renewables don't do it because they always come combined with natural gas power plants
Care to explain why you think that solar, tide, hydro, wind would necessarily need to be combined with natural gas plants?
Hydro doesn't need to be, but just about every major river that can be dammed already is, and environmentalists aren't really happy about that. Like it or not, hydro isn't really going to replace much more of our fossil-fuel-based power generation.
The rest tend to be inconsistent so unless you have a way to store power (variations on hydro usually), you end up needing fossil-fuels in order to take up the slack. Diversity of sources will probably help, but only so far. You still end up with a ton of idle fossil-fuel plants even in the best case just so that if you get a week of cloudy non-windy weather you don't have blackouts. Nobody likes paying for idle plants, so the pressure is always there to run them and build fewer renewable plants.
The main problem with renewables is that for the most part they're just not ready yet unless you want a significant increase in energy costs. In some situations they're becoming competitive, but I've yet to hear about anybody who has a plan for having them handle baseline load for any significant area.
The big advantage of nuclear is that it works just like coal/etc - you fuel it up and you run it as much as you want to day or night, and you can build one right now. The main downside is that people have a lot of irrational fear about them. There are also what I'd consider legitimate concerns - from an engineering standpoint they certainly can be built/operated safely, but in practice there can be motivation to cut corners.
Absolutely. The government needs to stop regulating deals between private individuals. If Jim over here wants to sell himself into slavery so that his family can have some gruel to eat, that's his right. I have a nice little hut in the back for them to live in, and it is nice and safe with locks and barred windows. As long as they do good work I have every reason to keep them somewhat healthy.
"Molar water?" Even The Google has no idea what that is.
Perhaps you meant "pure water" or just "water?"
In any case, the resistivity is about 18 Mcm, which is pretty high. I would think that almost all the conductivity comes from H3O+ / OH- ions dissolved in the water, due to dissociation (which is why pure water has a pH of 7).
I'd think that impurities in an organic solvent of some kind would be far less of a problem than impurities in water. Water is a very polar solvent, and thus strongly promotes ionization of salts/etc. If you sprinkle salt in water you end up with sodium and chloride ions in solution, which are VERY conductive. If you sprinkle salt in toluene it will just sink to the bottom. Now, lots of organic compounds will dissolve just fine in toluene, but I imagine they'd make poor conductors.
That said, I can certainly believe that impurities would be a problem. A tiny spec of metal floating in the coolant could easily find its way into contact with a circuit of some kind, perhaps even inside a chip. I imagine that this sort of thing is much less likely to happen with air, which is much less viscous and thus can't hold anything but the lightest particles in suspension.
Well, the issue is not so much energy as it is power. A nuclear reactor doesn't really deliver any more power than a turbine - it just doesn't run out of fuel as often and doesn't require air (not an issue on the surface).
If your reactor delivers power in the MW range, and to fire at max speed your railgun needs power in the GW range, then you have a problem. Heat/wear has been an even bigger problem in the past as you've noted, but power is still a challenge if you want to sustain a high firing rate.
How about donating $1000 to maintain the ban on polygamy? Should such a person be fired? Or should they only be fired after it becomes accepted?
Also, people lobby all the time to create more potholes. What do you think cutting state Dept of Transportation spending is? Paying somebody to go out and dig holes in roads is a crime. Having an opinion on whether we should or shouldn't fill them in is just being involved.
I'm all for letting anybody get married/divorced who wants to get married/divorced. Lots of others disagree with me. I'm fine with letting them keep their jobs. If we fired everybody who was wrong about something, I'd be the only person left on the planet still employed.
With Linux, there's no reason to support 10-year-old releases unless some stodgy corporation is paying you big bucks to do so. When your release is getting old, just upgrade; it's free.
The problem with this is that it necessitates change. I don't mind it at all, but not all feel the same. You can't upgrade and keep your KDE 3.5. Gnome 3 is more recent so you can still avoid that, but I'm sure one day you won't.
