No. There is no "safe" place to shoot somebody. People have died from being shot in the foot. The purpose of firing a gun at a human being is to stop a threat. That may or may not involve killing the person. Whether they die is irrelevant, whether they have been stopped is the whole point. To that end, the correct target to aim at is the center of the chest.
A whole lot can go wrong when you fire a gun. Unless you 1) feel that a person absolutely must be stopped, 2) you are willing to potentially kill that person in order to stop them, and 3) you feel that the risk of injuring somebody who is not the target is justified by the threat, then you should not be pulling the trigger. "Shoot to wound" is something from the movies.
If all firearms disappeared over night, murders would then occur using some other kind of weapon.
If you want to argue that high-fire-rate weapons can kill a lot of people quickly, I invite you to imagine a person wielding a sword in a crowded shopping mall.
Anybody who thinks a knife is less deadly than a gun has obviously never tried to defend themselves against a knife attack. You have to get closer to the victim, true... But that's not of much consequence, as the victim is probably unarmed, and in this case they were little kids.
Are JPEG images necessarly tied to pixels? If they store images as just a bunch of cosine waves added together, can't they in theory be resolution independent?
Yes, in theory that is possible. But rendering the JPEG directly at a larger size isn't going to look any better than rendering it at the usual size and then upsampling it. In fact, it would probably look worse, because if you upsample, you at least get some interpolation that might help smooth the JPEG block artifacts -- if you render the image directly to a larger block size it'll probably make the block discontinuities look even worse (more visible).
You want to implement memory protection on a CPU that has no MMU. You could implement this in software by putting the CPU into perpetual single-step mode, and intercepting all memory accesses to implement the MMU in software. But this will be many thousands of times slower than a hardware MMU.
No, I don't believe everything I read, but the story is that his neighbor was complaining about his dogs, then his dogs were shot dead by somebody and no more than a day later the guy who complained about the dogs turned up dead. If that is what happened, don't you think it's deserving of some investigation at least?
Careful, dude. There are a lot of things that "can be implemented in software" but are much better achieved in hardware. I see no reason to prevent patents on hardware simply because the problem being solved could also be solved by some (possibly very hokey, inefficient) software.
Or, maybe neither of the phenomena causes the other. Maybe the eggs which hatch last are laid last, and eggs which are laid last are not as thick as eggs laid first, due to depletion of calcium in the mother's blood from laying previous eggs, thereby affecting the rate of gas exchange during incubation, which modulates brain development, yadda yadda...
If possible, design a component such that no errors will ever occur (except for hardware failure). An entire program can't be designed that way, obviously, but individual components can be. Collecting all resource allocation into a small, well defined set of locations will relieve the majority of other code from the need to handle errors. In such a design, the majority of functions will return "void".
Only my wife was a defendant. And it's not like being there would have helped. When the insurance company provides an attorney for you, the attorney is really working for the insurance company, not you. They are going to do whatever the insurance company thinks is best. I tried several times to provide information and influence the lawyer's decision making and was politely ignored.
It sucks, but we would have been more money out of pocket had we hired our own lawyer. As it happened we didn't pay a single dollar for any of this crap.
How on earth was she supposed to prove it? Let me give the context for this. My wife and I met with a certain kind of medical professional after our first son was born. This person said things to us that we found profoundly disturbing. After deliberating about it, my wife decided to file a complaint with the board which certifies people who work in this profession. It wasn't like badmouthing somebody on the Internet or in the newspaper, it was this formal process. Now obviously, when you file a complaint you have to describe what happened, which my wife did.
The board apparently studied our complaint and took some kind of action against this person (something mild, I don't remember exactly, but we got something in the mail about it). A couple of months later my wife was served with the lawsuit.
At the time, we had very little capacity to pay a high-priced lawyer. The person was asking for $150,000! So on a long shot we called our homeowner's insurance company and it turned out they would defend the case under our insurance policy. I have no idea if the attorney sucked, didn't care, or if something else went wrong but the jury found in favor of plaintiff for $5k. Our insurance company paid, but it was a disgusting experience.
I found out later that this woman has a history of serial lawsuits. Not long after our case ended, she sued the board which had sanctioned her. I know this, because she fucking subpoenaed me as a witness in that case. Did I mention she was representing herself? She asked me a bunch of nutty questions for two hours which I had no idea about -- like asking me to tell her what I thought some email meant, which I had never seen before (it had nothing to do with me or my wife).
