Exactly, some sort of receipt would be a good thing. But it shouldn't be very hard for them to provide enough of one, since they don't have to manage the conversion to dollars themselves.
One final issue - generally when you donate more than $10-20 to charity you get a tax receipt. $150-500 quite a bit more than $10.
The last time I donated to charity (clothing, not money), I got a receipt that said I had donated, but not how much -- I was responsible for filling in the details and providing any documentation of value I needed. If that's acceptable for Good Will, etc,., it should work for Folding@Home. They don't need to come up with a dollar figure, you can do that. They already tell you how many work units you did, right? Accounting for the electricity cost is your problem, but they should provide the details of who you donated it to that is needed for tax purposes.
Well, yeah, if you keep yourself out of the database then it won't have much info on you. Generally I do the same thing, privacy reasons and all that, but I'm aware there are (at least potentially) benefits I'm missing out on. So far, though, I'd rather have my privacy.
Ignoring for a moment the other ways to track me, I rather like being semi-permanently logged in to/. and a host of other sites. When I'm buying something, I don't want to have to go manually unblock the site so it can store my shopping cart data.
Does anyone know of a way to only block the "evil" cookies? I'd love something that blocked the tracking cookies, let the shopping cart ones through, and didn't require me to figure out which was which for each and every cookie.
Off the top of my head, break the problem into pieces. If it's trying to figure out what type of TV show you'll like, your political tastes are probably only peripherally relevant. Unique as you might be in total, I highly doubt that every piece of your tastes are unique. So maybe you correlate with group A for TV shows, B for techie news sites, C for... That just means that it needs to figure out what category the item in question belongs to before deciding whether or not you're likely to like it. And those categories don't have to be an exact match. Maybe your tastes are better described as things group X likes that group Y doesn't like, even though those two groups don't correlate strongly with each other (positively or negatively). Now, teasing out those correlations in a computationally tractable manner might be quite difficult (and I have no clue how to go about it), but I expect they're there to be found if some sufficiently clever person attacks the problem in the right way.
That would be the part where intelligent algorithms are needed. Hopefully it would observe that the content correlated poorly with love for things Man Was Not Meant to Know, and not bother using your Cthulhu fan club contacts to decide whether to recommend it for you.
Obviously, for this to be interesting, you need both good algorithms and (lots of) good data. Without either, the other doesn't help much.
By finding people with similar tastes, and showing you things they liked (well, more complex than that, but you get the idea). After all, if you have one in a million tastes, that means there are a couple thousand people online with similar tastes -- and several hundred of those even speak English. If the algorithms work well, then the computers have the potential advantage over humans of having *lots* of data to work with.
What? No, I'm not missing that at all. The OP asked why companies aren't clearly stating what licenses are compatible. The responder gave an answer amounting to saying that would be hard for the company. I explained why, if the company wanted to, it would be quite easy to do a good enough job to be useful (even if less than perfect). Given that they aren't doing it, and that it would be easy, one must conclude that there is some other reason -- like them not caring about other licenses than their own.
Then again, I suppose I could just throw my hands in the air and give the companies a free pass for not being helpful because it would be too difficult for them. But, I choose to instead point out why what they're doing is not just laziness but rather being actively greedy. It's remarkably hard to change things if you're unwilling to speak up about the problems, and instead just give up hope on anyone else improving their behavior.
Sure, I shouldn't count on them doing the right thing, nor do I expect them too -- but that's no excuse for me, you, or the original responder to give them a free pass because doing the right thing would be "hard." Especially when it wouldn't even be hard at all.
My point was that companies could improve things a lot by saying, for each of several popular licenses, whether their kit was compatible. Then, if you want to use some other license, that's fine, but you're on your own forming a legal opinion about the EULA. If, however, you choose one of the ones they did, you don't have anything to worry about. In the current situation, even if you aren't lazy and put in a large amount of effort, you might come to a different (but entirely reasonable) reading of the license than the company -- and that's undesirable for all concerned, no matter who the court eventually sides with.
So pick a few of the popular ones. Half a dozen at most should be enough for most people; the rest can either infer from the list or form their own legal opinion.
License compatibility is based on legal opinion.
Yes, and if the legal opinion of the license author were made known, in a legally binding way, then there wouldn't be any difficulties (as long as you were willing to accept that opinion, obviously). If EULA section blah reads "... and we promise that code released under the GPLv2 does not violate this EULA..." (in legal translation, I suppose), then it wouldn't matter whether I thought the rest of the EULA was compatible or not, because the only other relevant opinion is known and in my favor. Of course, that would require a company that was actually interested in helping its customers and developers, rather than simply protecting itself every way it can think of and then some.
