...
Anyone would be confused by their units. Even if they did mean to refer to the kilogram-force (why?) they still got it wrong. The shorthand for kilogram-force is "kgf" or "kp", not "kg" or "g". The article's "700g" is a measure of mass, not weight.
Better: A 100.0kg person at the North Pole weighs the same as a 100.7kg person at Peru's Nevado Huascaran summit.
For weights that are comparing against a known mass there is no problem. The 1 kg of material you want to buy will always weigh the same as the 1 kg on the other side of the scale weight...
True, but not all scales work that way. Many use a spring or pressure sensor rather than comparing weights directly, and thus wouldn't take variations in the gravitational field into account. These scales would need to be calibrated against a known mass if they were used in a different location than they were manufactured in. (Of course, they probably needed to be calibrated anyway...)
Is there some term that is used to distinguish connectors with / without this functionality, so I can buy the right kind?
I gave up on finding USB Charging Specification-compatible chargers a while ago and just picked up a "charge-only" USB cable, which does the same thing as the adapter in this article: short the D+ and D- pins on the device side. This lets any standards-compliant (i.e. non-Apple) device know that it's safe to charge at full speed, so it should fix the problem so long as your charger can handle the current.
You can tell whether an Android device is charging properly by looking at the Battery pages in Settings. It should say "Charging (AC)" to indicate a full-speed charge, or "Charging (USB)" to indicate that it's limiting itself to 100mA.
This is the part I don't understand about the model. In order for our universe to be the 3D surface of a 4D black hole, everything in our universe would have to somehow be constrained to move along the event horizon. Otherwise the event horizon would be just one of many possible 3D subspaces to consider within the larger 4D universe.
What is it that forces matter and energy in our universe to stay on the event horizon, rather than either escaping or falling into the black hole? I don't recall hearing about any such effect where 3D black holes are concerned.
a black hole in such a universe would have a 3-d "surface"
I'm trying to decide whether this makes any more sense than a square circle. 3D surface is a contradiction in terms. A surface is 2 dimensional by definition.
The term "surface" normally refers to a two-dimensional shape in 3D space, but it can be generalized to any number of dimensions (a hypersurface). One example would be a hypersphere (x**2 + y**2 + z**2 + w**2 = 1), which has three orthogonal directions of movement along the hypersurface and encloses a four-dimensional space. Movement tangent to the hypersphere it would seem like movement in normal 3D Euclidean space, except that if you travel far enough in any direction you'll eventually end up back where you started.
Once something has passed the edge of the visible universe it is effectively lost to us
Only until we build a bigger telescope.
It's not a matter of how large or sensitive the telescope is; if something is far enough away, the expansion of the space between the object and ourselves causes the distance between us to increase faster than the speed of light, meaning light from the object can never reach us. Once something reaches that distance it's cut off from us for good (or at least as long as the universe continues to expand).
It's not really the same because anything that collides with a black hole will cease to exist.... Even if the collapsed star's gravity did not stop the photons from exiting it would effectively vanish out of existence.
These are one and the same thing. Black holes are not particularly special; the event horizon isn't some solid barrier things crash into. It's merely the point of no return, beyond which escape velocity exceeds the speed of light. Objects which enter a black hole "cease to exist" in exactly the same sense as objects which pass beyond the visible universe: any effect involving the object would need to propagate faster than the speed of light to reach us.
Yes, but once the functionality is in a shared library, it's no more work to use the library from the toolkit than it would be for the toolkit to describe to X what it wants X to do with the library. On the contrary, by removing a layer of indirection it's easier for the toolkit to get the effect it wants, since it has the full API of the library to work with rather than just the portion exposed via an extension to the X11 protocol.
Compositing is used to implement various helpful window management features which are not possible when drawing directly to the screen, like the "Exposé" / "Present Windows" effect and live previews during task switching. It also allows various accessibility improvements, like contrast-enhancing visual filters and smooth screen magnification, and is required for the efficient implementation of translucent windows, including merely non-rectangular windows with antialiased borders.
