We're in a magical time, a time of transition. Gone are the days when every kid knew what a brake like was, and carried a jack knife. Not yet arrived are the days when every kid carries a pocket computer and knows how to crack a password.
Reaction time isn't just the 0.whatever seconds it takes you to react to a stimulus you're alert to. It's also the amount of time it takes you to notice the stimulus in the first place. IIRC the majority of collisions are caused by distracted drivers.
You're right, reaction time isn't an issue if you're driving properly. But people don't drive properly, and that doesn't only cause reaction time to be important, but also impacts negatively on their reaction time.
It's easier to create a realistic stressful situation with VR than it is without. By "easier" I mean easier to sell to an ethics review board and get volunteers for.
That's all well and good, but the manufacturers want to sell IP cameras and other "Internet of things" stuff to regular consumers. Regular consumers don't "VPN into" things. They tap on an app on their smartphone.
Besides which, unless you're extraordinarily lucky or for some reason spend multiple times the regular rate for Internet service, you have a dynamic DNS address. That means something on your network, whether it's a toaster, desktop computer or your router, has to talk to an external DDNS server to update your IP address in order for you to have something to VPN into. So there's something, probably Chinese made, on your LAN "phoning home" anyway.
It used to be for sure. Configuring port forwarding manually isn't for the average person. Now there are a bunch of ways to poke holes in NAT though. One of the most common is a UPnP IGD, which is a protocol for asking the router to pretty please forward a port for you.
Trivial, is it? As the GP explained, the vast majority of people do not have static IP addresses so it's absolutely necessary to use a DDNS type service. Since the DDNS service has to be a server somewhere that DOES have a static IP address, that is indeed what the kids today call "the cloud."
"The article is about the cost of overall administrative overhead"
Yup. There's a question worth asking. And don't stop at CEOs and presidents. What do all of the people who "supervise" make combined, compared to the cost of the people actually doing the work, and the material costs of actually making something?
I just got an e-mail that someone was appointed to a newly created office of assistant dean (of the faculty, not the university). Apparently the vice dean got tired of doing all the work so he needed to have an assistant dean. Presumably the actual dean hasn't done much for years, ever since the position of vice dean was created. I'm actually afraid to look and see if there's actually another layer in there somewhere....
The signal they detected has a quickly increasing frequency and amplitude, then a ringdown. You can't get that with a supernova, but it fits well what's expected from a couple of black holes spiraling together and merging. Thriip.
IIRC six a year was what they were expecting when they designed LIGO. There was some speculation (not sure how reliable) that that rate might be a bit low, since there were (I believe) two or three other possibilities already in the data.
Yes, a non-random sample is called a biased sample. Variance decreases your confidence in your answer. Bias makes you get the wrong answer, confidently.
The size of the population actually has nothing to do with the size of your required sample, although the complexity of the things you're trying to explain can. You can make valid statistical inferences with only two samples, although you'll have extremely low confidence. Our current sample of thousands of exoplanets would be more than good enough to make meaningful inferences about other solar systems, except that it's so biased. Extending that to the prevalence of life is even worse because our current sample size is one, plus a lot of supposition.
Mmm, you seem correct up until that last bit. Stallman's problem with non-GPL open licenses is that they lack or don't have as strong protections against redistribution. He does have an essay where he says that proprietary software is morally wrong. IIRC he also makes the link to licenses that don't prevent proprietary software. If not in that particular essay, then in other writings.
Given a sample of edge-on solar systems it's not hard to correct for the bias and get the expected number of total solar systems. The large-close bias is a much bigger problem. How do you estimate the number of Earth sized or smaller planets in the habitable zone when we haven't been using good enough instruments for long enough to detect them?
Inferring things about populations from tiny samples is kind of how statistics works. The problem here is that statistics assumes your sample is random, and our sample of exoplanets is definitely biased.
He doesn't even say that, according to the article. He says Earth may be a "mild statistical anomaly." We were pretty sure about that just from the composition of our own solar system.
Not sure where you got "not practical" out of that. It's a piece about an interesting paper to get more efficiency out of the setup. A plain old solar sail (even without the lasers) works pretty well if you're going far enough or you need a little bit of fuel-free thrust for things like station keeping or attitude control.
Slashdot Mirror
We're in a magical time, a time of transition. Gone are the days when every kid knew what a brake like was, and carried a jack knife. Not yet arrived are the days when every kid carries a pocket computer and knows how to crack a password.
This is why measures of dispersion are important. Never trust a number without a confidence interval.
Slashdot hasn't been a tech site for a decade. It was infested by script kiddies in 2003 and now they're all cranky middle aged political pundits.
I'll bet you that out of 10000 miles of driving there's at least one where any human driver will be tuned out. Hopefully not all at the same time.