I don't know what happened after that. What a terrible shitty thing it was. Makes you scared to speak up when somebody is abusive to you. We got slimed.
False, my wife was sued for libel and lost, despite her statements being completely true. In fact, the court seemed totally uninterested in whether it was true. It was a jury trial, if that matters.
Actually, it isn't far-fetched to assume that lawmakers will do something idiotic that causes a bunch of consequences they didn't intend. While I can easily see Facebook trying to language-lawyer this shit to their advantage, I'd give it 50/50 chance the law actually does imply the goofy stuff Facebook says it does.
I believe that laws should always be enforced in full and to the letter, along with all unintended consequences. This way, broken laws can be quickly identified and fixed (or repealed). It also would prevent prosecutors from selectively enforcing obscure provisions of the law to target specific individuals.
When judges and juries start making exceptions for cases that are "obviously not what was meant" we just encourage more sloppy law-making.
If it was within our power to resurrect the dead, there would be no need for laws against murder. Killing somebody would just be a way of inconveniencing them temporarily -- murder would be a type of harassment.
Belcher was no victim. He murdered his girlfriend in cold blood and then took the easy way out by killing himself before he could be brought to justice. I suppose his act could be described as rational (having just destroyed his entire life, what reason was there to go on?) But it wasn't brave, or honorable.
It's possible that repeated hits to the head fucked him up, but we don't allow "I was fucked up your honor" as an excuse for other people, such as drunk drivers.
ALL context switches are expensive. The primary effect of a context switch is that each context has its own memory address layout.
No, that's not correct. Context switches between threads within the same process (or between one kernel thread and another), or context switches due to system calls, do not alter the page tables and do not flush the TLB. The vast majority of context switches are due to system calls, not scheduling. In a system call, the overhead is primarily due to switching in and out of supervisor mode.
This creates a LOT of extra work for the CPU as it cannot rely on cached data (addresses are different, the same data may not be in the same location in a new context) and consequent cache invalidation, etc.
Again, incorrect. CPU cache (not TLB) is tagged by physical, not virtual, address. Changes to the page tables are irrelevant to the cache.
People are mentioning tcpdump, wireshark, etc. Why not sniff something a bit more lower level, a bit less documented, and therefore a bit more interesting?
Buy a cheap logic analyzer (here's one for $50). For even more fun buy a Bus Pirate, which works kind of like the old Game Genie game modification device from the 90's. Connect probes to conductors on various devices and try to figure out how they communicate at the electrical level, then modify the signals themselves to try to make new things happen!
How are you going to throw a big enough rock? Consider Meteor Crater in Arizona. This is a crater with diameter less than a mile -- big enough to obliterate most of the core of a city, but not region-wide destruction kind of size. The meteor required to do that was 50 meters across and made of iron, making its mass somewhere around 500 million kilograms. The escape velocity from the surface of the moon is 2.4 kilometers per second. The energy required to accelerate 500 million kilograms to 2.4 km/s is 400 gigawatt hours.
Suppose you wanted to be able to launch the rock within 30 seconds of a Russian sneak attack. That means a net power of 48 terawatts. Oh, and multiply that by the number of rocks you plan on throwing. How do you propose we produce 48 terawatts up on the Moon?
You do realize that the use of force is not the only means of human cooperation, right?
Actually, it is. Because those who are willing to be violent will obliterate those who are unwilling. That has nothing to do with human morals and everything to do with the laws of physics ("bullet through the brain causes a splattering of gray matter everywhere").
Until you find a way for the "cooperators" to control the "violence users" in a way that doesn't involve violence, you're just wrong.
Ah, a proponent of the "one true way" style of thinking, where the one true way was defined several decades ago by people who are clearly smarter than the rest of us "clueless" users. Who the hell are we to defy the wisdom of a bunch of bearded UNIX hackers? How convincing.
Slashdot Mirror
No. There is no "safe" place to shoot somebody. People have died from being shot in the foot. The purpose of firing a gun at a human being is to stop a threat. That may or may not involve killing the person. Whether they die is irrelevant, whether they have been stopped is the whole point. To that end, the correct target to aim at is the center of the chest.
