What-if 0007: "Everybody Out"

What if there was a forum for discussing these?

Moderators: Moderators General, Prelates, Magistrates

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Tue Sep 04, 2012 5:24 am UTC

J Thomas wrote:
Max™ wrote:On the subject of relativity precision, GPS affects us every day and ONLY exists because of relativity, and the accuracy of GPS is directly related to the level of precision in our understanding of relativity.


I'm sorry. This is utter and complete nonsense. I can't blame you for not thinking it through since so many other people repeat it uncritically. With so much repetition it must seem like it doesn't need any thought, it must be true or they wouldn't say it so much.

Our engineers don't need a complete theory to explain in detail how everything works before they can use it. If they built a GPS system without knowing about relativity, they could and would measure the systematic errors and adjust for them. They have to do that *anyway*, because they get lots of small errors from unknown sources, that add up. The systematic ones are the easiest to deal with. The idea that it's so very important that there is a theory which sort of predicts one of the systematic errors is ludicrous.

No, this is not nonsense, see, I understand relativity pretty well after the last twenty some years of study.

The effects gravity and velocity would have on GPS systems would lead to a discovery of relativity if it was not already known. It isn't "just an error" as you imply.

The fact that you think it "sort of predicts a systematic error" is ludicrous.
Current theories about what happens inside an iron nucleus tend to suppose that maybe something the size of an alpha particle tends to have some stability in there, and there's some sort of binding energy mediated by some sort of particle.... Some theories suppose an almost crystalline structure of smaller pieces. Others emphasize quarks. It might turn out that individual protons and neutrons maintain their identity inside a nucleus, or maybe they have no more existence than individual H2O molecules in liquid water. You can look at the electromagnetic radiation that comes out of nuclei to get some idea how they're organized, but when you induce them to radiate that stuff they are not in the same states they're in when they aren't radiating it.... It was only natural to assume that they have a structure analogous to electron shells, and interpret everything in those terms, and it's possible to get reasonable results that way.

There's only one sound theoretical model which explains particle physics, there are lots of theoretical structures and far more hypothetical explanations, but at least when I discuss a theory, I mean an experimentally tested and sound explanation for a certain set of phenomena.

So OK, I ask you again, when a nucleus emits a beta particle, did it create an electron from scratch just at that moment, or was the electron already there? Perhaps inside a neutron? And when a nucleus emits a positron, did it create that positron from scratch just at that moment, or was the positron already there? Perhaps inside a proton? Do you understand experiments which confirm one of those interpretations over the other?

The nucleus underwent a transition which released enough energy for spontaneous production of another particle.

I don't say the iron atom in the new state would have the same energy it had when it contained one more neutrino. How could it? That would violate various physical laws.

Iron atoms don't "contain" neutrinos, they are released in certain particle interactions, but stating they are contained within a particular atom is extremely inaccurate.

I probably don't understand all the subtleties of the conservation laws. I notice that when talented amateurs discuss this sort of thing typically some of them are confused. Maybe I could learn something here. If it's possible to set up special conditions that result in atoms preferentially emitting neutrinos (of a type which is currently unknown) in a single direction, how does that violate symmetry? It's an unknown type of neutrino, imagine that something as simple as an electric or magnetic field affected the direction that atoms emit these neutrinos. How does that violate symmetry?

A symmetry is best explained as the ability to rotate or reverse the orientation of an experiment without detecting a variation.

Sometimes a particle doesn't care if you point it up, left, rotate it 180 or 360 degrees, or move it two feet to the left or right.

Sometimes a particle doesn't care if it is yesterday or today.

Sometimes a particle doesn't care if it has right handed spin or left handed spin. If you look at an interaction in a mirror, you shouldn't be able to distinguish it from a non-reflected experiment.


When those symmetries are violated, that relates to a conservation law, as Emmy Noether noted.

But no matter, even if the neutrinos were emitted in all directions equally, all we need is a neutrino mirror and we can get well over half of them going in roughly the right direction. Currently I have no idea how to make a neutrino mirror, of course. And if it's possible I don't know how heavy it would be or how much energy would be required to maintain it. But hey, tell somebody in 1812 that you have a magnet you can turn on and off instantly, and they probably wouldn't believe it until they saw it. There were people making guesses that electricity was somehow connected to magnetism, but nobody had found a connection. Do you know that neutrino mirrors violate physical law?