If you're a big company running proprietary binaries it becomes more of a challenge to upgrade, which is why Red Hat gets paid big bucks to backport, but they'll only do it for 10 years from initial release (which is almost as good as Windows).
There are ways to generate true random numbers (people sell RNGs based on nuclear decay and thermal noise for several hundred dollars), and those will get you a true one-time pad.
I'd argue that those are ways to generate numbers that we think are random, but that only means that no pattern has been detected. I don't see any way to prove that one of these devices actually generates truly random numbers.
But otherwise I agree - strictly speaking a One Time Pad only works with random numbers. Perfectly implementing one is probably impossible, but it can of course be awfully good in practice.
I don't disagree with most of what you wrote, but most of the people still running XP would be having just as many problems if they were running Linux (maybe not if they were running something completely seamless like ChromeOS, but that doesn't have a 10 year track record yet so who knows).
Upgrade cost is probably not a big factor for most people running XP, at least not in the US (maybe in China/etc it is). The real issues are old hardware, not wanting to wipe/reinstall/reconfigure everything, and not wanting to deal with a lot of change. Any of those issues are potentially a problem with Linux as well.
Sure, the upgrade is free, and there are lots of options. You may still have problems with insufficient RAM if your system is old enough, though at least driver issues probably won't be a problem (on an old PC you probably don't care about 3D acceleration using proprietary drivers). Many distros still struggle with in-place upgrades to varying degrees, and unless you are running something like RHEL you're going to have to do more frequent update cycles than on Windows (potentially free, but still more hassle). Dealing with change is certainly a challenge on Linux, or so my Ubuntu-loving friends tell me (and I went through it back in the KDE 4.0 days as well).
I'm not knocking anybody here, either. This stuff is hard, or at least boring and unprofitable. The boring part keeps FOSS developers from backporting fixes to 10-year-old distro releases, and the unprofitable part keeps everybody else from doing it as well.
So, who is backporting security patches to linux 2.0, or KDE 3.0?
Linux 2.0 has a continuous sequence of security patches and updates through to the present day. Linux 2.4, 2.6... 3.6, 3.8. If you haven't kept up to date you that's not the fault of the linux developers. All the steps have been made available, for free, and continue to be. Which isn't the case with Windows XP.
Link? I'm not aware of anybody backporting patches to Linux 2.0. Certainly there are 2.6 longterm kernels still around.
All of the stuff you're saying is a great reason to buy Linux and not Windows. I just don't get how MS should have to support XP forever when they've always only guaranteed 10 years from date of obsolescence, which is what they've done.
If you bought a PC with XP then you've either gotten at least 10 years of support, or you intentionally chose not to use a newer version like Vista. And if you stuck with MS after seeing Vista, well, I'd say that MS is hardly the source of all of your problems...
Well, half the reason I ditched KDE for a few years was that KDE 4 was a huge performance hog in its early days. It has gotten better, but I doubt it is comparable to 3.5. But, once I upgraded my hardware I returned to it.
I also wouldn't say that Win7 is remarkably slow compared to XP. Then again, I haven't run it on any systems with very limited RAM.
MS supports their OSes for 10 years from obsolesce. They have had the same lifecycle for ages - anybody buying a PC with XP on it less than 10 years ago knew this date was coming. If they bought it more than 10 years ago then they knew that it would be good for at least 10 years, which was a promise that was kept.
Sure, I avoid Windows anytime I can. Still, I can't really say that MS doesn't support their products. I mean, should they also be supporting Win98, Win95, and Win 3.1? How about Win/386 and MS DOS 2.1 (now supporting sub-directories!)?
Sure, you can do it, but is it really fair to EXPECT a vendor to support an OS for so long? MS provides free security support for Windows for longer than any other consumer OS vendor out there. RHEL is probably the next closest option I'm aware of.
