So can you run Linux on it then? You could ssh to it from Windows and get some real work done. According to TFA it does run Linux: To put it simply, the Killer NIC is actually a Linux system-on-chip. That basically means that it is its own little Linux computer inside what is likely your Windows box. The engineers at Bigfoot have given you access to this machine.
Fit the stolen car with your own number plates complete with relevant RFID information and presto! A digital signature on the RFID tag should be sufficient to prevent that.
On tolled roads another option could be to have speed-dependent pricing: Have the price go up when exceeding the desired most common speed, and have an absolute speed limit a bit above that. A few people will be willing to pay the additional price for the additional speed, while the majority will just keep at the maximum normal-price speed instead.
But companies (and their consumers, and their government) want to have cars that go 200 MPH, even if deaths increase as a result. You hear arguments like that all the time. But I've checked it: Before cars were invented, the death rate was 1 death per person. There's absolutely no indication that cars have increased that rate. Indeed, the fact that many car owners are still alive may be seen as an indication that owning cars actually decreases death rates.
Troll?! Don't you guys have a sense of humor? You must be new here. Slashdot humor always takes forms like "In Soviet Russia...", "Imagine a Beowulf cluster..." or "1.... 2. ??? 3. Profit!"
Further, it appears to work very well for them, without sacrificing their employees ability to work.
Of course should it not work well, Microsoft wouldn't tell you. Or would you really expect them to say "well, we have security problems caused by this MS product..."? There are a lot of reasons why they won't do that. First, it would of course make bad advertising for the products. Second, it would also make bad advertising of MS itself (along the lines of "they can't even keep their own network safe"). And third, it would give attackers a hint where to target their attack.
So the fact that MS claims that it works well in reality doesn't tell you much. Maybe it works well, maybe it doesn't, but all you really can say is that it isn't so bad that they can't cover all problems which might occur.
Geothermal and tidal energy are two non-nuclear energy sources that spring to mind. AFAIK geothermal energy comes from the decay of radioactive nuclei inside the earth. The original heat from formation of the earth (which was indeed gravitational) has radiated away long ago.
Not all our energy is nuclear: Tidal power plants don't use nuclear power.
However it's quite obvious that the sun must be shut down as quickly as possible: First, as you aready said, it's using nuclear energy, and of course nuclear energy is known to be bad. But for the sun, it's not even abstract: The sun is known to continuously send radioactive radiation. Fortunately the earths magnetic field and atmosphere are saving us from most of it, but what if the magnetic field fails? Also note that we are already quite certain that the sun will end up destroying all life on earth when (not: if) it finally fails. So we really shouldn't tolerate such a dangerous nuclear reactor so close to earth.:-)
The problem with solar panels is PRICE. That is the nut that needs to be cracked. If a panel is only 10% efficient, but not much more expensive than a similar sized roofing slate, it suddenly becomes economical to re-roof the house the next time the roof needs attention. However, it's not even near that price. An 80 watt panel currently costs around GBP£250 which is fantastically expensive. There is no way I could even remotely afford solar roofing at those prices, however much I'd like it. The price per peak watt needs to be about a tenth of what it is now.
According to the article, the new solar cells would cost about $3/watt. I don't really know the exchange rate between $ and £, but AFAIR it's about a factor of two. That would give £120 for an 80 watt panel. Still not 1/10th, but at least less than half the price.
Also note that in your calculation you also have to add the savings in electricity. That is, you don't have to have equal price to conventional roofing, but you have to have the equal proce of conventional roofing plus the saved payment of electricity in some reasonable amount of time. So if the money saved on your electricity bill in, say, the next three years makes up for the extra cost of solar cell roofing, it's still more profitable to do it. (Well, actually you'll also have to account for the interest you would have gotten for the money during that time, because the money you pay for electricity next year will still give interest this year, while the money you spent on the solar panel will not.)
No, the odd-numbered Googles are the development versions (basically, almost everything is in beta), while the even-numbered ones are the release versions.