I know that turning a neutrino mirror on right here would induce thrust uh... well, not sure which part of the planet you're on, or where the sun is relative to you, but right now where I am, the thrust would be almost straight up.

Sure, and if sugar wasn't stable against numerous effects and last a certain length of time you couldn't have a candy cane. But you can still suck on it, when it isn't being quite so stable.

Yeah, except sugar doesn't comprise most of the visible matter in the universe, and isn't an elementary particle, so this analogy is rather useless.

I don't propose endless free energy. If it turns out that we can convert an iron nucleus entirely into neutrinos, then after it is converted entirely into neutrinos it will be gone. That would be the end. Since neutrinos aren't observed to have charge, after emitting neutrinos an atom should still have all its charges. But it might not continue to be stable that way, and might eventually eject some charges. That could give radioactivity etc. Hoping that the whole thing could become neutrinos is pretty much a best-case wish.

>.>

Charge is a property of a particle, not something which would be emitted. Releasing neutrinos will have to reduce the mass of the constituent particles, and that will definitely change it from being iron.

Turning it entirely into neutrinos doesn't sound like an experiment I want to be around, go do that on a moon, one of the ones around Jupiter.

You aren't qualified to decide that. If the future path of physics runs qualitatively like its past and present, we will find that our current understanding of physics is fundamentally incomplete ad maybe flawed, but it still works adequately in a limited range of circumstances. What will be possible with the new physics, in circumstances we currently don't know how to set up? You don't know. If you did know, you would understand the new physics, and you do not.

You're proposing that while there are no wizards today, potentially there could be wizards later, and since I can't say what wizards can do, I can't reasonably say that it is unlikely there will be wizards in the future?

You don't know what is possible outside the circumstances you understand. You don't know what it takes to get outside the circumstances you understand.

Now, you're not qualified to say whether I know what the boundaries of my own understanding are.

I must have been unclear. Look, do you use electricity o make liquid hydrogen? Yes. Where does the electricity come from? A little bit is hydroelectric, and some is nuclear, and a whole lot comes from burning fossil fuels. Yes, you use fossil fuels to make liquid hydrogen. Similarly with hydrazine. To make the precursors you do various endothermic reactions, and the energy comes mostly from fossil fuels.

You need energy, not fossil fuels, that fossil fuels are cheap doesn't mean we can't make rocket fuel without them. There is nothing about fossil fuels which is required for production of energy, so your argument is broken.

You ever play one of those games where you drop coins into water and try to land them in the right spot? Crowbars are fine if you don't care what you hit. Just keep dropping them until you're sure you hit what you want. You can probably get a lot better precision with masers or maybe with particle beams. But that aside, it's in general easy to attack earth from space, particularly if you have spare mass to drop. And it's hard to attack space from earth. If you live on earth, you are better off not to create a culture in space that considers itself separate from your culture.

I'm pretty good at the coin drop games actually, but you can make a guided crowbar just as easily, though that isn't your point.

Well see, ecologists have harvested their low hanging fruit, and to make their next breakthroughs they need to build artificial sealed ecosystems and study them. We need about 100 glassed structures about one square kilometer each, a cost of perhaps $1 billion apiece. If all the money isn't available right away we could start out building a few of them and build the rest over time. We have to get this to test our ecological theories. It will be worth the money, I promise.

You disagree? You think the money should be spent on physics experiments instead? OK, is there a way we can resolve this scientifically to decide which experiments are worth more -- before we actually get the experimental results and find out what we win?

Why is there an either/or?

All science is a worthwhile expenditure compared to blowing shit up or bailing out bankers when they fuck up the economy.

You're making a very disingenuous argument here.
mu

rmsgrey
Posts: 3655
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: What-if 0007: Everybody Out

Postby rmsgrey » Tue Sep 04, 2012 12:14 pm UTC

Actually, preferential radiation in one direction only requires P-symmetry violation, and has been observed experimentally (so has CP-symmetry violation). In both cases, the violations require weak-force interactions - strong and electromagnetic interactions preserve both symmetries.