I'm a huge proponent of Linux for many reasons, including the ability to self-support. However, I think it is a bit much to demand that a vendor continue to support an OS that came out in 2001.
So yes, Linux IS still recieving backports of fixes.
And keep in mind 2.2.0 is from 1999 and 2.4.0 is from 2001. 2.6.0 came out in 2003.
Some comparing off that, Linux *IS* actively supporting software as old as Windows XP.
What distro are you thinking of? The Linux kernel only has longterm support back to 2.6.32. RHEL offers 10 years from initial release, which is almost as good as MS (10 years from next version release).
Nobody said the fee would be "nominal." For just about anybody the cost to maintain support is going to be less than the cost of just buying a new computer. Companies don't buy extended support because they're too cheap to replace their PCs - they do it as an interim measure to delay having to update internal applications.
I bought a netbook in 2010 that had XP on it so support until 2020 please.
You do realize that Windows is already the longest-supported consumer operating system around, right? Not even RHEL comes close.
They support their OS for 10 years from the day the replacement version (Vista in this case) was released.
I don't really have a problem with the 10-years-from-last-sale thing, but that probably wouldn't be that big of a deal for MS to deal with, since they're already not all that far off (even your 2010 netbook has been supported for 4 years). The folks who will really have trouble are those like Apple/Samsung/etc selling smartphones. Apple is the leader here as far as I can see, but even they will drop support a year or two after the last sale (a few years after first sale). Google supports their Nexus phones for about 1.5 years after first sale, or about 6 months after last sale. Most of the other Android vendors barely support their phones at all, with them often being obsolete while they're still being sold.
Now, support would mean security updates, not new features, but 10 years from last sale would be a revolution in terms of expectations in the software industry.
That probably wouldn't work for software. For starters, it can often be delivered online so now you need a great firewall to block it. Even for physical shipments, if the OS sells for $100 and the box costs $2 to produce, then they can afford the odd seizure at the port if it is sold direct to consumer.
The FDA has a lot of trouble intercepting shipments of pharmaceuticals to individuals for just that reason. You would never get a pallet of them through without documentation, but how many nondescript 8" boxes declared as containing toys are they going to open at the docks?
Perhaps if Microsoft hadn't developed a series of OS's that treated security like a two-bit whore, none of this would be a major issue. Proprietary software can be done right, with minimal effort to support it for decades.
So, who is backporting security patches to linux 2.0, or KDE 3.0?
No OS is free from vulnerabilities. Sure, I'd take Linux over Windows any day, but if you're running ancient FOSS you probably have a ton of vulnerabilities that you'd never be able to keep up with on your own.
Sure, lots of proprietary software is supported for decades, but most of that stuff just doesn't get much attention from the hacking community. Just look at SCADA - it was considered secure for ages until some government decided it was worth going after, and then you had the mother of all worms come along. Since then I haven't heard a peep, and it isn't because it is any more secure. The fact that the widget inventory system at your warehouse has never had a security update doesn't mean that it doesn't contain vulnerabilities.
Due to its nature, Windows will always be a target (well, until everybody finally gives up on using it). The same is true of Linux and OSX, and if you run a version of either that isn't being actively maintained then you're going to be vulnerable.
one-time pad... if properly implemented. Big, big if. Barriers are almost insurmountable unless you are very paranoid and have lots of resources.
It really depends on what you want to do with it. If your goal is to upload HD movies to your friends, then yes you're going to be spending a LOT of time on key generation and management.
On the other hand, if you're just trading the odd short message, then 1MB of random data will last you quite a while and that isn't too hard to generate with a very strong PRNG. If you want to pull numbers out of a hat one at a time it is a bit more of a pain. Really the RNG is probably the biggest practical limitation, assuming that the amount of data to encrypt in the future is much smaller than your capacity to store key data.
It just doesn't have anywhere near the convenience of public key crypto, however.
Yup. AES-256 will only fall if somebody finds an algorithmic weakness that reduces the complexity to something lower than brute force or something like a quantum algorithm.