My remote X11 clients run on these machines and present their UIs on my local X11 display server running on my laptop. While it is probably true that these clients are not transmitting XDrawLine and XFillArc protocol elements to render their UIs, they are still mostly assembling pre-rendered bitmaps, widgets, and font glyph assets to send down the wire for rendering on the local server. How is this going to work on Wayland?
Actually, what the clients are doing right now is assembling bitmaps, widgets, and font glyph assets into a pixmap on the client side, most likely without the benefits of GPU acceleration, and sending the result as an uncompressed pixmap over the wire to the X server, which hands it off to a compositor, which combines the pixmap with images from other applications and hands the result back to the X server. If they're luck enough not to need any special transformations or compositing effects, the clients may be able to leave the rendering of the individual font glyphs to the server, but that's about it.
With Wayland the clients are doing the same work to assemble the surfaces for their windows, but they get to use the local GPU to do it, and the result is compressed by a local off-screen Wayland proxy server using modern video codecs before being transmitted over the network for compositing.
On a desktop, the only advantage to Wayland is that it facilitates implementing a pretty compositing desktop. This is a fad that is already starting to fade from fashion.
Distracting toys like "wobbly windows" may be fading from fashion, but composited desktops are here to stay.
If all toolkits have to implement something then X is the proper place to put it.
No, if all toolkits have to implement something then a shared library is the proper place to put it. For the most part, that's what's been happening. For example, both Qt and GTK+ need to render fonts, so they rely on FreeType.
The problem with putting drawing primitives and font rendering in the X server is that applications generally want to compose these operations together in ways the standard APIs don't support, which would imply either lots of round-trips and wasted effort or moving the application's drawing code into the X server (like DPS). When both the X server and the application are local, it's obviously both easier and more efficient to hand all of the rendering over to the application. The interesting part is that even if the application is remote, it can be more efficient to communicate a compressed form of the result over the network rather than all the drawing commands and raw data required to assemble the same image locally.
Looking back I see that I wasn't entirely clear. I did actually mean to include this possibility; a 51% attack can result in supposedly settled transactions being reversed, along with transactions which have not yet made it into a block.
Any miner can choose which block to base their new block on; it doesn't have to be the latest one in the dominant blockchain. However, honest nodes will prefer the branch of the blockchain with the highest total difficulty, so by choosing an older block you're starting at a disadvantage. The effort required to revert the blockchain to an earlier point goes up exponentially based on the number of blocks you're trying to replace; at 51% you essentially have a 50/50 chance of being in charge of the next block, so the effort required would scale with approximately 2^n, where n is the number of blocks. The task becomes easier as you control more of the network; at 100% you would require "only" the same effort as it took to produce those blocks in the first place (plus one additional block to break the tie).
A malevolent agent would need to control at least than 51% to outvote and falsify transaction history...
Note that "falsify" in this context is still limited to blocking or reversing existing (valid) transactions. A person with 51% of the hashing power of the entire network could spend bitcoins from his own accounts multiple times, or allow someone else to do the same, or prevent someone (or everyone) else from spending their bitcoins. He still wouldn't be able to spend anyone else's bitcoins without their private key, no matter how much of the mining he controls. The winning miner chooses the transactions which go into the next block, but the block itself is still subject to certain rules, including the rule that the transactions must have valid signatures. A block which will only be accepted by your own mining nodes isn't of much use to anyone.
Also, few intel processors have built-in RNGs, at least not ones the Linux kernel can use. None of the machines we've bought in the last 5 years have them. When was the last major intel x86 processor to have one? P2/P3 based systems?
How about all the latest Ivy Bridge-based processors, like the Core I3/i5/i7, via the RdRand opcode?
There is no longer a separate HW RNG, but that's only because it was moved into the core. And yes, the RdRand opcode is based on NSA-influenced NIST standards and may well be compromised.
With perfect risk assessment being possible, no insurance company would issue a policy until that assessment was done...