Because that would be hard, create a lot of problematic debris, and be unlikely to guarantee changing the position of the massive bits enough to be helpful.
Using its normal station keeping thrusters to move it out of the way, however, would probably be feasible. We should know with plenty of notice whether Apophis will hit anything, and if so what. (If it's a decommissioned satellite, we could send up a tug -- more work, but it shouldn't be too hard given notice.)
Yeah, it's clearly typical science journalism. For now, though I'm willing to believe there might be something there. I'd like to see an official NASA report on the story.
Of course, no one's mentioned that we'll know much more precisely what will happen by 2029 -- not only whether there's a concern at all, but which satellite would be hit. In which case we could, you know, move the satellite. The do have some station-keeping capability, after all. And even the dead ones could be moved by a tug, given a small amount of notice and a really good reason (I think this qualifies).
The NASA NEO site gives 21M tons, which is much more plausible.
The many observations are what was involved in the NASA JPL orbit determination. My impression is he just said "suppose it hits a satellite, what will that do to the orbit?" and then ran those numbers plus the JPL numbers through the simulator.
I also get the impression the 1/450 number is 1/450 odds of a collision with Earth, *if* it hits a satellite.
I don't know how many satellites there are in geosync, but I'm pretty sure you're right that it's less than 40,000. That might be about right for total satellites, though -- not that that's all that relevant.
I too would like to see an official NASA comment on this.
The NASA NEO site gives 2.1E10 kg, or 2.1E7 tons -- 21M tons, unless I've screwed up the units somewhere or that site is wrong (both possible:) ).
The same site gives vImpact for Apophis at 12.59km/s. I haven't looked at the approach trajectory in detail, but geosynchronous orbit is only 3.07 km/s, so the relative velocity is dominated by Apophis (moving at less than 12.59, but more than the 5.87 km/s vInfinity; I'm too lazy to work out the exact number). It's orbital velocity wrt the Sun is about the same as Earth's, or 30 km/s -- so the 5-10 km/s collision velocity is 15-30% of its orbital velocity, roughly.
It's a small effect, to be sure, but it has a very, very long lever to work with. I'd be reluctant to say he's wrong without actually doing the math myself in far greater detail than either of us has done here.
Collisions in space are actually quite predictable. The asteroid is huge and fast, so the entire satellite gets obliterated -- no random debris falling off, because odds are that the satellite is either entirely within the path or entirely outside it. Supersonic (relative to speed of sound in asteroid / satellite, not something irrelevant like Earth's atmosphere) collisions are basically completely inelastic (details more complex, but reasonably well understood).
Satellites don't vary in mass all that much. The big ones are a few tons to a few tens of tons, once you ignore the ISS. The little ones don't matter, so you ignore them.
Telescope observations can most definitely produce the many nines of precision needed for this work. It goes something like this: on day one, it's within this error bar. On day two, within that error bar. On day a few thousand, this other error bar. Individually, the error bar is large, but as they spread out, the path through every one of them gets rather precisely defined. Imagine positioning a set of 1 meter wide gates across the US -- sure, you can't measure the position of the bowling ball you rolled through them to better than 1 meter at any one point, but by the time it's gone through *all* of them, you have sub-ppm accuracy on its exact angle. Extend the scale a bit and you get the precision needed.
Calling the pocket is the easy part: if it hits, then the piece of the Earth pointed in that direction will be the Atlantic. Sure, it might strike a glancing blow and hit at the edge, but thanks to foreshortening the odds are against that.
The satellite would weigh a few tons. The asteroid weighs 21 million tons. The course change won't be that significant. Which is exactly why it's an interesting case -- if the course change was significant, it would miss us by rather a lot. Remember, small changes get magnified by close interactions with other bodies. So a small change while deep in Earth's gravity well changes the exact location it will be in by rather a lot some time later.
As for the Atlantic, don't forget projection distortions -- the bits of the planet near the horizon are less likely to get hit, per unit planet surface area, because they get foreshortened from the perspective of the incoming asteroid.
I want to see the math. What miscalculation did NASA make? Did they use centimeters instead of meters? Was it a simple math error? Did they use an incorrect statistic?
Why did the kid have access to this information? Why wouldn't he have access to the info? Scientific data gets published. You know, so that other people can read it and check the results. And correct them if they're wrong. Like in this case (though as others have pointed out, it may be less of a correction and more of a clarification).