Also, there is always the one-time pad. That is completely invulnerable to brute-force attack if properly implemented.
Yup. If you have a guard check somebody's fingerprints it would be extremely difficult to sneak through. If you stick a fingerprint scanner next to a door in an empty building, that is a different story.
The guard isn't too likely to be fooled by a gummy bear...
I'm not saying it isn't possible, but depending where on the protocol stack the USB port is intercepted, it might still be vulnerable. You also introduce the risk of vulnerabilities in your antivirus software (which is probably closed-source), and the risk of breaking things if you deploy a bad update (why would an OS update require testing, but an antivirus update not?).
Antivirus really seems like a technical solution to a non-technical problem: unresponsive software vendors. It lets you pick MS for their UI, but somebody else for their support.
In the US a contract that involves some required length of service is not easily broken either. However, the thing about a contract is that neither party has the right to unilaterally change it. Adding a new fee to a contract (the cancellation fee) would certainly be considered a material change by a court, and that means that either the company would be forced to honor the original terms, or the contract would be completely void. A court is basically going to do whatever seems fair in such a situation.
This is true of any contract. If I sign a contract that forced me to buy a car from you for $4000, and you want to charge me $4001, then you'd find it almost impossible to get a court to force me to uphold the new deal. Now, if I noticed a bug splat on the window that wasn't previously there a court might still force me to buy it - there is no permanent damage and the gist of the contract remains intact. That would be considered a non-material change to the deal. The same would probably be true if you asked me to meet you at a different office to close the deal down the street from the originally-agreed-upon office. Courts try to look at a change and decide if it was the sort of thing that might cause somebody to rethink the deal in the first place. Money is almost always considered a material change in a contract, regardless of how little.
So, when AT&T or whoever gives you a GSM smartphone for $50 with a two year contract, and then two months later tells you they are raising rates by 50 cents a month, if you threaten them enough you might be able to break your contract and sell the phone to somebody else at a considerable profit. If they don't let you out you might be forced to dispute charges, or go to court over it (and possibly end up with something on your credit history - which is wrong, but...).
I suspect that even in the US a court would tend to take a negative view of such tactics.
The problem is that it never goes to court. They'll charge you, and now you have to try to dispute the charge on your credit card bill. If you manage to dispute it successfully, then you'll get a negative letter in your credit history. If you fail, then you're out the money and you have to sue to get it back, or you could refuse to pay your credit card bill and again have a negative letter in your credit history.
You do have the right to write a letter in your credit history explaining your side of the dispute. However, nobody will bother to read it, because companies don't care if you were ripped off or not. Most likely, they plan on ripping you off someday as well, and the fact that you don't let yourself be pushed around is reason enough to charge you a higher interest rate, or refuse to do business with you.
So, then you can try to go to court to get the record taken out of your credit history, but then everybody yells free speech, since all the record says is that you didn't pay a bill which you disputed, which is factually correct.
Basically the whole US credit system is based around effectively moving justice from the courts to a reputation system controlled by big companies.
Not exactly controversial, though the thing about missile defense is that you usually have to engage them when they're well out over international waters if you want to prevent them from hitting you. So, there is a certain element of aggressiveness as you're attacking something that has not yet violated your territory.
I'm not sure where the boost phase of these things ends either. During the boost phase itself you can only tell what general direction it is going in - if fired over your nation then the ballistic trajectory will at first fall short, then fall on your territory, and then overshoot your territory. You don't know where it will fall until the boost is over. Of course, you can always just fire on it as soon as it is likely to come close or overfly - if it never gets enough energy there is no risk of it hitting you.
More like the Gulf of Thailand, but either way most of it is international waters. There is the whole Spratly Island thing, but that isn't near where MH370 lost contact, and everybody and their uncle is sailing through there just to try to assert their claims.
Slashdot Mirror
renewables don't do it because they always come combined with natural gas power plants
Care to explain why you think that solar, tide, hydro, wind would necessarily need to be combined with natural gas plants?