A whole lot can go wrong when you fire a gun. Unless you 1) feel that a person absolutely must be stopped, 2) you are willing to potentially kill that person in order to stop them, and 3) you feel that the risk of injuring somebody who is not the target is justified by the threat, then you should not be pulling the trigger. "Shoot to wound" is something from the movies.
If all firearms disappeared over night, murders would then occur using some other kind of weapon.
If you want to argue that high-fire-rate weapons can kill a lot of people quickly, I invite you to imagine a person wielding a sword in a crowded shopping mall.
Lets have some sanity about guns as a privileged and not a right
So, treason is cool now? Passing laws in direct violation of Constitutionally enumerated rights is fine?
While we're at it, let's make a list of other Constitutional amendments we prefer to ignore. How about those pesky 13th, 19th, and 24th as well?
Anybody who thinks a knife is less deadly than a gun has obviously never tried to defend themselves against a knife attack. You have to get closer to the victim, true... But that's not of much consequence, as the victim is probably unarmed, and in this case they were little kids.
Are JPEG images necessarly tied to pixels? If they store images as just a bunch of cosine waves added together, can't they in theory be resolution independent?
Yes, in theory that is possible. But rendering the JPEG directly at a larger size isn't going to look any better than rendering it at the usual size and then upsampling it. In fact, it would probably look worse, because if you upsample, you at least get some interpolation that might help smooth the JPEG block artifacts -- if you render the image directly to a larger block size it'll probably make the block discontinuities look even worse (more visible).
You want to implement memory protection on a CPU that has no MMU. You could implement this in software by putting the CPU into perpetual single-step mode, and intercepting all memory accesses to implement the MMU in software. But this will be many thousands of times slower than a hardware MMU.
No, I don't believe everything I read, but the story is that his neighbor was complaining about his dogs, then his dogs were shot dead by somebody and no more than a day later the guy who complained about the dogs turned up dead. If that is what happened, don't you think it's deserving of some investigation at least?
Actually, a detective who had been investigating the OJ case for 30 or so years [...]
The murders occurred in 1994. You're a whole decade off.
Careful, dude. There are a lot of things that "can be implemented in software" but are much better achieved in hardware. I see no reason to prevent patents on hardware simply because the problem being solved could also be solved by some (possibly very hokey, inefficient) software.
Or, maybe neither of the phenomena causes the other. Maybe the eggs which hatch last are laid last, and eggs which are laid last are not as thick as eggs laid first, due to depletion of calcium in the mother's blood from laying previous eggs, thereby affecting the rate of gas exchange during incubation, which modulates brain development, yadda yadda...
If possible, design a component such that no errors will ever occur (except for hardware failure). An entire program can't be designed that way, obviously, but individual components can be. Collecting all resource allocation into a small, well defined set of locations will relieve the majority of other code from the need to handle errors. In such a design, the majority of functions will return "void".
Only my wife was a defendant. And it's not like being there would have helped. When the insurance company provides an attorney for you, the attorney is really working for the insurance company, not you. They are going to do whatever the insurance company thinks is best. I tried several times to provide information and influence the lawyer's decision making and was politely ignored.
It sucks, but we would have been more money out of pocket had we hired our own lawyer. As it happened we didn't pay a single dollar for any of this crap.
How on earth was she supposed to prove it? Let me give the context for this. My wife and I met with a certain kind of medical professional after our first son was born. This person said things to us that we found profoundly disturbing. After deliberating about it, my wife decided to file a complaint with the board which certifies people who work in this profession. It wasn't like badmouthing somebody on the Internet or in the newspaper, it was this formal process. Now obviously, when you file a complaint you have to describe what happened, which my wife did.
The board apparently studied our complaint and took some kind of action against this person (something mild, I don't remember exactly, but we got something in the mail about it). A couple of months later my wife was served with the lawsuit.
At the time, we had very little capacity to pay a high-priced lawyer. The person was asking for $150,000! So on a long shot we called our homeowner's insurance company and it turned out they would defend the case under our insurance policy. I have no idea if the attorney sucked, didn't care, or if something else went wrong but the jury found in favor of plaintiff for $5k. Our insurance company paid, but it was a disgusting experience.