That's about the one thing I wouldn't do for speaker wire. For a balanced signal-level line, maybe, but I'd have to look at the details. But for speakers, more copper really is better (within reason), and you won't get much copper with sane amounts of Cat6. Don't get me wrong, Cat6 is excellent cable, but it's engineered for all the wrong things.
NASA and Marquardt agree that... [it] will crash into the Atlantic ocean Ah, so there's only a 1 in 450 chance of it hitting earth, but we know which ocean it will land in if it does (7 years after it hits the satellite).
Yes, actually, that's the easy part. We know very precisely when and from what direction it will be coming, the question is will it go left, right, or straight down the middle? (Metaphorically speaking... I don't know the details, for all I know we're above and to the left of the center track.)
Once you know when and what direction, you know which hemisphere. Once you account for projection distortion, that puts the odds as pretty good it lands in an area well less than half of the Earth's surface. Something the size of, say, the Atlantic Ocean.
Yeah, for speakers, it's hard to imagine an environment where you'd need shielding -- with the one exception of long runs near 60Hz mains power.
But, I've also seen noisy enough environments that I had indicator LEDs lighting up purely from the noise power. No, it wasn't turning on some high impedance input -- it was being directly rectified by the LED and producing enough light you could see the results. Of course, that particular problem was solved by fixing the noise issue, but my point is that even commercial parts from reputable suppliers can create a nasty enough environment to be worth worrying about.
Black holes of this size do not give off meaningful amounts of Hawking radiation. Their temperature is far, far below the cosmic microwave background temperature -- so even if they didn't capture matter, they would grow by absorbing background radiation. A one solar mass black hole is at only 60 nanokelvins; heavier black holes are colder. Perfect black bodies at that temperature glow very, very dimly.
Did you read the first half of the sentence? Because you missed the whole point of my comment. As I said, in the sense it is normally used "Turing complete" is insufficient. My point was the phrase is usually used slightly less formally, with unspoken conditions -- namely, what they usually really mean is "X would be Turing complete if it had access to infinite memory." Babbage's analytical engine had a Turing-complete instruction set, but without infinite memory is not Turing complete in the formal sense, even though people speak of it as such. The same is true for every computer ever built. Plenty of computers have been built that are "Turing complete" in the informal sense, but have memory limitations that are sufficient not only to make them not Turing complete in the formal sense, but to mean that they can't run a large program like Linux.
If you're going to be all pedantic about it (which I really don't mind; as should be obvious I do it often enough myself) you really need to read and respond to the whole comment, or at least the entire sentence. Taking quotes that far out of context is childish at best.
Slashdot Mirror
Exactly, some sort of receipt would be a good thing. But it shouldn't be very hard for them to provide enough of one, since they don't have to manage the conversion to dollars themselves.
They're a university. I'm sure they have that taken care of.
All the joys of hashcash, with none of the headaches of writing a cross-platform solution! And there's a third party involved, too!
The last time I donated to charity (clothing, not money), I got a receipt that said I had donated, but not how much -- I was responsible for filling in the details and providing any documentation of value I needed. If that's acceptable for Good Will, etc,., it should work for Folding@Home. They don't need to come up with a dollar figure, you can do that. They already tell you how many work units you did, right? Accounting for the electricity cost is your problem, but they should provide the details of who you donated it to that is needed for tax purposes.
Well, yeah, if you keep yourself out of the database then it won't have much info on you. Generally I do the same thing, privacy reasons and all that, but I'm aware there are (at least potentially) benefits I'm missing out on. So far, though, I'd rather have my privacy.
just turn off your cookies!
Ignoring for a moment the other ways to track me, I rather like being semi-permanently logged in to /. and a host of other sites. When I'm buying something, I don't want to have to go manually unblock the site so it can store my shopping cart data.
Does anyone know of a way to only block the "evil" cookies? I'd love something that blocked the tracking cookies, let the shopping cart ones through, and didn't require me to figure out which was which for each and every cookie.
Off the top of my head, break the problem into pieces. If it's trying to figure out what type of TV show you'll like, your political tastes are probably only peripherally relevant. Unique as you might be in total, I highly doubt that every piece of your tastes are unique. So maybe you correlate with group A for TV shows, B for techie news sites, C for... That just means that it needs to figure out what category the item in question belongs to before deciding whether or not you're likely to like it. And those categories don't have to be an exact match. Maybe your tastes are better described as things group X likes that group Y doesn't like, even though those two groups don't correlate strongly with each other (positively or negatively). Now, teasing out those correlations in a computationally tractable manner might be quite difficult (and I have no clue how to go about it), but I expect they're there to be found if some sufficiently clever person attacks the problem in the right way.