Hydro doesn't need to be, but just about every major river that can be dammed already is, and environmentalists aren't really happy about that. Like it or not, hydro isn't really going to replace much more of our fossil-fuel-based power generation.
The rest tend to be inconsistent so unless you have a way to store power (variations on hydro usually), you end up needing fossil-fuels in order to take up the slack. Diversity of sources will probably help, but only so far. You still end up with a ton of idle fossil-fuel plants even in the best case just so that if you get a week of cloudy non-windy weather you don't have blackouts. Nobody likes paying for idle plants, so the pressure is always there to run them and build fewer renewable plants.
The main problem with renewables is that for the most part they're just not ready yet unless you want a significant increase in energy costs. In some situations they're becoming competitive, but I've yet to hear about anybody who has a plan for having them handle baseline load for any significant area.
The big advantage of nuclear is that it works just like coal/etc - you fuel it up and you run it as much as you want to day or night, and you can build one right now. The main downside is that people have a lot of irrational fear about them. There are also what I'd consider legitimate concerns - from an engineering standpoint they certainly can be built/operated safely, but in practice there can be motivation to cut corners.
Absolutely. The government needs to stop regulating deals between private individuals. If Jim over here wants to sell himself into slavery so that his family can have some gruel to eat, that's his right. I have a nice little hut in the back for them to live in, and it is nice and safe with locks and barred windows. As long as they do good work I have every reason to keep them somewhat healthy.
"Molar water?" Even The Google has no idea what that is.
Perhaps you meant "pure water" or just "water?"
In any case, the resistivity is about 18 Mcm, which is pretty high. I would think that almost all the conductivity comes from H3O+ / OH- ions dissolved in the water, due to dissociation (which is why pure water has a pH of 7).
I'd think that impurities in an organic solvent of some kind would be far less of a problem than impurities in water. Water is a very polar solvent, and thus strongly promotes ionization of salts/etc. If you sprinkle salt in water you end up with sodium and chloride ions in solution, which are VERY conductive. If you sprinkle salt in toluene it will just sink to the bottom. Now, lots of organic compounds will dissolve just fine in toluene, but I imagine they'd make poor conductors.
That said, I can certainly believe that impurities would be a problem. A tiny spec of metal floating in the coolant could easily find its way into contact with a circuit of some kind, perhaps even inside a chip. I imagine that this sort of thing is much less likely to happen with air, which is much less viscous and thus can't hold anything but the lightest particles in suspension.
Well, the issue is not so much energy as it is power. A nuclear reactor doesn't really deliver any more power than a turbine - it just doesn't run out of fuel as often and doesn't require air (not an issue on the surface).
If your reactor delivers power in the MW range, and to fire at max speed your railgun needs power in the GW range, then you have a problem. Heat/wear has been an even bigger problem in the past as you've noted, but power is still a challenge if you want to sustain a high firing rate.
How about donating $1000 to maintain the ban on polygamy? Should such a person be fired? Or should they only be fired after it becomes accepted?
Also, people lobby all the time to create more potholes. What do you think cutting state Dept of Transportation spending is? Paying somebody to go out and dig holes in roads is a crime. Having an opinion on whether we should or shouldn't fill them in is just being involved.
I'm all for letting anybody get married/divorced who wants to get married/divorced. Lots of others disagree with me. I'm fine with letting them keep their jobs. If we fired everybody who was wrong about something, I'd be the only person left on the planet still employed.
With Linux, there's no reason to support 10-year-old releases unless some stodgy corporation is paying you big bucks to do so. When your release is getting old, just upgrade; it's free.
The problem with this is that it necessitates change. I don't mind it at all, but not all feel the same. You can't upgrade and keep your KDE 3.5. Gnome 3 is more recent so you can still avoid that, but I'm sure one day you won't.