So, while directed neutrino emissions would break a symmetry, it's a symmetry that's known to break, and the type of interaction that's known to break it.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Tue Sep 04, 2012 12:59 pm UTC

Yeah, but the types of processes he's positing are a bit beyond typical neutrino emissions, and I did know about the handedness of neutrinos already. Some models posit sterile neutrinos as being right-handed, as I recall.
mu

rmsgrey
Posts: 3655
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: What-if 0007: Everybody Out

Postby rmsgrey » Tue Sep 04, 2012 2:43 pm UTC

Max™ wrote:Yeah, but the types of processes he's positing are a bit beyond typical neutrino emissions, and I did know about the handedness of neutrinos already. Some models posit sterile neutrinos as being right-handed, as I recall.


If anything not involving the weak force can generate neutrinos, then it's something currently unknown, so there's no reason not to allow it to break parity conservation as well...

Me, I favour the rainbow drive - that's the one where you fart rainbows, and get propelled at faster-than-light speeds to wherever you most want to be - it also relies on as yet undiscovered physics.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Tue Sep 04, 2012 3:45 pm UTC

Max™ wrote:
J Thomas wrote:
Max™ wrote:On the subject of relativity precision, GPS affects us every day and ONLY exists because of relativity, and the accuracy of GPS is directly related to the level of precision in our understanding of relativity.


I'm sorry. This is utter and complete nonsense. I can't blame you for not thinking it through since so many other people repeat it uncritically. With so much repetition it must seem like it doesn't need any thought, it must be true or they wouldn't say it so much.

Our engineers don't need a complete theory to explain in detail how everything works before they can use it. If they built a GPS system without knowing about relativity, they could and would measure the systematic errors and adjust for them. They have to do that *anyway*, because they get lots of small errors from unknown sources, that add up. The systematic ones are the easiest to deal with. The idea that it's so very important that there is a theory which sort of predicts one of the systematic errors is ludicrous.

No, this is not nonsense, see, I understand relativity pretty well after the last twenty some years of study.

The effects gravity and velocity would have on GPS systems would lead to a discovery of relativity if it was not already known. It isn't "just an error" as you imply.


It might easily be true that after we had a working GPS system, physicists who studied the systematic errors would invent relativity from it. It is not true that you have to understand relativity to get a working GPS system.

Current theories about what happens inside an iron nucleus tend to suppose that maybe something the size of an alpha particle tends to have some stability in there, and there's some sort of binding energy mediated by some sort of particle.... Some theories suppose an almost crystalline structure of smaller pieces. Others emphasize quarks. It might turn out that individual protons and neutrons maintain their identity inside a nucleus, or maybe they have no more existence than individual H2O molecules in liquid water. You can look at the electromagnetic radiation that comes out of nuclei to get some idea how they're organized, but when you induce them to radiate that stuff they are not in the same states they're in when they aren't radiating it.... It was only natural to assume that they have a structure analogous to electron shells, and interpret everything in those terms, and it's possible to get reasonable results that way.

There's only one sound theoretical model which explains particle physics, there are lots of theoretical structures and far more hypothetical explanations, but at least when I discuss a theory, I mean an experimentally tested and sound explanation for a certain set of phenomena.


Well, I probably don't understand this very well, but it looks like you have these various particles that can be studied in isolation, and then you have nuclei. And the theories about what happens inside nuclei are wild extrapolations from what you know about particles in isolation, but the particular wild extrapolations that get chosen are compatible with some meager experimental evidence. So it makes a certain sense to figure that there are a finite number of point particles zooming around inside the vast empty spaces inside each nucleus, and they are so sparse in there that you can imagine everything happening from pairwise interactions because it will be vanishingly rare that three of them get close enough to each other at the same time to have three-way effects. All the details make mathematical sense. It's analogous to the math for ideal gases. With ideal gases you can start out imagining that gas molecules are pointsize, but when you compare it to real results you get errors which go away when you assume that each gas molecule has a volume that it excludes other gas molecules from. So molecules are not points after all. But the math for particles inside nuclei works about as well without excluded volume as it does with it, so you might as well think of them as points.

OK, so there's this theory which is compatible with current experimental data. that's what it means for it to be experimentally tested. All of the current theoretical structures and hypothetical explanations are compatible with current experimental data -- the ones that aren't have been disproved.