Ridden in an airplane lately?
Reaction time isn't just the 0.whatever seconds it takes you to react to a stimulus you're alert to. It's also the amount of time it takes you to notice the stimulus in the first place. IIRC the majority of collisions are caused by distracted drivers.
You're right, reaction time isn't an issue if you're driving properly. But people don't drive properly, and that doesn't only cause reaction time to be important, but also impacts negatively on their reaction time.
It's easier to create a realistic stressful situation with VR than it is without. By "easier" I mean easier to sell to an ethics review board and get volunteers for.
I wonder if a "not having your head up your ass workshop" would get as many attendees as a "mindfulness workshop."
That's all well and good, but the manufacturers want to sell IP cameras and other "Internet of things" stuff to regular consumers. Regular consumers don't "VPN into" things. They tap on an app on their smartphone.
Besides which, unless you're extraordinarily lucky or for some reason spend multiple times the regular rate for Internet service, you have a dynamic DNS address. That means something on your network, whether it's a toaster, desktop computer or your router, has to talk to an external DDNS server to update your IP address in order for you to have something to VPN into. So there's something, probably Chinese made, on your LAN "phoning home" anyway.
It used to be for sure. Configuring port forwarding manually isn't for the average person. Now there are a bunch of ways to poke holes in NAT though. One of the most common is a UPnP IGD, which is a protocol for asking the router to pretty please forward a port for you.
Not much of an internet of things then, is it?
Trivial, is it? As the GP explained, the vast majority of people do not have static IP addresses so it's absolutely necessary to use a DDNS type service. Since the DDNS service has to be a server somewhere that DOES have a static IP address, that is indeed what the kids today call "the cloud."
Kind of like social networking, Web x.0 and "the cloud?" People get paid the big bucks to come up with these things!
"The article is about the cost of overall administrative overhead"
Yup. There's a question worth asking. And don't stop at CEOs and presidents. What do all of the people who "supervise" make combined, compared to the cost of the people actually doing the work, and the material costs of actually making something?
I just got an e-mail that someone was appointed to a newly created office of assistant dean (of the faculty, not the university). Apparently the vice dean got tired of doing all the work so he needed to have an assistant dean. Presumably the actual dean hasn't done much for years, ever since the position of vice dean was created. I'm actually afraid to look and see if there's actually another layer in there somewhere....
The signal they detected has a quickly increasing frequency and amplitude, then a ringdown. You can't get that with a supernova, but it fits well what's expected from a couple of black holes spiraling together and merging. Thriip.
IIRC six a year was what they were expecting when they designed LIGO. There was some speculation (not sure how reliable) that that rate might be a bit low, since there were (I believe) two or three other possibilities already in the data.
Yes, a non-random sample is called a biased sample. Variance decreases your confidence in your answer. Bias makes you get the wrong answer, confidently.
The size of the population actually has nothing to do with the size of your required sample, although the complexity of the things you're trying to explain can. You can make valid statistical inferences with only two samples, although you'll have extremely low confidence. Our current sample of thousands of exoplanets would be more than good enough to make meaningful inferences about other solar systems, except that it's so biased. Extending that to the prevalence of life is even worse because our current sample size is one, plus a lot of supposition.
Mmm, you seem correct up until that last bit. Stallman's problem with non-GPL open licenses is that they lack or don't have as strong protections against redistribution. He does have an essay where he says that proprietary software is morally wrong. IIRC he also makes the link to licenses that don't prevent proprietary software. If not in that particular essay, then in other writings.
Given a sample of edge-on solar systems it's not hard to correct for the bias and get the expected number of total solar systems. The large-close bias is a much bigger problem. How do you estimate the number of Earth sized or smaller planets in the habitable zone when we haven't been using good enough instruments for long enough to detect them?
Inferring things about populations from tiny samples is kind of how statistics works. The problem here is that statistics assumes your sample is random, and our sample of exoplanets is definitely biased.
He doesn't even say that, according to the article. He says Earth may be a "mild statistical anomaly." We were pretty sure about that just from the composition of our own solar system.
Oh, so you have seen them! One AC to DC converter for several lights. Exactly as I suggested in the... GGGGP?
You made up the "for the whole building" yourself as a straw man.
Solar sails are indeed not a great idea for lifting things off planets. Of course, that's not what we're discussing here.
Time dilation is insignificant at 5% of light speed. It's about 1/10 of 1%.
Not sure where you got "not practical" out of that. It's a piece about an interesting paper to get more efficiency out of the setup. A plain old solar sail (even without the lasers) works pretty well if you're going far enough or you need a little bit of fuel-free thrust for things like station keeping or attitude control.
You know we've launched solar sails, right?