If you're a big company running proprietary binaries it becomes more of a challenge to upgrade, which is why Red Hat gets paid big bucks to backport, but they'll only do it for 10 years from initial release (which is almost as good as Windows).
There are ways to generate true random numbers (people sell RNGs based on nuclear decay and thermal noise for several hundred dollars), and those will get you a true one-time pad.
I'd argue that those are ways to generate numbers that we think are random, but that only means that no pattern has been detected. I don't see any way to prove that one of these devices actually generates truly random numbers.
But otherwise I agree - strictly speaking a One Time Pad only works with random numbers. Perfectly implementing one is probably impossible, but it can of course be awfully good in practice.
I don't disagree with most of what you wrote, but most of the people still running XP would be having just as many problems if they were running Linux (maybe not if they were running something completely seamless like ChromeOS, but that doesn't have a 10 year track record yet so who knows).
Upgrade cost is probably not a big factor for most people running XP, at least not in the US (maybe in China/etc it is). The real issues are old hardware, not wanting to wipe/reinstall/reconfigure everything, and not wanting to deal with a lot of change. Any of those issues are potentially a problem with Linux as well.
Sure, the upgrade is free, and there are lots of options. You may still have problems with insufficient RAM if your system is old enough, though at least driver issues probably won't be a problem (on an old PC you probably don't care about 3D acceleration using proprietary drivers). Many distros still struggle with in-place upgrades to varying degrees, and unless you are running something like RHEL you're going to have to do more frequent update cycles than on Windows (potentially free, but still more hassle). Dealing with change is certainly a challenge on Linux, or so my Ubuntu-loving friends tell me (and I went through it back in the KDE 4.0 days as well).
I'm not knocking anybody here, either. This stuff is hard, or at least boring and unprofitable. The boring part keeps FOSS developers from backporting fixes to 10-year-old distro releases, and the unprofitable part keeps everybody else from doing it as well.
So, who is backporting security patches to linux 2.0, or KDE 3.0?
Linux 2.0 has a continuous sequence of security patches and updates through to the present day. Linux 2.4, 2.6 ... 3.6, 3.8. If you haven't kept up to date you that's not the fault of the linux developers. All the steps have been made available, for free, and continue to be. Which isn't the case with Windows XP.
Link? I'm not aware of anybody backporting patches to Linux 2.0. Certainly there are 2.6 longterm kernels still around.
All of the stuff you're saying is a great reason to buy Linux and not Windows. I just don't get how MS should have to support XP forever when they've always only guaranteed 10 years from date of obsolescence, which is what they've done.
If you bought a PC with XP then you've either gotten at least 10 years of support, or you intentionally chose not to use a newer version like Vista. And if you stuck with MS after seeing Vista, well, I'd say that MS is hardly the source of all of your problems...
Well, half the reason I ditched KDE for a few years was that KDE 4 was a huge performance hog in its early days. It has gotten better, but I doubt it is comparable to 3.5. But, once I upgraded my hardware I returned to it.
I also wouldn't say that Win7 is remarkably slow compared to XP. Then again, I haven't run it on any systems with very limited RAM.
MS supports their OSes for 10 years from obsolesce. They have had the same lifecycle for ages - anybody buying a PC with XP on it less than 10 years ago knew this date was coming. If they bought it more than 10 years ago then they knew that it would be good for at least 10 years, which was a promise that was kept.
Sure, I avoid Windows anytime I can. Still, I can't really say that MS doesn't support their products. I mean, should they also be supporting Win98, Win95, and Win 3.1? How about Win/386 and MS DOS 2.1 (now supporting sub-directories!)?
Sure, you can do it, but is it really fair to EXPECT a vendor to support an OS for so long? MS provides free security support for Windows for longer than any other consumer OS vendor out there. RHEL is probably the next closest option I'm aware of.
I'm a huge proponent of Linux for many reasons, including the ability to self-support. However, I think it is a bit much to demand that a vendor continue to support an OS that came out in 2001.
