No, in that case, you're defending someone else's wanton disregard for the rights of others or for the value of human life. Which is still disgusting. The law says it's murder; the dictionary agrees that it's malicious. Revenge has nothing to do with it. Nobody who displays such a profound disregard for other people should be permitted to be part of society.
Main Entry: malice
Pronunciation: 'ma-l&s
Function: noun
1 a : the intention or desire to cause harm (as death, bodily injury, or property damage) to another through an unlawful or wrongful act without justification or excuse b : wanton disregard for the rights of others or for the value of human life c : an improper or evil motive or purpose
Sounds to me like killing someone by driving while drunk qualifies.
People seem to suspend so much understanding of basic science when pondering spaceflight...all of our fictional models presume some effortless way to move very far distances and of course no adverse health affects on the human body.
It sounds like your idea of predictions of future technology is based on Star Trek. If you're going to judge possibilities for space flight based on science fiction, at least read some of the good stuff. Pretty much every technical objection that can be made against manned space flight has been tackled by authors at some point -- and their solutions are sometimes very plausible.
If you think that it's as hard to check code for correctness as it is to write the code in the first place, you can't be a developer, or you're not thinking clearly.
Have you ever developed any software on a large scale? Checking your own code for correctness -- bug identification, testing, fixing -- is always the longest part of development. Checking somebody else's software is even worse. Something that has been intentionally inserted and hidden could slip by even intense scrutiny.
If it wasn't for the huge amount of piracy going on, they wouldn't have it as an excuse. If you really hate the RIAA so much, you should not only vote with your wallet by not giving them money, you should also refrain from obtaining their products illegally.
As long as people continue to use their products, the record industry will have control of the situation.
Seriously, even when we reach this point, there will still be major obstacles to actually implementing such automated control. One is refusal (emotional, but no less real for being so) to accept a machine's decisions; as another poster pointed out, "Sorry, ma'am, the computer decided your kid was expendable" won't be well received.
I don't disagree with this. However, that's more of an argument for why it won't be used, rather than why it shouldn't -- which was where I was coming from.
Because I can understand, subconsciously, what a human will do, how a human will react,
The way I see it, humans aren't very good at this anyway. If we were, then there wouldn't be nearly as many accidents as there are. Yes, a computer may make a mistake -- but I think that it's obvious that they could do a better job in many situations than a human. It seems very likely that the number of accidents that could be prevented by quick reaction time, improved awareness, and high-quality physics predictions would be greater than the number of accidents averted by human intuition.
Except that the right response is NOT to brake every time. Example: sometime last year, a deer jumped out on I-70 and ran into my lane, the middle lane. I managed to jump into the lane he had vacated and avoided collision. Braking would have been disastrous or fatal.
As I indicated in my first post, an obstacle leaves the driver with many options: swerving, braking, driving over it, speeding up. It turns out that computers are really bad, so far, at making that decision.
What, are you saying that computers aren't capable of:
And, of course, don't forget that at 75mph, the computer will be able to stop at least 100ft sooner than a human. That's a lot of space. Sure, collision avoidance is a tough problem -- but it's not unsolvable.
Anyone can pull out scenarios where a computer might fail to make the right decision; and anyone can come up with scenarios where a computer would succeed in safely avoiding an accident that a human would not. I contend that at some point in the future, the benefits of having a computer in control will far outweigh the risks, and at that point we would be wise to use them instead of foolishly trusting our human superiority.
First off, the situation you mention didn't require you to hit anything. So even if the computer just knew to avoid everything, it could have solved that particular problem. (And it's doubtful whether your course of action was the only possible one to avoid a collision -- it may have been the only one you found in a split second, but computers are very good at considering lots of possibilities, and cars are more maneuverable than most people think.) Again, situations where a collision is inevitable are extremely rare, compared to situations where quick reactions can avoid a collision.
If we give control of the car to the computer, it must do better than the human in all situations. "Sorry, Ma'am. The cars AI chose to hit your child instead of the dog" is not acceptable.
I honestly don't understand this reasoning. Why is it acceptable to have human-caused deaths, but not AI-caused ones? (And don't try to say that neither is, because your statement says that even if an AI only fails in one uncommon situation and the human fails in dozens of common situations, then you still can't use the AI.) If computer-controlled cars were safer [which I'd define as causing fewer accidents or fatalities], it would seem like a no-brainer to use them.
Can we make the computer fast enough and smart enough to choose correctly, every time?
I suspect that we can make a computer that will do a better job of it than humans. If you think that avoiding the kid is hardwired into people's brains, you obviously haven't driven much lately. Chances are that the kid would get killed at least one time out of three.
Now, instead of your rather contrived example -- situations where you have to hit something are extremely rare if appropriate safety precautions are taken -- let's take the situation where a kid runs out into the street in front of you, and you need to brake immediately to avoid hitting him. A computer-controlled car should be able to sense an obstacle on the road and brake every time. Do you trust humans to do the same? (If your answer is yes, please pick up a newspaper and take a look for pedestrian traffic fatalities.)
It's obviously true that computer control can be safer than human control in some situations. It's also true that with current technology, a human controller will be safer in some situations than a computer. With the current state of tech, computers probably aren't up to the task. Someday, however, that's going to change.
