Apparently, enough people find these interesting and vote them up every time so they keep appearing on the front page of Slashdot. I know, democracy is a bitch. Personally it doesn't bother me too much.
The problem is probably that funding research is a different budget than funding launches. So they will prefer to postpone the research and pay more for the Russians to do it, rather than save money in the long run by letting SpaceX finish their program and then do the launches much cheaper. That's the way bureaucracy works.
That may be a good idea for valves that open and close regularly. But for valves that only open in an emergency, it's probably better to just put them in parallel. 2s2p is more likely to fail closed than 2p. Anyway, I'm sure they thought of that.
Thank you for your exhaustive reply, that was very informative indeed. I wish they would just build lots of those reactors and get rid of the old ones, rather than extending the old ones because "nucular is dangerous so we don't want to build any more".
Then what happened in Fukushima? That meltdown was't so benign, was it? Overheating, production of hydrogen, explosions, etc... When all that could have been avoided if the whole thing had been designed with materials that melt at much lower temperatures so the material is dispersed when the reactor starts to overheat.
If the valves have to open in an emergency, and a single valve has a probability of failure of 1/10, three parallel valves bring the probability down to 1/1000.
If the valves have to close in an emergency, put them in series and the result is the same.
They put a lot of effort into preventing a meltdown, and that's obviously a good thing, but why don't they do anything to limit the effects of a meltdown if it ever does happen? That would make all the other safety mechanisms a lot less critical. AFAIK, if a meltdown does occur in modern reactors, all the fuel just becomes a big hot lump of critical mass at the bottom of the reactor that's very hard to deal with and causes all sorts of trouble due to the enormous temperatures involved.. Why not put a wide cone of very heat resistant metal at the bottom, so the fuel falls onto the cone and spreads out in a very wide circle along the bottom, for example? Or install the entire reactor vessel on top of a huge pool of water so that, if the vessel melts down, everything just drops into the huge pool. I can think of so many low tech ways of dispersing fuel automatically whenever the container becomes too hot, using materials that perform well under normal operating conditions but melt when the temperature exceeds a threshold, automatically creating a safe and stable situation. Sure, the reactor becomes a write-off, but there will be no explosions, no release of radioactivity, etc.
What's the price of one of those valves? And how much is it compared to the construction of the entire plant? And compared to the amount they are already over budget? Can't they just install three times as many valves as required, rather than doing more studies that cost ten times as much? I'm not an expert, but they seem to be penny wise pound stupid. Make everything to the exact required specifications and then go over budget when things don't quite work as planned, instead of taking a huge margin and adding some extra safety redundancy well beyond the regulations. How hard can it be to keep a reactor from overheating? All you need to do is take the fuel apart so it's no longer critical, and pump cool water through the vessel. I can think of a dozen ways to do that, all of them together costing less than what they are already over budget. But nooooo, we must save costs, so we have to install the minimum amount of safety equipment we can get away with, and then end up spending way more after all when some of the assumptions turn out to be flawed.
Well, the Dungeness power plant is just about as close to France as the British could get. And France and England are not the only ones. The one in Belgium (Doel) is pretty close to the Netherlands. If you look at a map of nuclear power plants, there seem to be a lot more near borders than would be statistically "expected". I wonder why...
And when will you make a 17 inch laptop again? Mine's still going strong after 5 years after replacing the HDD with an SSD, but I would have bought an upgrade two years ago if you only made one.
You'll find that most people slow when they see an impending collision - reducing the speed tends to be the best option.
That's a fallacy that has caused many accidents already. I took an advanced driving course a few years ago, and they showed us some videos of the stupid things people do, just to show that no matter how good you are, you always have to assume other drivers are likely to be idiots.
In one of those videos, a car started to cross an intersection, but then the driver got scared because the approaching traffic was closer/faster than she thought, so, instead of accelerating out of the way, she actually hit the brakes (reducing the speed is the best option, right?) and stopped in the middle of the intersection. Of course she got hit by the approaching traffic that had slowed down a bit, but never expected her to come to a complete standstill. And yes, it was not a malfunction in the car, she really did hit the brakes. Quite unbelievable, but true.
So you are right that many people tend to do this. But you should not consider that to be a good thing. The Google car might have been better off not braking but actually accelerating under those circumstances. But once again, I don't have the exact data. The three examples just seemed fishy and kind of familiar to situations I have encountered with bad (but officially very safe) drivers.