Except that for operator overloading, I've never seen any argument which cannot be used for named functions as well. Indeed, I've only ever seen one argument: Operators not doing what the operator symbol suggests are confusing. Of course they are, but that's the same with functions:
bigint add(bigint a, bigint b) {
bigint result; // code setting result to the product of a and b
return result; }
There somewhat is. There's a namespace inside std that includes templated functions for ==, !=, >, >=, and = is !; == is !; etc.)
It's not very fine-grained unfortunately; like you can't specify "I want to provide the other operators for this type but not this type" I think. You can specify it (you can even specify that you only want some of them, e.g. only operator!=). It's just not straightforward:
namespace helper {
using std::relops::operator!=;
using std::relops::operator>;// etc.
struct use_relops {}; }
class MyClass: helper::use_relops {//... };
Now, the templates from std::relops will be found through Koenig lookup for expressions involving MyClass instances (or instances of classes derived from MyClass), but will not be found otherwise (unless you've brought them into the current scope through another way, of course).
Is it weird at all that the Friends theme song started to play in my head as I read that comment?We've found out that you've placed an unauthorized copy of a copyrighted song in your head. Please delete that song and pay us $1000, or we will sue you. The RIAA.
Of course everything ever said about experts is questionable, because either it's said by the experts themselves and thus biased, or it's said by non-experts and thus unreliable.:-)
I'm still waiting for the computer with one button: "Do What I Mean"
That's a terrible interface. It gives you the option of the computer either doing what you mean (by pressing the button), or the computer not doing what you mean (by not pressing the button). Do you really need that choice? I don't ever want the computer not to do what I mean. Therefore the ideal computer has no button. It just always does what I mean.
He doesn't deny that different users are a good thing (that's why he includes the "change user" into the lock mode). However I'm not sure how he imagines this combination to work. I see two possibilities:
Either changing user from within lock first logs user 1 off, and then logs user 2 in. But then, what do you do if you definitively don't want toget logged out for some reason?
Or the second user gets logged in while the first user's programs still continue running. But then, without a logoff option user 1's programs might unnessecarily continue to run and eat ressources from user 2. And no, rebooting (or power off and on again) might not be an option because user 3 might still have programs running which he does not want to have terminated.
Of course, user 2 could just end all his programs before locking the screen, but he might not want to do all that work (and besides, there may be background processes running for him which he does not even know about).
The problem is that he has fallen for the much to common fallacy that the opposite of the wrong must be the right. To provide as many choices as possible is obviously wrong, therefore he thinks the opposite, that is to provide as little choice as possible, must be right.
What about the following rule?
The right amount of choice is best.
Of course that's a rule which isn't as easily followed as either "provide as much choice as possible" or "provide as little choice as possible", but doing things right is almost always a bit harder than just going to one extreme.
Slashdot Mirror
To put it simply, the Killer NIC is actually a Linux system-on-chip. That basically means that it is its own little Linux computer inside what is likely your Windows box. The engineers at Bigfoot have given you access to this machine.
Of course also quantum mechanics doesn't allow you to do FTL communication. See http://en.wikipedia.org/wiki/No-communication_theo rem
It has a built-in time machine which sends the packets a few milliseconds to the past.
Fit the stolen car with your own number plates complete with relevant RFID information and presto! A digital signature on the RFID tag should be sufficient to prevent that.
On tolled roads another option could be to have speed-dependent pricing: Have the price go up when exceeding the desired most common speed, and have an absolute speed limit a bit above that. A few people will be willing to pay the additional price for the additional speed, while the majority will just keep at the maximum normal-price speed instead.
SCNR
Of course should it not work well, Microsoft wouldn't tell you. Or would you really expect them to say "well, we have security problems caused by this MS product
So the fact that MS claims that it works well in reality doesn't tell you much. Maybe it works well, maybe it doesn't, but all you really can say is that it isn't so bad that they can't cover all problems which might occur.
So the sun powers the revolution of the moon around the earth? How?
Not all our energy is nuclear: Tidal power plants don't use nuclear power.