So yes, Linux IS still recieving backports of fixes.
And keep in mind 2.2.0 is from 1999 and 2.4.0 is from 2001. 2.6.0 came out in 2003.
Some comparing off that, Linux *IS* actively supporting software as old as Windows XP.
What distro are you thinking of? The Linux kernel only has longterm support back to 2.6.32. RHEL offers 10 years from initial release, which is almost as good as MS (10 years from next version release).
Nobody said the fee would be "nominal." For just about anybody the cost to maintain support is going to be less than the cost of just buying a new computer. Companies don't buy extended support because they're too cheap to replace their PCs - they do it as an interim measure to delay having to update internal applications.
I bought a netbook in 2010 that had XP on it so support until 2020 please.
You do realize that Windows is already the longest-supported consumer operating system around, right? Not even RHEL comes close.
They support their OS for 10 years from the day the replacement version (Vista in this case) was released.
I don't really have a problem with the 10-years-from-last-sale thing, but that probably wouldn't be that big of a deal for MS to deal with, since they're already not all that far off (even your 2010 netbook has been supported for 4 years). The folks who will really have trouble are those like Apple/Samsung/etc selling smartphones. Apple is the leader here as far as I can see, but even they will drop support a year or two after the last sale (a few years after first sale). Google supports their Nexus phones for about 1.5 years after first sale, or about 6 months after last sale. Most of the other Android vendors barely support their phones at all, with them often being obsolete while they're still being sold.
Now, support would mean security updates, not new features, but 10 years from last sale would be a revolution in terms of expectations in the software industry.
That probably wouldn't work for software. For starters, it can often be delivered online so now you need a great firewall to block it. Even for physical shipments, if the OS sells for $100 and the box costs $2 to produce, then they can afford the odd seizure at the port if it is sold direct to consumer.
The FDA has a lot of trouble intercepting shipments of pharmaceuticals to individuals for just that reason. You would never get a pallet of them through without documentation, but how many nondescript 8" boxes declared as containing toys are they going to open at the docks?
Perhaps if Microsoft hadn't developed a series of OS's that treated security like a two-bit whore, none of this would be a major issue. Proprietary software can be done right, with minimal effort to support it for decades.
So, who is backporting security patches to linux 2.0, or KDE 3.0?
No OS is free from vulnerabilities. Sure, I'd take Linux over Windows any day, but if you're running ancient FOSS you probably have a ton of vulnerabilities that you'd never be able to keep up with on your own.
Sure, lots of proprietary software is supported for decades, but most of that stuff just doesn't get much attention from the hacking community. Just look at SCADA - it was considered secure for ages until some government decided it was worth going after, and then you had the mother of all worms come along. Since then I haven't heard a peep, and it isn't because it is any more secure. The fact that the widget inventory system at your warehouse has never had a security update doesn't mean that it doesn't contain vulnerabilities.
Due to its nature, Windows will always be a target (well, until everybody finally gives up on using it). The same is true of Linux and OSX, and if you run a version of either that isn't being actively maintained then you're going to be vulnerable.
one-time pad ... if properly implemented.
Big, big if. Barriers are almost insurmountable unless you are very paranoid and have lots of resources.
It really depends on what you want to do with it. If your goal is to upload HD movies to your friends, then yes you're going to be spending a LOT of time on key generation and management.
On the other hand, if you're just trading the odd short message, then 1MB of random data will last you quite a while and that isn't too hard to generate with a very strong PRNG. If you want to pull numbers out of a hat one at a time it is a bit more of a pain. Really the RNG is probably the biggest practical limitation, assuming that the amount of data to encrypt in the future is much smaller than your capacity to store key data.
It just doesn't have anywhere near the convenience of public key crypto, however.
Yup. AES-256 will only fall if somebody finds an algorithmic weakness that reduces the complexity to something lower than brute force or something like a quantum algorithm.