Yes, in all of them the other driver was legally at fault. That's not the point. If you drive very slowly on the highway for no good reason, you are creating a dangerous situation. If you brake while the road is clear just because something might happen on the side of the road, you are creating a dangerous situation. People just don't expect you to do things like that.
Yes, and if you read the reports, I noticed a few where technically the autonomous car may not have been "at fault" (legally) but the car did something that normal drivers would not expect and this resulted in the accident.
- December 2012: rear-ended at approx. 20-25 mph while driving past a disabled vehicle and emergency vehicles, which were stationary on the shoulder. In other words, the road was clear, but there were some emergency vehicles on the shoulder, so the car needlessly slowed down to a ridiculously low speed and got rear-ended by a human driver who did not expect this.
- Feburary 2015: a car rolled through a stop sign and hit the rear quarter panel of the Google car. Prior to the collision, the Google car had applied the brakes in response to the detection of the other vehicle. I don't have the exact data, but could it be that the human rolling through the stop sign had judged that he would be able to pass behind the Google car, but hit it because the Google car braked unexpectedly?
- April 2015: Google car starts rolling a bit to get a better view before making right turn on red, then stops again for approaching traffic. Reminds me of an accident I almost had one day behind a human driver who started to move forward and then stopped again for traffic that was still way out, while the two of us could have easily passed. I was looking to the left, judging that I could easily follow the car ahead with plenty of time for the approaching traffic, and only noticed in the nick of time that the idiot in front had actually stopped. His overly cautious behaviour almost caused a crash in which I officially would have been to blame.
So basically, it looks like the Google cars are driving like typical inexperienced, nervous and scared drivers who brake for all sorts of things and cause others to "cause" accidents. Which is only be be expected, really. It will be some time before they really become as fluent as human drivers.
I'm sure glad Einstein devoted more than 1% of his time to prove that time flows differently for different observers, speeds cannot be simply added together, momentum has to be calculated differently, etc.
If a researcher perfoms an experiment and gets a very strange, unexpected result, what should he do? Say "that result is clearly impossible, so I shall just disregard it"?
No, he will try to repeat the experiment, gather data, and try to figure out what's going on. Maybe (most likely) there's a perfectly valid explanation within existing scientific frameworks, maybe it's a setup or measurement error, or maybe, just maybe, this is a new effect that hadn't been discovered yet. So the scientist tries to figure that out, and tells others about the experiments so they can try the same thing and see if they get similar results.
That's how science works.
I'm sure you would have called the theory of relativity "pseudoscience" back in the day of Newtonian physics. New things do get discovered sometimes. As long as it's being researched using scientific methods, that's science and not pseudoscience. Yes, they probably will be wrong. That doesn't mean it's not science.
Why don't the GIMP maintainers just accept to take back control, then "update" the project by removing all the source code files and replacing them with funny cat pictures?
The big problem is that everyone will now remember the fake study and continue to believe it, because the rectification doesn't get nearly as much coverage. People are still refusing to vaccinate children because they're afraid of autism even though the author of that study actually confessed having made the whole thing up.
I like the way they start out by saying how hard it would be to land on Mars. You mean nobody has thought of that yet? Quick, somebody call Elon Musk and warn him before he sends over a rocket full of people with no way of landing there!
Also, they actually say they might have problems with plants producing too much oxygen. OK, hang on a minute there... Too much oxygen? On Mars? Somehow I don't think that will be such a major problem. Especially when combined with that other problem of not being able to make enough CO2...
I'm not saying it's going to be a picknick. It will be a hell of a challenge to just grow food and get breathable air. It just seems funny how the article emphasizes non-issues while disregarding much bigger problems.
Oh, and we shouldn't send over women because women live longer and are therefore more likely to develop cancer! Right, pick people with the shortest possible lifespan to maximize their... errr... oh, wait...
Slashdot Mirror
If it has a half life longer than humanity's been in existence, it can't be very radioactive.
And what does a few grams of plutonium have to do with nuclear power plants blowing up?
Apparently, enough people find these interesting and vote them up every time so they keep appearing on the front page of Slashdot. I know, democracy is a bitch. Personally it doesn't bother me too much.