Accurate assessment has a cost as well, both to perform the assessment and in losing customers who aren't willing to complete it. It is always possible to compensate for uncertainty in the assessment by assuming a higher level of risk, constrained on one side by the actual cost to the insurance company and on the other by competition for insurance customers. If insurance companies always waited for perfect information there would be no insurance companies.
... once the assessment does get done, anyone that knows they are not at risk, will just cancel their policy once they know they are not at risk.
In the pre-conception insurance policies I was referring to, I had in mind a single lump-sum payment rather than monthly premiums, so there would be no risk of anyone cancelling their policy once the know the results of the screening. That's how insurance already works for other one-time events, e.g. insuring your outdoor concert against poor weather. For those without the means to cover such a payment up front, it could be treated as a loan; the point is that the customer's entire obligation to the insurer is set in stone before the event, just like the insurer's obligation to the customer.
For more traditional medical insurance, I would think of the periodic premiums as covering the total cost of anything discovered during that particular period. If something is discovered and the customer subsequently cancels their insurance, that's fine. Even if they were only a customer for one month, their first month's premium covered the expected lifetime cost of anything new discovered during that month.
Insurance is useless for those who don't have risk. Paying insurance to cover you against diabetes would be stupid if you could see through a DNA test that you will never get it.
Exactly. Insurance isn't meant to cover those who without risk, any more than it's meant to cover those facing a known cost. It's for the cases in between, where you have a low risk of a high cost and would prefer a predictable premium.
When we do it your way, it doesn't work at all. People who aren't going to need insurance will get it really cheap. People who do will be unable to afford it.
That is exactly how insurance is supposed to work. If you aren't high-risk, there is no reason for you to pay a lot for insurance. Conversely, if you are high-risk, you should pay more, because you're expected to cost more. Insurance isn't a charity or transfer scheme, it's a way to take a risk and make it into a fixed cost. Instead of a 0.1% chance of paying $1,000,000, you pay $1,000. The expected cost (risk * cost) does not change, apart from overhead.
The only way it works at all is either if you sign up before any risk assessments like this can be done...
Exactly. The proper time to take out insurance on a child's genetics would be before the genetics are determined, based on the biological parents' risk factors. If you're a high-risk parent due to inheritable genetic factors you can either pay more for the insurance or consider e.g. adoption rather than passing on your own genes. The insurance should cover the total lifetime cost of treating any genetic disorders which are discovered. (This is part of the problem with "preexisting conditions": the original insurer should be on the hook for the total cost of treating anything discovered while you're a customer, not just the ongoing costs so long as you maintain your insurance.)
What is needed is regulation of the insurance companys to reduce the scope of their discrimination.
Discrimination, at least insofar as it relates to risk, is pretty much the entire point of an insurance company. Assessing actuarial risk to determine the expected cost of insuring someone is fundamental to the job. If you want charity rather than insurance, just say so; please don't ruin actual insurance for those who want it.
The existence of the militia, and the definition I used, predate the Constitution, much less the Second Militia Act of 1792. See the history of militia in the United States: "[The term "militia"] has historically been used to describe all able-bodied men who are not members of the Army or Navy (Uniformed Services)."
Then what... is a "well regulated militia"? One guy regulating himself?
In the language of the time, it meant every able-bodied male of military age, with the training and supplies necessary to operate as an effective military force in time of need. There was no question of whether weapons were limited to the militia, because the militia was everyone deemed capable of using them.
In any case, the right is not restricted to the militia: "the right of the people to keep and bear arms, shall not be infringed." "The people" is an even more all-inclusive term than "well regulated militia".
I don't disagree. My own preference would be an open-source native client with no ties to the browser, something stable which can be audited and won't be replaced every time Mega updates their web site. However, the existing system isn't exactly handing the key out to every web site you visit, just the components which are expected to have access to it. It's about as secure as can be expected of a plain web app.
Only if you define "someone with access to my computer" to include "anyone who runs a web server I visit".
The article doesn't say that any web site can access the key, just the browser itself (via bookmarklets or third-party extensions) and Mega. Both of which are obvious.