So OK, I ask you again, when a nucleus emits a beta particle, did it create an electron from scratch just at that moment, or was the electron already there? Perhaps inside a neutron? And when a nucleus emits a positron, did it create that positron from scratch just at that moment, or was the positron already there? Perhaps inside a proton? Do you understand experiments which confirm one of those interpretations over the other?

The nucleus underwent a transition which released enough energy for spontaneous production of another particle.


Yes, I think that's a valid interpretation. So, do you think there are protons inside a nucleus? Or is it just that when some experiment comes out compatible with a proton inside a nucleus, or an alpha particle leaves a nucleus etc, that the nucleus underwent a transition that spontaneously produced a brand new proton or alpha particle on the spot? Does it make any difference which of those is true? I can imagine physics theories which are so fuzzy that it makes no difference whether things actually exist inside nuclei or instead whether nuclei simply produce whatever you observe just in time for you to observe them. Is your theory like that?

I don't say the iron atom in the new state would have the same energy it had when it contained one more neutrino. How could it? That would violate various physical laws.

Iron atoms don't "contain" neutrinos, they are released in certain particle interactions, but stating they are contained within a particular atom is extremely inaccurate.


Are you saying that there have been experiments that prove there are no neutrinos inside nuclei, waiting to get out? Are you sure your interpretation is the only possible way to look at it that can be valid?

I probably don't understand all the subtleties of the conservation laws. I notice that when talented amateurs discuss this sort of thing typically some of them are confused. Maybe I could learn something here. If it's possible to set up special conditions that result in atoms preferentially emitting neutrinos (of a type which is currently unknown) in a single direction, how does that violate symmetry? It's an unknown type of neutrino, imagine that something as simple as an electric or magnetic field affected the direction that atoms emit these neutrinos. How does that violate symmetry?

A symmetry is best explained as the ability to rotate or reverse the orientation of an experiment without detecting a variation.

Sometimes a particle doesn't care if you point it up, left, rotate it 180 or 360 degrees, or move it two feet to the left or right.

Sometimes a particle doesn't care if it is yesterday or today.

Sometimes a particle doesn't care if it has right handed spin or left handed spin. If you look at an interaction in a mirror, you shouldn't be able to distinguish it from a non-reflected experiment.

When those symmetries are violated, that relates to a conservation law, as Emmy Noether noted.


Yes. So imagine that there are some sort of conditions you can create that result in a lot of neutrinos produced all heading north. Then you rotate your equipment 90 degrees, and now you get a lot of neutrinos all heading east. They go one direction, the things that they split off from get pushed the opposite direction. The particles don't care which direction they go, they only care how you set up conditions and you can just as easily set up those conditions in any direction. Which conservation law does this violate?

But no matter, even if the neutrinos were emitted in all directions equally, all we need is a neutrino mirror and we can get well over half of them going in roughly the right direction. Currently I have no idea how to make a neutrino mirror, of course. And if it's possible I don't know how heavy it would be or how much energy would be required to maintain it. But hey, tell somebody in 1812 that you have a magnet you can turn on and off instantly, and they probably wouldn't believe it until they saw it. There were people making guesses that electricity was somehow connected to magnetism, but nobody had found a connection. Do you know that neutrino mirrors violate physical law?

I know that turning a neutrino mirror on right here would induce thrust uh... well, not sure which part of the planet you're on, or where the sun is relative to you, but right now where I am, the thrust would be almost straight up.


Yes. The neutrinos go one direction and we get the equal and opposite force. Does that violate anything?

Sure, and if sugar wasn't stable against numerous effects and last a certain length of time you couldn't have a candy cane. But you can still suck on it, when it isn't being quite so stable.

Yeah, except sugar doesn't comprise most of the visible matter in the universe, and isn't an elementary particle, so this analogy is rather useless.


People keep changing their elementary particles as they get new data. Dalton thought atoms were immutable elements. The Curies etc disproved that before the controversy died down about whether atoms were real. Then we got immutable protons, electrons, and neutrons. As far as I know electrons are still supposed to be elementary particles, but that's just because the data hasn't come in yet to disprove it. Of course, electrons can disappear if you let positrons approach them, and you figure they get made from scratch every time a nucleus changes just the right way.
I think you're talking about relative stability. Each of these particles has a lot of energy locked up in it, but usually you need to spend a lot to extract what's locked up. So they're reasonably stable. Find a way to tunnel through that barrier and maybe you can harvest energy cheap.