I found out later that this woman has a history of serial lawsuits. Not long after our case ended, she sued the board which had sanctioned her. I know this, because she fucking subpoenaed me as a witness in that case. Did I mention she was representing herself? She asked me a bunch of nutty questions for two hours which I had no idea about -- like asking me to tell her what I thought some email meant, which I had never seen before (it had nothing to do with me or my wife).
I don't know what happened after that. What a terrible shitty thing it was. Makes you scared to speak up when somebody is abusive to you. We got slimed.
When laws conflict, both laws should be nullified. This would have the nice side effect of reducing the number of laws.
False, my wife was sued for libel and lost, despite her statements being completely true. In fact, the court seemed totally uninterested in whether it was true. It was a jury trial, if that matters.
Actually, it isn't far-fetched to assume that lawmakers will do something idiotic that causes a bunch of consequences they didn't intend. While I can easily see Facebook trying to language-lawyer this shit to their advantage, I'd give it 50/50 chance the law actually does imply the goofy stuff Facebook says it does.
I believe that laws should always be enforced in full and to the letter, along with all unintended consequences. This way, broken laws can be quickly identified and fixed (or repealed). It also would prevent prosecutors from selectively enforcing obscure provisions of the law to target specific individuals.
When judges and juries start making exceptions for cases that are "obviously not what was meant" we just encourage more sloppy law-making.
If it was within our power to resurrect the dead, there would be no need for laws against murder. Killing somebody would just be a way of inconveniencing them temporarily -- murder would be a type of harassment.
Strange -- do you also expect murder trials to bring the victims back to life?
Belcher was no victim. He murdered his girlfriend in cold blood and then took the easy way out by killing himself before he could be brought to justice. I suppose his act could be described as rational (having just destroyed his entire life, what reason was there to go on?) But it wasn't brave, or honorable.
It's possible that repeated hits to the head fucked him up, but we don't allow "I was fucked up your honor" as an excuse for other people, such as drunk drivers.
ALL context switches are expensive. The primary effect of a context switch is that each context has its own memory address layout.
No, that's not correct. Context switches between threads within the same process (or between one kernel thread and another), or context switches due to system calls, do not alter the page tables and do not flush the TLB. The vast majority of context switches are due to system calls, not scheduling. In a system call, the overhead is primarily due to switching in and out of supervisor mode.
This creates a LOT of extra work for the CPU as it cannot rely on cached data (addresses are different, the same data may not be in the same location in a new context) and consequent cache invalidation, etc.
Again, incorrect. CPU cache (not TLB) is tagged by physical, not virtual, address. Changes to the page tables are irrelevant to the cache.
People are mentioning tcpdump, wireshark, etc. Why not sniff something a bit more lower level, a bit less documented, and therefore a bit more interesting?
Buy a cheap logic analyzer (here's one for $50). For even more fun buy a Bus Pirate, which works kind of like the old Game Genie game modification device from the 90's. Connect probes to conductors on various devices and try to figure out how they communicate at the electrical level, then modify the signals themselves to try to make new things happen!
"Cape Carnival?"
How are you going to throw a big enough rock? Consider Meteor Crater in Arizona. This is a crater with diameter less than a mile -- big enough to obliterate most of the core of a city, but not region-wide destruction kind of size. The meteor required to do that was 50 meters across and made of iron, making its mass somewhere around 500 million kilograms. The escape velocity from the surface of the moon is 2.4 kilometers per second. The energy required to accelerate 500 million kilograms to 2.4 km/s is 400 gigawatt hours.
Suppose you wanted to be able to launch the rock within 30 seconds of a Russian sneak attack. That means a net power of 48 terawatts. Oh, and multiply that by the number of rocks you plan on throwing. How do you propose we produce 48 terawatts up on the Moon?
You do realize that the use of force is not the only means of human cooperation, right?
Actually, it is. Because those who are willing to be violent will obliterate those who are unwilling. That has nothing to do with human morals and everything to do with the laws of physics ("bullet through the brain causes a splattering of gray matter everywhere").
Until you find a way for the "cooperators" to control the "violence users" in a way that doesn't involve violence, you're just wrong.
Ah, a proponent of the "one true way" style of thinking, where the one true way was defined several decades ago by people who are clearly smarter than the rest of us "clueless" users. Who the hell are we to defy the wisdom of a bunch of bearded UNIX hackers? How convincing.