The browser prevents sites from accessing other sites' local data. It would be interesting if they managed to find a way around that protection, but they didn't. The system is working as designed.
The private schools should survive on their own resources; parents sending their kids to private schools should not be able to direct education taxes towards them. This, again, goes to the universality of the benefit of an educated populace. All should pay, and only the publicly owned schools should receive the money.
You may want to rethink that. This isn't "all pay equally", this is "send your kids to public school or pay double". At the very least, parents who take direct responsibility for the education of the portion of the populace which they are personally responsible for (their own kids) should be able to count that toward their part of the cost of providing universal education. They shouldn't be forced to pay twice, once for the education their kids are actually getting and again for the one which would otherwise have been provided by the state.
Note that I'm not saying that tax money should go to private schools (like with vouchers). Rather, if the education part of your taxes amounts to the cost of publicly educating three kids, and you pay to send two kids to an accredited private school, or home school them according to the required curriculum, that should count as 2/3 of the taxes, since the public school has two fewer students to worry about. This would be capped at your share of the public education cost, no matter how much extra you paid (i.e. no refunds).
Parents will probably still end up paying more for a private school, but at least they won't have to pay for both, which should reduce the obstacles to sending kids to private schools and consequently improve the general level of education. The public schools, in the meantime, are still getting just as much money per student as before; they just don't get paid extra for students they aren't responsible for educating.
Slashdot Mirror
He doesn't have to worry as much about getting robbed.
Unless...someone steals his laptop.
Actually, if the wallet is deterministic and encrypted with a passphrase—the default for the Electrum client, among others—and he either memorized the seed or recorded it somewhere safe, then even that wouldn't be a problem. Of course, he'd still be out the cost of replacing the laptop, but he wouldn't lose his bitcoins.
Physical theft of the device is less of a threat these days than malware. Malicious software running on your device has a better chance of being able to access your private keys.
Why would they need to store the models? Document it thoroughly, make sure it works the way the application says it does, and then archive the digital data. The submitter can keep the prototype.
What if it says Charging (wireless)?
Then you must have one of the new Nexus devices with inductive charging, and I can only assume you already know what to do with it and whether it's working properly.
That means a 100kg person weighs 700g more near the North Pole
You are confused by the units they are using. This is actually the kilogram-force [wikipedia.org], a non-SI unit of weight...
Anyone would be confused by their units. Even if they did mean to refer to the kilogram-force (why?) they still got it wrong. The shorthand for kilogram-force is "kgf" or "kp", not "kg" or "g". The article's "700g" is a measure of mass, not weight.
Better: A 100.0kg person at the North Pole weighs the same as a 100.7kg person at Peru's Nevado Huascaran summit.
For weights that are comparing against a known mass there is no problem. The 1 kg of material you want to buy will always weigh the same as the 1 kg on the other side of the scale weight ...
True, but not all scales work that way. Many use a spring or pressure sensor rather than comparing weights directly, and thus wouldn't take variations in the gravitational field into account. These scales would need to be calibrated against a known mass if they were used in a different location than they were manufactured in. (Of course, they probably needed to be calibrated anyway...)
Is there some term that is used to distinguish connectors with / without this functionality, so I can buy the right kind?
I gave up on finding USB Charging Specification-compatible chargers a while ago and just picked up a "charge-only" USB cable, which does the same thing as the adapter in this article: short the D+ and D- pins on the device side. This lets any standards-compliant (i.e. non-Apple) device know that it's safe to charge at full speed, so it should fix the problem so long as your charger can handle the current.
You can tell whether an Android device is charging properly by looking at the Battery pages in Settings. It should say "Charging (AC)" to indicate a full-speed charge, or "Charging (USB)" to indicate that it's limiting itself to 100mA.
This is the part I don't understand about the model. In order for our universe to be the 3D surface of a 4D black hole, everything in our universe would have to somehow be constrained to move along the event horizon. Otherwise the event horizon would be just one of many possible 3D subspaces to consider within the larger 4D universe.