I don't propose endless free energy. If it turns out that we can convert an iron nucleus entirely into neutrinos, then after it is converted entirely into neutrinos it will be gone. That would be the end. Since neutrinos aren't observed to have charge, after emitting neutrinos an atom should still have all its charges. But it might not continue to be stable that way, and might eventually eject some charges. That could give radioactivity etc. Hoping that the whole thing could become neutrinos is pretty much a best-case wish.

>.>

Charge is a property of a particle, not something which would be emitted. Releasing neutrinos will have to reduce the mass of the constituent particles, and that will definitely change it from being iron.


You can emit charged particles. You're talking some sort of definitional game here, and I don't see that it's useful. Supposing that it turns out a nucleus can release a neutrino and afterward it has 26 positive charges in its nucleus, I'm going to call it iron. You can call it something else if you want to. Currently we have a handful of stable iron isotopes and a handful of radioactive isotopes. If we find a way to make a few more, that shouldn't be such a big deal.

Turning it entirely into neutrinos doesn't sound like an experiment I want to be around, go do that on a moon, one of the ones around Jupiter.


When I think about it, you'd surely have at least 26 positrons left over. Maybe some heavier particles with positive charge.

You aren't qualified to decide that. If the future path of physics runs qualitatively like its past and present, we will find that our current understanding of physics is fundamentally incomplete ad maybe flawed, but it still works adequately in a limited range of circumstances. What will be possible with the new physics, in circumstances we currently don't know how to set up? You don't know. If you did know, you would understand the new physics, and you do not.

You're proposing that while there are no wizards today, potentially there could be wizards later, and since I can't say what wizards can do, I can't reasonably say that it is unlikely there will be wizards in the future?


Yes, definitely. Since you can't say what wizards can do, it's very very likely that something will come along that can be called a wizard. It's already happened -- in WWII they called physicists wizards when they came up with radar etc.

You don't know what is possible outside the circumstances you understand. You don't know what it takes to get outside the circumstances you understand.

Now, you're not qualified to say whether I know what the boundaries of my own understanding are.


You might be right. If we lived long enough, how would we tell?

How about this -- assuming we both live for 200 more years, imagine it's 2212, and there has been no advance in physics that could be called "revolutionary". Everything has proceeded smoothly to extend the 2012 consensus about physics in predictable ways, with no big surprises and no radical new insights, then you are right. But if discoveries are made that cause any fundamental change in physics thinking, then I am right.

If you thought we'd live long enough to determine that, and you had the chance to make a side bet at even odds about which way it would turn out, which side would you choose?

I must have been unclear. Look, do you use electricity o make liquid hydrogen? Yes. Where does the electricity come from? A little bit is hydroelectric, and some is nuclear, and a whole lot comes from burning fossil fuels. Yes, you use fossil fuels to make liquid hydrogen. Similarly with hydrazine. To make the precursors you do various endothermic reactions, and the energy comes mostly from fossil fuels.

You need energy, not fossil fuels, that fossil fuels are cheap doesn't mean we can't make rocket fuel without them. There is nothing about fossil fuels which is required for production of energy, so your argument is broken.


If we do it this year, it will be largely fossil fuels. And since energy is fungible, if you buy up all the geothermal energy to use for space travel, the price of fossil fuels will rise. If you use more energy for space, somebody will have to live on less energy or else we will burn more fossil fuels faster. It doesn't have to be that way in the long run. But until we develop large alternate energy industries, the energy that gets used will be largely fossil fuels. A hundred years from now the space program won't use fossil fuel, because the fossil fuel will be mostly gone. This year it burns fossil fuel to get the energy to do most of what it does.

Well see, ecologists have harvested their low hanging fruit, and to make their next breakthroughs they need to build artificial sealed ecosystems and study them. We need about 100 glassed structures about one square kilometer each, a cost of perhaps $1 billion apiece. If all the money isn't available right away we could start out building a few of them and build the rest over time. We have to get this to test our ecological theories. It will be worth the money, I promise.

You disagree? You think the money should be spent on physics experiments instead? OK, is there a way we can resolve this scientifically to decide which experiments are worth more -- before we actually get the experimental results and find out what we win?