Also, there is always the one-time pad. That is completely invulnerable to brute-force attack if properly implemented.
Yup. If you have a guard check somebody's fingerprints it would be extremely difficult to sneak through. If you stick a fingerprint scanner next to a door in an empty building, that is a different story.
The guard isn't too likely to be fooled by a gummy bear...
I'm not saying it isn't possible, but depending where on the protocol stack the USB port is intercepted, it might still be vulnerable. You also introduce the risk of vulnerabilities in your antivirus software (which is probably closed-source), and the risk of breaking things if you deploy a bad update (why would an OS update require testing, but an antivirus update not?).
Antivirus really seems like a technical solution to a non-technical problem: unresponsive software vendors. It lets you pick MS for their UI, but somebody else for their support.
In the US a contract that involves some required length of service is not easily broken either. However, the thing about a contract is that neither party has the right to unilaterally change it. Adding a new fee to a contract (the cancellation fee) would certainly be considered a material change by a court, and that means that either the company would be forced to honor the original terms, or the contract would be completely void. A court is basically going to do whatever seems fair in such a situation.
This is true of any contract. If I sign a contract that forced me to buy a car from you for $4000, and you want to charge me $4001, then you'd find it almost impossible to get a court to force me to uphold the new deal. Now, if I noticed a bug splat on the window that wasn't previously there a court might still force me to buy it - there is no permanent damage and the gist of the contract remains intact. That would be considered a non-material change to the deal. The same would probably be true if you asked me to meet you at a different office to close the deal down the street from the originally-agreed-upon office. Courts try to look at a change and decide if it was the sort of thing that might cause somebody to rethink the deal in the first place. Money is almost always considered a material change in a contract, regardless of how little.
So, when AT&T or whoever gives you a GSM smartphone for $50 with a two year contract, and then two months later tells you they are raising rates by 50 cents a month, if you threaten them enough you might be able to break your contract and sell the phone to somebody else at a considerable profit. If they don't let you out you might be forced to dispute charges, or go to court over it (and possibly end up with something on your credit history - which is wrong, but...).
I suspect that even in the US a court would tend to take a negative view of such tactics.
The problem is that it never goes to court. They'll charge you, and now you have to try to dispute the charge on your credit card bill. If you manage to dispute it successfully, then you'll get a negative letter in your credit history. If you fail, then you're out the money and you have to sue to get it back, or you could refuse to pay your credit card bill and again have a negative letter in your credit history.
You do have the right to write a letter in your credit history explaining your side of the dispute. However, nobody will bother to read it, because companies don't care if you were ripped off or not. Most likely, they plan on ripping you off someday as well, and the fact that you don't let yourself be pushed around is reason enough to charge you a higher interest rate, or refuse to do business with you.
So, then you can try to go to court to get the record taken out of your credit history, but then everybody yells free speech, since all the record says is that you didn't pay a bill which you disputed, which is factually correct.
Basically the whole US credit system is based around effectively moving justice from the courts to a reputation system controlled by big companies.
Any appearance of bigotry so high up the chain of command cannot be tolerated if an organization truly wishes to be inclusive.
You're suggesting a zero-tolerance policy for intolerance?
Not exactly controversial, though the thing about missile defense is that you usually have to engage them when they're well out over international waters if you want to prevent them from hitting you. So, there is a certain element of aggressiveness as you're attacking something that has not yet violated your territory.
I'm not sure where the boost phase of these things ends either. During the boost phase itself you can only tell what general direction it is going in - if fired over your nation then the ballistic trajectory will at first fall short, then fall on your territory, and then overshoot your territory. You don't know where it will fall until the boost is over. Of course, you can always just fire on it as soon as it is likely to come close or overfly - if it never gets enough energy there is no risk of it hitting you.
More like the Gulf of Thailand, but either way most of it is international waters. There is the whole Spratly Island thing, but that isn't near where MH370 lost contact, and everybody and their uncle is sailing through there just to try to assert their claims.