The problem is probably that funding research is a different budget than funding launches. So they will prefer to postpone the research and pay more for the Russians to do it, rather than save money in the long run by letting SpaceX finish their program and then do the launches much cheaper. That's the way bureaucracy works.
SpaceX is more likely to sell them a trampoline.
Especially after the Russians refused to sell their rockets to Elon when he went there many years ago.
But then again, money is money, and they need funding for manned missions to develop the Mars project, so who knows.
That may be a good idea for valves that open and close regularly. But for valves that only open in an emergency, it's probably better to just put them in parallel. 2s2p is more likely to fail closed than 2p. Anyway, I'm sure they thought of that.
Thank you for your exhaustive reply, that was very informative indeed. I wish they would just build lots of those reactors and get rid of the old ones, rather than extending the old ones because "nucular is dangerous so we don't want to build any more".
Then what happened in Fukushima? That meltdown was't so benign, was it? Overheating, production of hydrogen, explosions, etc... When all that could have been avoided if the whole thing had been designed with materials that melt at much lower temperatures so the material is dispersed when the reactor starts to overheat.
If the valves have to open in an emergency, and a single valve has a probability of failure of 1/10, three parallel valves bring the probability down to 1/1000.
If the valves have to close in an emergency, put them in series and the result is the same.
They put a lot of effort into preventing a meltdown, and that's obviously a good thing, but why don't they do anything to limit the effects of a meltdown if it ever does happen? That would make all the other safety mechanisms a lot less critical. AFAIK, if a meltdown does occur in modern reactors, all the fuel just becomes a big hot lump of critical mass at the bottom of the reactor that's very hard to deal with and causes all sorts of trouble due to the enormous temperatures involved.. Why not put a wide cone of very heat resistant metal at the bottom, so the fuel falls onto the cone and spreads out in a very wide circle along the bottom, for example? Or install the entire reactor vessel on top of a huge pool of water so that, if the vessel melts down, everything just drops into the huge pool. I can think of so many low tech ways of dispersing fuel automatically whenever the container becomes too hot, using materials that perform well under normal operating conditions but melt when the temperature exceeds a threshold, automatically creating a safe and stable situation. Sure, the reactor becomes a write-off, but there will be no explosions, no release of radioactivity, etc.
What's the price of one of those valves? And how much is it compared to the construction of the entire plant? And compared to the amount they are already over budget? Can't they just install three times as many valves as required, rather than doing more studies that cost ten times as much? I'm not an expert, but they seem to be penny wise pound stupid. Make everything to the exact required specifications and then go over budget when things don't quite work as planned, instead of taking a huge margin and adding some extra safety redundancy well beyond the regulations. How hard can it be to keep a reactor from overheating? All you need to do is take the fuel apart so it's no longer critical, and pump cool water through the vessel. I can think of a dozen ways to do that, all of them together costing less than what they are already over budget. But nooooo, we must save costs, so we have to install the minimum amount of safety equipment we can get away with, and then end up spending way more after all when some of the assumptions turn out to be flawed.
Well, the Dungeness power plant is just about as close to France as the British could get. And France and England are not the only ones. The one in Belgium (Doel) is pretty close to the Netherlands. If you look at a map of nuclear power plants, there seem to be a lot more near borders than would be statistically "expected". I wonder why...
And when will you make a 17 inch laptop again? Mine's still going strong after 5 years after replacing the HDD with an SSD, but I would have bought an upgrade two years ago if you only made one.
You'll find that most people slow when they see an impending collision - reducing the speed tends to be the best option.
That's a fallacy that has caused many accidents already. I took an advanced driving course a few years ago, and they showed us some videos of the stupid things people do, just to show that no matter how good you are, you always have to assume other drivers are likely to be idiots.
In one of those videos, a car started to cross an intersection, but then the driver got scared because the approaching traffic was closer/faster than she thought, so, instead of accelerating out of the way, she actually hit the brakes (reducing the speed is the best option, right?) and stopped in the middle of the intersection. Of course she got hit by the approaching traffic that had slowed down a bit, but never expected her to come to a complete standstill. And yes, it was not a malfunction in the car, she really did hit the brakes. Quite unbelievable, but true.
So you are right that many people tend to do this. But you should not consider that to be a good thing. The Google car might have been better off not braking but actually accelerating under those circumstances. But once again, I don't have the exact data. The three examples just seemed fishy and kind of familiar to situations I have encountered with bad (but officially very safe) drivers.