Why is there an either/or?


Because this year, physics can get a lot of money and biology can't. It's been that way since WWII. You can say that physics has to have the money because it's the only possible way that important physics problems will ever be solved, and we have to solve those physics problems quickly. I say the same thing for ecology problems and nobody pays the least bit of attention. We don't have a national priority to solve ecology problems.

You can say with a straight face that we have to spend gibgobs of money on physics because all the cheap physics has already been done, and it's like you don't even notice the disconnect there.

All science is a worthwhile expenditure compared to blowing shit up or bailing out bankers when they fuck up the economy


Agreed. But physics does get a big slice of money, while other disciplines mostly do not. You take it for granted that physics deserves to get it. For myself, if it's a choice between giving it to physics versus spending even more to blow stuff up, I want physics to have it. If it's a choice between physics keeping it or getting it spread more among disciplines, I want it spread. Better stlll would be more science funding all round.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Tue Sep 04, 2012 5:52 pm UTC

J Thomas wrote:It might easily be true that after we had a working GPS system, physicists who studied the systematic errors would invent relativity from it. It is not true that you have to understand relativity to get a working GPS system.

You have to account for relativistic effects to get accurate GPS systems, yes you could do it by trial and error, no that isn't particularly likely to happen anywhere near as quickly as it would if relativity had already been worked out beforehand.

Well, I probably don't understand this very well, but it looks like you have these various particles that can be studied in isolation, and then you have nuclei. And the theories about what happens inside nuclei are wild extrapolations from what you know about particles in isolation, but the particular wild extrapolations that get chosen are compatible with some meager experimental evidence.

Well, you can study protons and neutrons, but the particles they are composed of can't be isolated.

They aren't really wild extrapolations though.

So it makes a certain sense to figure that there are a finite number of point particles zooming around inside the vast empty spaces inside each nucleus, and they are so sparse in there that you can imagine everything happening from pairwise interactions because it will be vanishingly rare that three of them get close enough to each other at the same time to have three-way effects. All the details make mathematical sense. It's analogous to the math for ideal gases. With ideal gases you can start out imagining that gas molecules are pointsize, but when you compare it to real results you get errors which go away when you assume that each gas molecule has a volume that it excludes other gas molecules from. So molecules are not points after all. But the math for particles inside nuclei works about as well without excluded volume as it does with it, so you might as well think of them as points.

No, the math for point particles introduces numerous issues, I don't even know where to begin listing them. Your information is several decades out of date if you think point particles are still considered reasonable approximations.

OK, so there's this theory which is compatible with current experimental data. that's what it means for it to be experimentally tested. All of the current theoretical structures and hypothetical explanations are compatible with current experimental data -- the ones that aren't have been disproved.

Experimentally tested means it made predictions which were confirmed AND is compatible with prior data.

Not all hypothetical models are testable, certainly many are untestable without high energy experiments.

Yes, I think that's a valid interpretation. So, do you think there are protons inside a nucleus? Or is it just that when some experiment comes out compatible with a proton inside a nucleus, or an alpha particle leaves a nucleus etc, that the nucleus underwent a transition that spontaneously produced a brand new proton or alpha particle on the spot? Does it make any difference which of those is true? I can imagine physics theories which are so fuzzy that it makes no difference whether things actually exist inside nuclei or instead whether nuclei simply produce whatever you observe just in time for you to observe them. Is your theory like that?

There are blobs of stuff which correspond to quarks glued into proton-shaped bundles packed among neutron-shaped bundles with fuzzy clouds where electrons are likely to be found all around them.

When we shoot things into nucleons they bounce off in ways which indicate the above is close enough to be useful.

When we smash nucleons together hard enough to dislodge smaller pieces, the debris is consistent with models as I very roughly described above.

When we impart enough energy to nucleons before smashing them together, we find it produces "interesting" debris, things which aren't normally found associated with said nucleons, and are best though of as a product of the energies released in the collision.

When the energy densities in a collision are high enough, the stuff inside of nucleons can get smushed into new states which are too energetic to remain intact for long, and can only be identified by looking at what they decompose into.

Sufficiently developed models can produce very specific predictions of what should be found and in what quantities those results should be found in. The recent findings at the LHC of particles which correspond to Higgs decay events are further evidence that the standard model of particle physics accurately represents the real world.