What is it that forces matter and energy in our universe to stay on the event horizon, rather than either escaping or falling into the black hole? I don't recall hearing about any such effect where 3D black holes are concerned.
a black hole in such a universe would have a 3-d "surface"
I'm trying to decide whether this makes any more sense than a square circle. 3D surface is a contradiction in terms. A surface is 2 dimensional by definition.
The term "surface" normally refers to a two-dimensional shape in 3D space, but it can be generalized to any number of dimensions (a hypersurface). One example would be a hypersphere (x**2 + y**2 + z**2 + w**2 = 1), which has three orthogonal directions of movement along the hypersurface and encloses a four-dimensional space. Movement tangent to the hypersphere it would seem like movement in normal 3D Euclidean space, except that if you travel far enough in any direction you'll eventually end up back where you started.
Once something has passed the edge of the visible universe it is effectively lost to us
Only until we build a bigger telescope.
It's not a matter of how large or sensitive the telescope is; if something is far enough away, the expansion of the space between the object and ourselves causes the distance between us to increase faster than the speed of light, meaning light from the object can never reach us. Once something reaches that distance it's cut off from us for good (or at least as long as the universe continues to expand).
It's not really the same because anything that collides with a black hole will cease to exist. ... Even if the collapsed star's gravity did not stop the photons from exiting it would effectively vanish out of existence.
These are one and the same thing. Black holes are not particularly special; the event horizon isn't some solid barrier things crash into. It's merely the point of no return, beyond which escape velocity exceeds the speed of light. Objects which enter a black hole "cease to exist" in exactly the same sense as objects which pass beyond the visible universe: any effect involving the object would need to propagate faster than the speed of light to reach us.
Doesn't X contain a lot of shared libraries?
Yes, but once the functionality is in a shared library, it's no more work to use the library from the toolkit than it would be for the toolkit to describe to X what it wants X to do with the library. On the contrary, by removing a layer of indirection it's easier for the toolkit to get the effect it wants, since it has the full API of the library to work with rather than just the portion exposed via an extension to the X11 protocol.
Why are composited desktops important?
Compositing is used to implement various helpful window management features which are not possible when drawing directly to the screen, like the "Exposé" / "Present Windows" effect and live previews during task switching. It also allows various accessibility improvements, like contrast-enhancing visual filters and smooth screen magnification, and is required for the efficient implementation of translucent windows, including merely non-rectangular windows with antialiased borders.
My remote X11 clients run on these machines and present their UIs on my local X11 display server running on my laptop. While it is probably true that these clients are not transmitting XDrawLine and XFillArc protocol elements to render their UIs, they are still mostly assembling pre-rendered bitmaps, widgets, and font glyph assets to send down the wire for rendering on the local server. How is this going to work on Wayland?
Actually, what the clients are doing right now is assembling bitmaps, widgets, and font glyph assets into a pixmap on the client side, most likely without the benefits of GPU acceleration, and sending the result as an uncompressed pixmap over the wire to the X server, which hands it off to a compositor, which combines the pixmap with images from other applications and hands the result back to the X server. If they're luck enough not to need any special transformations or compositing effects, the clients may be able to leave the rendering of the individual font glyphs to the server, but that's about it.
With Wayland the clients are doing the same work to assemble the surfaces for their windows, but they get to use the local GPU to do it, and the result is compressed by a local off-screen Wayland proxy server using modern video codecs before being transmitted over the network for compositing.
On a desktop, the only advantage to Wayland is that it facilitates implementing a pretty compositing desktop. This is a fad that is already starting to fade from fashion.
Distracting toys like "wobbly windows" may be fading from fashion, but composited desktops are here to stay.
If all toolkits have to implement something then X is the proper place to put it.
No, if all toolkits have to implement something then a shared library is the proper place to put it. For the most part, that's what's been happening. For example, both Qt and GTK+ need to render fonts, so they rely on FreeType.