:-)
However it's quite obvious that the sun must be shut down as quickly as possible: First, as you aready said, it's using nuclear energy, and of course nuclear energy is known to be bad. But for the sun, it's not even abstract: The sun is known to continuously send radioactive radiation. Fortunately the earths magnetic field and atmosphere are saving us from most of it, but what if the magnetic field fails? Also note that we are already quite certain that the sun will end up destroying all life on earth when (not: if) it finally fails. So we really shouldn't tolerate such a dangerous nuclear reactor so close to earth.
According to the article, the new solar cells would cost about $3/watt. I don't really know the exchange rate between $ and £, but AFAIR it's about a factor of two. That would give £120 for an 80 watt panel. Still not 1/10th, but at least less than half the price.
Also note that in your calculation you also have to add the savings in electricity. That is, you don't have to have equal price to conventional roofing, but you have to have the equal proce of conventional roofing plus the saved payment of electricity in some reasonable amount of time. So if the money saved on your electricity bill in, say, the next three years makes up for the extra cost of solar cell roofing, it's still more profitable to do it. (Well, actually you'll also have to account for the interest you would have gotten for the money during that time, because the money you pay for electricity next year will still give interest this year, while the money you spent on the solar panel will not.)
No, the odd-numbered Googles are the development versions (basically, almost everything is in beta), while the even-numbered ones are the release versions.
It's not very fine-grained unfortunately; like you can't specify "I want to provide the other operators for this type but not this type" I think. You can specify it (you can even specify that you only want some of them, e.g. only operator!=). It's just not straightforward:
namespace helper
{
using std::relops::operator!=;
using std::relops::operator>;
struct use_relops {};
}
class MyClass: helper::use_relops
{
};
Now, the templates from std::relops will be found through Koenig lookup for expressions involving MyClass instances (or instances of classes derived from MyClass), but will not be found otherwise (unless you've brought them into the current scope through another way, of course).
Is it weird at all that the Friends theme song started to play in my head as I read that comment?We've found out that you've placed an unauthorized copy of a copyrighted song in your head. Please delete that song and pay us $1000, or we will sue you. The RIAA.
My AC days are over... I'm singing up right now!Hey, this is Slashdot, not the Eurovision Song Contest! :-)
Of course everything ever said about experts is questionable, because either it's said by the experts themselves and thus biased, or it's said by non-experts and thus unreliable. :-)
Just in case you didn't know that particular fact about cell phones: You can switch them off!
Modern cell phones tend to be small, don't they? So what would be an appropriate name for a small phone? ;-)
Obviously: Microphone!
And of course you cut and paste stuff on your computer without ever using scissors or glue ...
That's a terrible interface. It gives you the option of the computer either doing what you mean (by pressing the button), or the computer not doing what you mean (by not pressing the button). Do you really need that choice? I don't ever want the computer not to do what I mean. Therefore the ideal computer has no button. It just always does what I mean.
He doesn't deny that different users are a good thing (that's why he includes the "change user" into the lock mode). However I'm not sure how he imagines this combination to work. I see two possibilities:
Either changing user from within lock first logs user 1 off, and then logs user 2 in. But then, what do you do if you definitively don't want toget logged out for some reason?
Or the second user gets logged in while the first user's programs still continue running. But then, without a logoff option user 1's programs might unnessecarily continue to run and eat ressources from user 2. And no, rebooting (or power off and on again) might not be an option because user 3 might still have programs running which he does not want to have terminated.
Of course, user 2 could just end all his programs before locking the screen, but he might not want to do all that work (and besides, there may be background processes running for him which he does not even know about).
The problem is that he has fallen for the much to common fallacy that the opposite of the wrong must be the right. To provide as many choices as possible is obviously wrong, therefore he thinks the opposite, that is to provide as little choice as possible, must be right.
What about the following rule?
The right amount of choice is best.
Of course that's a rule which isn't as easily followed as either "provide as much choice as possible" or "provide as little choice as possible", but doing things right is almost always a bit harder than just going to one extreme.
If it's a Firefox-based WebOS, then you can just install Adblock.