Are you saying that there have been experiments that prove there are no neutrinos inside nuclei, waiting to get out? Are you sure your interpretation is the only possible way to look at it that can be valid?

I am saying that there is no interpretation which proposes such a thing and has made testable predictions which have been experimentally confirmed, so there is little benefit in examining such an interpretation.

Yes. So imagine that there are some sort of conditions you can create that result in a lot of neutrinos produced all heading north. Then you rotate your equipment 90 degrees, and now you get a lot of neutrinos all heading east. They go one direction, the things that they split off from get pushed the opposite direction. The particles don't care which direction they go, they only care how you set up conditions and you can just as easily set up those conditions in any direction. Which conservation law does this violate?

Making particles preferentially emit neutrinos in a single direction is more complicated than you seem to think.

Doing so in a manner which could allow you to extract thrust is stacking complications on top of complications on top of a particle which barely interacts with anything.

The bit about it probably violating conservation of momentum is a minor consideration at this point.
Yes. The neutrinos go one direction and we get the equal and opposite force. Does that violate anything?

Assuming you have a magical mirror that reflects particles which ignore everything else? I suppose not.

People keep changing their elementary particles as they get new data. Dalton thought atoms were immutable elements. The Curies etc disproved that before the controversy died down about whether atoms were real. Then we got immutable protons, electrons, and neutrons. As far as I know electrons are still supposed to be elementary particles, but that's just because the data hasn't come in yet to disprove it. Of course, electrons can disappear if you let positrons approach them, and you figure they get made from scratch every time a nucleus changes just the right way.
I think you're talking about relative stability. Each of these particles has a lot of energy locked up in it, but usually you need to spend a lot to extract what's locked up. So they're reasonably stable. Find a way to tunnel through that barrier and maybe you can harvest energy cheap.

The probability of the quarks in a proton tunneling into a different state (and collapsing into a black hole which then explodes) is low enough that it hasn't happened with any detectable rate in the last 13 billion years, and given the bit about supporting mass I mentioned, it shouldn't happen in anything less than 1040 years, as I recall.

You're essentially saying "if magic is real, we can use magic to get neutrinos out of atoms, and magically reflect them so we can magically fly around" at this point.

You can emit charged particles. You're talking some sort of definitional game here, and I don't see that it's useful. Supposing that it turns out a nucleus can release a neutrino and afterward it has 26 positive charges in its nucleus, I'm going to call it iron. You can call it something else if you want to. Currently we have a handful of stable iron isotopes and a handful of radioactive isotopes. If we find a way to make a few more, that shouldn't be such a big deal.

I'm not playing a definition game, I'm saying referring to "positive charges left in a nucleus" isn't sensible. Additionally I should note that making a particle emit a neutrino without changing the charge or mass or spin or whatnot basically amounts to "a wizard did it", if you don't want to call it magic, that's ok, just don't pretend it's physics.

When I think about it, you'd surely have at least 26 positrons left over. Maybe some heavier particles with positive charge.

O.o

Don't get around me with 26 positrons and a hunk of iron >.>, it's not going to be a very large explosion, but the gamma emissions will be nasty.

Yes, definitely. Since you can't say what wizards can do, it's very very likely that something will come along that can be called a wizard. It's already happened -- in WWII they called physicists wizards when they came up with radar etc.

No one who knew anything about science did that, laymen calling physicists wizards is amusing, but it is not significant as far as predicting future science.

You might be right. If we lived long enough, how would we tell?

How about this -- assuming we both live for 200 more years, imagine it's 2212, and there has been no advance in physics that could be called "revolutionary". Everything has proceeded smoothly to extend the 2012 consensus about physics in predictable ways, with no big surprises and no radical new insights, then you are right. But if discoveries are made that cause any fundamental change in physics thinking, then I am right.

You used the word consensus, a consensus doesn't really count for much.

I never said there would be no big surprises or insights, I said there will be no wizards.

You will not be right about wizards, I am confident of this.

If you thought we'd live long enough to determine that, and you had the chance to make a side bet at even odds about which way it would turn out, which side would you choose?

There will be no wizards, physics in 200 years will relate to current physics in an understandable manner, though it will have advanced, there will be no point where we discard all of current physics in favor of wizardry.