The problem with putting drawing primitives and font rendering in the X server is that applications generally want to compose these operations together in ways the standard APIs don't support, which would imply either lots of round-trips and wasted effort or moving the application's drawing code into the X server (like DPS). When both the X server and the application are local, it's obviously both easier and more efficient to hand all of the rendering over to the application. The interesting part is that even if the application is remote, it can be more efficient to communicate a compressed form of the result over the network rather than all the drawing commands and raw data required to assemble the same image locally.
Looking back I see that I wasn't entirely clear. I did actually mean to include this possibility; a 51% attack can result in supposedly settled transactions being reversed, along with transactions which have not yet made it into a block.
Any miner can choose which block to base their new block on; it doesn't have to be the latest one in the dominant blockchain. However, honest nodes will prefer the branch of the blockchain with the highest total difficulty, so by choosing an older block you're starting at a disadvantage. The effort required to revert the blockchain to an earlier point goes up exponentially based on the number of blocks you're trying to replace; at 51% you essentially have a 50/50 chance of being in charge of the next block, so the effort required would scale with approximately 2^n, where n is the number of blocks. The task becomes easier as you control more of the network; at 100% you would require "only" the same effort as it took to produce those blocks in the first place (plus one additional block to break the tie).
A malevolent agent would need to control at least than 51% to outvote and falsify transaction history ...
Note that "falsify" in this context is still limited to blocking or reversing existing (valid) transactions. A person with 51% of the hashing power of the entire network could spend bitcoins from his own accounts multiple times, or allow someone else to do the same, or prevent someone (or everyone) else from spending their bitcoins. He still wouldn't be able to spend anyone else's bitcoins without their private key, no matter how much of the mining he controls. The winning miner chooses the transactions which go into the next block, but the block itself is still subject to certain rules, including the rule that the transactions must have valid signatures. A block which will only be accepted by your own mining nodes isn't of much use to anyone.
Also, few intel processors have built-in RNGs, at least not ones the Linux kernel can use. None of the machines we've bought in the last 5 years have them. When was the last major intel x86 processor to have one? P2/P3 based systems?
How about all the latest Ivy Bridge-based processors, like the Core I3/i5/i7, via the RdRand opcode?
There is no longer a separate HW RNG, but that's only because it was moved into the core. And yes, the RdRand opcode is based on NSA-influenced NIST standards and may well be compromised.
With perfect risk assessment being possible, no insurance company would issue a policy until that assessment was done...
Accurate assessment has a cost as well, both to perform the assessment and in losing customers who aren't willing to complete it. It is always possible to compensate for uncertainty in the assessment by assuming a higher level of risk, constrained on one side by the actual cost to the insurance company and on the other by competition for insurance customers. If insurance companies always waited for perfect information there would be no insurance companies.
... once the assessment does get done, anyone that knows they are not at risk, will just cancel their policy once they know they are not at risk.
In the pre-conception insurance policies I was referring to, I had in mind a single lump-sum payment rather than monthly premiums, so there would be no risk of anyone cancelling their policy once the know the results of the screening. That's how insurance already works for other one-time events, e.g. insuring your outdoor concert against poor weather. For those without the means to cover such a payment up front, it could be treated as a loan; the point is that the customer's entire obligation to the insurer is set in stone before the event, just like the insurer's obligation to the customer.
For more traditional medical insurance, I would think of the periodic premiums as covering the total cost of anything discovered during that particular period. If something is discovered and the customer subsequently cancels their insurance, that's fine. Even if they were only a customer for one month, their first month's premium covered the expected lifetime cost of anything new discovered during that month.
Insurance is useless for those who don't have risk. Paying insurance to cover you against diabetes would be stupid if you could see through a DNA test that you will never get it.
Exactly. Insurance isn't meant to cover those who without risk, any more than it's meant to cover those facing a known cost. It's for the cases in between, where you have a low risk of a high cost and would prefer a predictable premium.
When we do it your way, it doesn't work at all. People who aren't going to need insurance will get it really cheap. People who do will be unable to afford it.