Wrong. Unless you're not a normal human, your eyes only give you a 24 Hz refresh rate.
This is false. Your eyes don't see things as "frames," so talking about the "refresh rate" of your eyes is meaningless. 24 frames per second is enough to give the illusion of motion, which is why you can have movies at that speed; however, your eyes can percieve quite a bit more than that. (On the same note, the 60 Hz claim of in the original post was just as nonsensical.)
However, the OP is dead wrong about computers not being fast enough. A well-designed sensor system can gather data much faster and more accurately than any human. It's figuring out what to do based on that data that is the hard part -- and it seems likely that in a controlled environment, computers could do a better job of that than humans.
It seems to me that appearance is probably one of the last reasons why most people would have the surgery -- after all, most (though admittedly not all) people who are candidates for the surgery could also wear contacts, and get the benefits of "no glasses".
Although I haven't done it yet, I'd like to do so in the future; the biggest reason being convenience. No more glasses to break or contacts to lose; no more groping for my glasses in the middle of the night; no more spending the time to put in and take out my contacts every day. Everyone who I've talked to who has had it done had the same reasons that I do. I have, however, been a bit worried about the failure rate, which is why I haven't yet done it.
Spoken like someone who has already forgotten what happened in 2000. Tens of thousands of Nader followers decided to get clever and vote for a candidate they disagreed with because they disliked Bush more. The result: Nader had a poorer showing than he might have had, and Bush still won.
Besides which, it's patently false that a libertarian vote is a "vote for Bush." There are quite a few people that would probably vote republican if they couldn't vote libertarian.
While the compression phase only has to run once, the decompression phase is what worries me. According to the tests, the decompression phase for bzip2 runs 5-8 times that of gzip.
Well, yes -- but the decompression time is still miniscule compared to the transmission time. If your compression is 7% more effective, then you decrease the transmission time by 7%. From their statistics, it looks like that's an overall win.
I did, however, forget that these standings also rely on the speed of the system used; if you're using a slower computer, the advantage of quicker decompression becomes more important. It looks to me like, on a typical broadband internet connection, bzip2 is probably better on a modern system, while gzip might win on sub-GHz machines.
Quite interesting on that page is the "ranking" section that lets you choose a network speed, and ranks the different algorithms according to time that it takes to compress/transmit/decompress a file (which is a typical usage of compression utilities, and probably one of the best measures of how effective it would be for web graphics.)
However, it doesn't take into account the fact that, for typical web distribution, the compression phase only has to be run once, and should probably not be counted. In that case, it looks like bzip2 slightly edges out gzip. Also, the default page is set up for a 1000kbit connection, which is probably faster than most internet connections -- which favors gzip. And, of course, it doesn't give benchmarks for image data, which might be different than results for source code.
Slashdot Mirror
As long as people continue to use their products, the record industry will have control of the situation.
Anyone can pull out scenarios where a computer might fail to make the right decision; and anyone can come up with scenarios where a computer would succeed in safely avoiding an accident that a human would not. I contend that at some point in the future, the benefits of having a computer in control will far outweigh the risks, and at that point we would be wise to use them instead of foolishly trusting our human superiority.
Lots of us. Unless X.org offers something new and improved, I can't see any reason why I would want to switch.
Now, instead of your rather contrived example -- situations where you have to hit something are extremely rare if appropriate safety precautions are taken -- let's take the situation where a kid runs out into the street in front of you, and you need to brake immediately to avoid hitting him. A computer-controlled car should be able to sense an obstacle on the road and brake every time. Do you trust humans to do the same? (If your answer is yes, please pick up a newspaper and take a look for pedestrian traffic fatalities.)
It's obviously true that computer control can be safer than human control in some situations. It's also true that with current technology, a human controller will be safer in some situations than a computer. With the current state of tech, computers probably aren't up to the task. Someday, however, that's going to change.
However, the OP is dead wrong about computers not being fast enough. A well-designed sensor system can gather data much faster and more accurately than any human. It's figuring out what to do based on that data that is the hard part -- and it seems likely that in a controlled environment, computers could do a better job of that than humans.
That site seems to be down. Could somebody post a mirror?
Although I haven't done it yet, I'd like to do so in the future; the biggest reason being convenience. No more glasses to break or contacts to lose; no more groping for my glasses in the middle of the night; no more spending the time to put in and take out my contacts every day. Everyone who I've talked to who has had it done had the same reasons that I do. I have, however, been a bit worried about the failure rate, which is why I haven't yet done it.
Besides which, it's patently false that a libertarian vote is a "vote for Bush." There are quite a few people that would probably vote republican if they couldn't vote libertarian.
I did, however, forget that these standings also rely on the speed of the system used; if you're using a slower computer, the advantage of quicker decompression becomes more important. It looks to me like, on a typical broadband internet connection, bzip2 is probably better on a modern system, while gzip might win on sub-GHz machines.
However, it doesn't take into account the fact that, for typical web distribution, the compression phase only has to be run once, and should probably not be counted. In that case, it looks like bzip2 slightly edges out gzip. Also, the default page is set up for a 1000kbit connection, which is probably faster than most internet connections -- which favors gzip. And, of course, it doesn't give benchmarks for image data, which might be different than results for source code.