That would be the part where intelligent algorithms are needed. Hopefully it would observe that the content correlated poorly with love for things Man Was Not Meant to Know, and not bother using your Cthulhu fan club contacts to decide whether to recommend it for you.
Obviously, for this to be interesting, you need both good algorithms and (lots of) good data. Without either, the other doesn't help much.
By finding people with similar tastes, and showing you things they liked (well, more complex than that, but you get the idea). After all, if you have one in a million tastes, that means there are a couple thousand people online with similar tastes -- and several hundred of those even speak English. If the algorithms work well, then the computers have the potential advantage over humans of having *lots* of data to work with.
What? No, I'm not missing that at all. The OP asked why companies aren't clearly stating what licenses are compatible. The responder gave an answer amounting to saying that would be hard for the company. I explained why, if the company wanted to, it would be quite easy to do a good enough job to be useful (even if less than perfect). Given that they aren't doing it, and that it would be easy, one must conclude that there is some other reason -- like them not caring about other licenses than their own.
Then again, I suppose I could just throw my hands in the air and give the companies a free pass for not being helpful because it would be too difficult for them. But, I choose to instead point out why what they're doing is not just laziness but rather being actively greedy. It's remarkably hard to change things if you're unwilling to speak up about the problems, and instead just give up hope on anyone else improving their behavior.
Sure, I shouldn't count on them doing the right thing, nor do I expect them too -- but that's no excuse for me, you, or the original responder to give them a free pass because doing the right thing would be "hard." Especially when it wouldn't even be hard at all.
My point was that companies could improve things a lot by saying, for each of several popular licenses, whether their kit was compatible. Then, if you want to use some other license, that's fine, but you're on your own forming a legal opinion about the EULA. If, however, you choose one of the ones they did, you don't have anything to worry about. In the current situation, even if you aren't lazy and put in a large amount of effort, you might come to a different (but entirely reasonable) reading of the license than the company -- and that's undesirable for all concerned, no matter who the court eventually sides with.
There are thousands of different licenses.
So pick a few of the popular ones. Half a dozen at most should be enough for most people; the rest can either infer from the list or form their own legal opinion.
License compatibility is based on legal opinion.
Yes, and if the legal opinion of the license author were made known, in a legally binding way, then there wouldn't be any difficulties (as long as you were willing to accept that opinion, obviously). If EULA section blah reads "... and we promise that code released under the GPLv2 does not violate this EULA ..." (in legal translation, I suppose), then it wouldn't matter whether I thought the rest of the EULA was compatible or not, because the only other relevant opinion is known and in my favor. Of course, that would require a company that was actually interested in helping its customers and developers, rather than simply protecting itself every way it can think of and then some.
Because that would be hard, create a lot of problematic debris, and be unlikely to guarantee changing the position of the massive bits enough to be helpful.
Using its normal station keeping thrusters to move it out of the way, however, would probably be feasible. We should know with plenty of notice whether Apophis will hit anything, and if so what. (If it's a decommissioned satellite, we could send up a tug -- more work, but it shouldn't be too hard given notice.)
Yeah, it's clearly typical science journalism. For now, though I'm willing to believe there might be something there. I'd like to see an official NASA report on the story.
Of course, no one's mentioned that we'll know much more precisely what will happen by 2029 -- not only whether there's a concern at all, but which satellite would be hit. In which case we could, you know, move the satellite. The do have some station-keeping capability, after all. And even the dead ones could be moved by a tug, given a small amount of notice and a really good reason (I think this qualifies).
The NASA NEO site gives 21M tons, which is much more plausible.
The many observations are what was involved in the NASA JPL orbit determination. My impression is he just said "suppose it hits a satellite, what will that do to the orbit?" and then ran those numbers plus the JPL numbers through the simulator.
I also get the impression the 1/450 number is 1/450 odds of a collision with Earth, *if* it hits a satellite.
I don't know how many satellites there are in geosync, but I'm pretty sure you're right that it's less than 40,000. That might be about right for total satellites, though -- not that that's all that relevant.
I too would like to see an official NASA comment on this.
The NASA NEO site gives 2.1E10 kg, or 2.1E7 tons -- 21M tons, unless I've screwed up the units somewhere or that site is wrong (both possible :) ).
The same site gives vImpact for Apophis at 12.59km/s. I haven't looked at the approach trajectory in detail, but geosynchronous orbit is only 3.07 km/s, so the relative velocity is dominated by Apophis (moving at less than 12.59, but more than the 5.87 km/s vInfinity; I'm too lazy to work out the exact number). It's orbital velocity wrt the Sun is about the same as Earth's, or 30 km/s -- so the 5-10 km/s collision velocity is 15-30% of its orbital velocity, roughly.