Yes, in all of them the other driver was legally at fault. That's not the point. If you drive very slowly on the highway for no good reason, you are creating a dangerous situation. If you brake while the road is clear just because something might happen on the side of the road, you are creating a dangerous situation. People just don't expect you to do things like that.
Yes, and if you read the reports, I noticed a few where technically the autonomous car may not have been "at fault" (legally) but the car did something that normal drivers would not expect and this resulted in the accident.
- December 2012: rear-ended at approx. 20-25 mph while driving past a disabled vehicle and emergency vehicles, which were stationary on the shoulder. In other words, the road was clear, but there were some emergency vehicles on the shoulder, so the car needlessly slowed down to a ridiculously low speed and got rear-ended by a human driver who did not expect this.
- Feburary 2015: a car rolled through a stop sign and hit the rear quarter panel of the Google car. Prior to the collision, the Google car had applied the brakes in response to the detection of the other vehicle. I don't have the exact data, but could it be that the human rolling through the stop sign had judged that he would be able to pass behind the Google car, but hit it because the Google car braked unexpectedly?
- April 2015: Google car starts rolling a bit to get a better view before making right turn on red, then stops again for approaching traffic. Reminds me of an accident I almost had one day behind a human driver who started to move forward and then stopped again for traffic that was still way out, while the two of us could have easily passed. I was looking to the left, judging that I could easily follow the car ahead with plenty of time for the approaching traffic, and only noticed in the nick of time that the idiot in front had actually stopped. His overly cautious behaviour almost caused a crash in which I officially would have been to blame.
So basically, it looks like the Google cars are driving like typical inexperienced, nervous and scared drivers who brake for all sorts of things and cause others to "cause" accidents. Which is only be be expected, really. It will be some time before they really become as fluent as human drivers.
Anyway, who cares about sunset when you're that far away, the sun is just a star in the sky that's a bit brighter than the others
Dwarf moons?
Also, good luck finding the back doors if they were written by contestants in the underhanded code contest.
I'm sure glad Einstein devoted more than 1% of his time to prove that time flows differently for different observers, speeds cannot be simply added together, momentum has to be calculated differently, etc.
No, it's not pseudoscience.
If a researcher perfoms an experiment and gets a very strange, unexpected result, what should he do? Say "that result is clearly impossible, so I shall just disregard it"?
No, he will try to repeat the experiment, gather data, and try to figure out what's going on. Maybe (most likely) there's a perfectly valid explanation within existing scientific frameworks, maybe it's a setup or measurement error, or maybe, just maybe, this is a new effect that hadn't been discovered yet. So the scientist tries to figure that out, and tells others about the experiments so they can try the same thing and see if they get similar results.
That's how science works.
I'm sure you would have called the theory of relativity "pseudoscience" back in the day of Newtonian physics. New things do get discovered sometimes. As long as it's being researched using scientific methods, that's science and not pseudoscience. Yes, they probably will be wrong. That doesn't mean it's not science.
Why don't the GIMP maintainers just accept to take back control, then "update" the project by removing all the source code files and replacing them with funny cat pictures?
The big problem is that everyone will now remember the fake study and continue to believe it, because the rectification doesn't get nearly as much coverage. People are still refusing to vaccinate children because they're afraid of autism even though the author of that study actually confessed having made the whole thing up.
I like the way they start out by saying how hard it would be to land on Mars. You mean nobody has thought of that yet? Quick, somebody call Elon Musk and warn him before he sends over a rocket full of people with no way of landing there!
Also, they actually say they might have problems with plants producing too much oxygen. OK, hang on a minute there... Too much oxygen? On Mars? Somehow I don't think that will be such a major problem. Especially when combined with that other problem of not being able to make enough CO2...
I'm not saying it's going to be a picknick. It will be a hell of a challenge to just grow food and get breathable air. It just seems funny how the article emphasizes non-issues while disregarding much bigger problems.
Oh, and we shouldn't send over women because women live longer and are therefore more likely to develop cancer! Right, pick people with the shortest possible lifespan to maximize their... errr... oh, wait...
They'd better not send American citizens over there, because they'll actually have to keep paying taxes on anything they earn on Mars.