That is exactly how insurance is supposed to work. If you aren't high-risk, there is no reason for you to pay a lot for insurance. Conversely, if you are high-risk, you should pay more, because you're expected to cost more. Insurance isn't a charity or transfer scheme, it's a way to take a risk and make it into a fixed cost. Instead of a 0.1% chance of paying $1,000,000, you pay $1,000. The expected cost (risk * cost) does not change, apart from overhead.
The only way it works at all is either if you sign up before any risk assessments like this can be done ...
Exactly. The proper time to take out insurance on a child's genetics would be before the genetics are determined, based on the biological parents' risk factors. If you're a high-risk parent due to inheritable genetic factors you can either pay more for the insurance or consider e.g. adoption rather than passing on your own genes. The insurance should cover the total lifetime cost of treating any genetic disorders which are discovered. (This is part of the problem with "preexisting conditions": the original insurer should be on the hook for the total cost of treating anything discovered while you're a customer, not just the ongoing costs so long as you maintain your insurance.)
What is needed is regulation of the insurance companys to reduce the scope of their discrimination.
Discrimination, at least insofar as it relates to risk, is pretty much the entire point of an insurance company. Assessing actuarial risk to determine the expected cost of insuring someone is fundamental to the job. If you want charity rather than insurance, just say so; please don't ruin actual insurance for those who want it.
The existence of the militia, and the definition I used, predate the Constitution, much less the Second Militia Act of 1792. See the history of militia in the United States: "[The term "militia"] has historically been used to describe all able-bodied men who are not members of the Army or Navy (Uniformed Services)."
But that would mean your right to bear arms terminates at age 45, and if you were listed as an exception.
No, it wouldn't, because it is the right of the people to keep and bear arms, not just the militia.
Then what ... is a "well regulated militia"? One guy regulating himself?
In the language of the time, it meant every able-bodied male of military age, with the training and supplies necessary to operate as an effective military force in time of need. There was no question of whether weapons were limited to the militia, because the militia was everyone deemed capable of using them.
In any case, the right is not restricted to the militia: "the right of the people to keep and bear arms, shall not be infringed." "The people" is an even more all-inclusive term than "well regulated militia".
I don't disagree. My own preference would be an open-source native client with no ties to the browser, something stable which can be audited and won't be replaced every time Mega updates their web site. However, the existing system isn't exactly handing the key out to every web site you visit, just the components which are expected to have access to it. It's about as secure as can be expected of a plain web app.
Only if you define "someone with access to my computer" to include "anyone who runs a web server I visit".
The article doesn't say that any web site can access the key, just the browser itself (via bookmarklets or third-party extensions) and Mega. Both of which are obvious.
The browser prevents sites from accessing other sites' local data. It would be interesting if they managed to find a way around that protection, but they didn't. The system is working as designed.
The private schools should survive on their own resources; parents sending their kids to private schools should not be able to direct education taxes towards them. This, again, goes to the universality of the benefit of an educated populace. All should pay, and only the publicly owned schools should receive the money.
You may want to rethink that. This isn't "all pay equally", this is "send your kids to public school or pay double". At the very least, parents who take direct responsibility for the education of the portion of the populace which they are personally responsible for (their own kids) should be able to count that toward their part of the cost of providing universal education. They shouldn't be forced to pay twice, once for the education their kids are actually getting and again for the one which would otherwise have been provided by the state.
Note that I'm not saying that tax money should go to private schools (like with vouchers). Rather, if the education part of your taxes amounts to the cost of publicly educating three kids, and you pay to send two kids to an accredited private school, or home school them according to the required curriculum, that should count as 2/3 of the taxes, since the public school has two fewer students to worry about. This would be capped at your share of the public education cost, no matter how much extra you paid (i.e. no refunds).
Parents will probably still end up paying more for a private school, but at least they won't have to pay for both, which should reduce the obstacles to sending kids to private schools and consequently improve the general level of education. The public schools, in the meantime, are still getting just as much money per student as before; they just don't get paid extra for students they aren't responsible for educating.