It's a small effect, to be sure, but it has a very, very long lever to work with. I'd be reluctant to say he's wrong without actually doing the math myself in far greater detail than either of us has done here.
Collisions in space are actually quite predictable. The asteroid is huge and fast, so the entire satellite gets obliterated -- no random debris falling off, because odds are that the satellite is either entirely within the path or entirely outside it. Supersonic (relative to speed of sound in asteroid / satellite, not something irrelevant like Earth's atmosphere) collisions are basically completely inelastic (details more complex, but reasonably well understood).
Satellites don't vary in mass all that much. The big ones are a few tons to a few tens of tons, once you ignore the ISS. The little ones don't matter, so you ignore them.
Telescope observations can most definitely produce the many nines of precision needed for this work. It goes something like this: on day one, it's within this error bar. On day two, within that error bar. On day a few thousand, this other error bar. Individually, the error bar is large, but as they spread out, the path through every one of them gets rather precisely defined. Imagine positioning a set of 1 meter wide gates across the US -- sure, you can't measure the position of the bowling ball you rolled through them to better than 1 meter at any one point, but by the time it's gone through *all* of them, you have sub-ppm accuracy on its exact angle. Extend the scale a bit and you get the precision needed.
Calling the pocket is the easy part: if it hits, then the piece of the Earth pointed in that direction will be the Atlantic. Sure, it might strike a glancing blow and hit at the edge, but thanks to foreshortening the odds are against that.
The satellite would weigh a few tons. The asteroid weighs 21 million tons. The course change won't be that significant. Which is exactly why it's an interesting case -- if the course change was significant, it would miss us by rather a lot. Remember, small changes get magnified by close interactions with other bodies. So a small change while deep in Earth's gravity well changes the exact location it will be in by rather a lot some time later.
As for the Atlantic, don't forget projection distortions -- the bits of the planet near the horizon are less likely to get hit, per unit planet surface area, because they get foreshortened from the perspective of the incoming asteroid.
That's about the one thing I wouldn't do for speaker wire. For a balanced signal-level line, maybe, but I'd have to look at the details. But for speakers, more copper really is better (within reason), and you won't get much copper with sane amounts of Cat6. Don't get me wrong, Cat6 is excellent cable, but it's engineered for all the wrong things.
NASA and Marquardt agree thatYes, actually, that's the easy part. We know very precisely when and from what direction it will be coming, the question is will it go left, right, or straight down the middle? (Metaphorically speaking... I don't know the details, for all I know we're above and to the left of the center track.)
Once you know when and what direction, you know which hemisphere. Once you account for projection distortion, that puts the odds as pretty good it lands in an area well less than half of the Earth's surface. Something the size of, say, the Atlantic Ocean.
Yeah, for speakers, it's hard to imagine an environment where you'd need shielding -- with the one exception of long runs near 60Hz mains power.
But, I've also seen noisy enough environments that I had indicator LEDs lighting up purely from the noise power. No, it wasn't turning on some high impedance input -- it was being directly rectified by the LED and producing enough light you could see the results. Of course, that particular problem was solved by fixing the noise issue, but my point is that even commercial parts from reputable suppliers can create a nasty enough environment to be worth worrying about.
In astonomy and astrophysics, ignition usually refers to fusion, rather than a chemical process.
Black holes of this size do not give off meaningful amounts of Hawking radiation. Their temperature is far, far below the cosmic microwave background temperature -- so even if they didn't capture matter, they would grow by absorbing background radiation. A one solar mass black hole is at only 60 nanokelvins; heavier black holes are colder. Perfect black bodies at that temperature glow very, very dimly.
Did you read the first half of the sentence? Because you missed the whole point of my comment. As I said, in the sense it is normally used "Turing complete" is insufficient. My point was the phrase is usually used slightly less formally, with unspoken conditions -- namely, what they usually really mean is "X would be Turing complete if it had access to infinite memory." Babbage's analytical engine had a Turing-complete instruction set, but without infinite memory is not Turing complete in the formal sense, even though people speak of it as such. The same is true for every computer ever built. Plenty of computers have been built that are "Turing complete" in the informal sense, but have memory limitations that are sufficient not only to make them not Turing complete in the formal sense, but to mean that they can't run a large program like Linux.
If you're going to be all pedantic about it (which I really don't mind; as should be obvious I do it often enough myself) you really need to read and respond to the whole comment, or at least the entire sentence. Taking quotes that far out of context is childish at best.