Because this year, physics can get a lot of money and biology can't. It's been that way since WWII. You can say that physics has to have the money because it's the only possible way that important physics problems will ever be solved, and we have to solve those physics problems quickly. I say the same thing for ecology problems and nobody pays the least bit of attention. We don't have a national priority to solve ecology problems.

You can say with a straight face that we have to spend gibgobs of money on physics because all the cheap physics has already been done, and it's like you don't even notice the disconnect there.

I'm curious where you get the idea that ecological/environmental/climatological sciences aren't getting millions of dollars a year?

Agreed. But physics does get a big slice of money, while other disciplines mostly do not. You take it for granted that physics deserves to get it. For myself, if it's a choice between giving it to physics versus spending even more to blow stuff up, I want physics to have it. If it's a choice between physics keeping it or getting it spread more among disciplines, I want it spread. Better stlll would be more science funding all round.

Image
>.>
mu

KrytenKoro
Posts: 1487
Joined: Tue Apr 05, 2011 2:58 pm UTC

Re: What-if 0007: Everybody Out

Postby KrytenKoro » Tue Sep 04, 2012 7:09 pm UTC

rmsgrey wrote:
Max™ wrote:Yeah, but the types of processes he's positing are a bit beyond typical neutrino emissions, and I did know about the handedness of neutrinos already. Some models posit sterile neutrinos as being right-handed, as I recall.


If anything not involving the weak force can generate neutrinos, then it's something currently unknown, so there's no reason not to allow it to break parity conservation as well...

Me, I favour the rainbow drive - that's the one where you fart rainbows, and get propelled at faster-than-light speeds to wherever you most want to be - it also relies on as yet undiscovered physics.

I've heard they are currently studying cats as possible sources of natural rainbow drives.
From the elegant yelling of this compelling dispute comes the ghastly suspicion my opposition's a fruit.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Wed Sep 05, 2012 5:11 am UTC

Max™ wrote:
J Thomas wrote:It might easily be true that after we had a working GPS system, physicists who studied the systematic errors would invent relativity from it. It is not true that you have to understand relativity to get a working GPS system.

You have to account for relativistic effects to get accurate GPS systems, yes you could do it by trial and error, no that isn't particularly likely to happen anywhere near as quickly as it would if relativity had already been worked out beforehand.


You could easily be right. But the experiment has not been done. I am in general dissatisfied about this topic. When I first looked at it, I found several reports from people who claimed they were engineers who had participated in developing GPS, and they said that there were at that time no relativistic corrections, and in fact the system did work adequately without relativistic correction. I noticed that there were a large number of other corrections needed. http://en.wikipedia.org/wiki/Error_anal ... ing_System I imagined that perhaps without an explicit relativistic correction, the relativistic errors might get rolled into the other error correction and corrected as if they had other sources. But it looked like a whole lot of work to check whether that could work.

Then I saw that there were Creation Science guys who for some inexplicable reason wanted to doubt relativity, and they argued that GPS could work without relativity because if it couldn't they'd look bad. Meanwhile the anti-CS people used GPS as their trump card, arguing that nobody could make GPS work unless they thoroughly understood relativity. The fact that GPS cannot work without relativity proves that relativity is 100% correct and also important. And it was clear that at least 99% of the people who argued about it didn't actually understand GPS any better than I did. When two sides of an ideological divide start shouting their slogans, truth suffers. And I can't trust CS guys not to provide falsified reports that appear to support their position.

Here is a paper that explains some things, unfortunately it would take some time to be sure I understood it.
http://tycho.usno.navy.mil/ptti/1996/Vol%2028_16.pdf
They appear to say that at that time, GPS did not use relativistic corrections because for users at or near the surface of the earth the error would be less than 1 cm. Because of their methods of calculation, some of the errors cancelled while others were averaged out by the Kalman filter. Relativistic errors would accumulate to unacceptable levels over the course of a day, but they were not allowed to accumulate -- they were discarded with each successive recalculation in about 0.1 second, and the resulting errors amounted to around 2.5 millimeters.

In principle, the critics of GPS in the relativity debate have not been completely wrong. The
neglected gamma factor could hurt us. The OCS software should be reformulated. Nevertheless, in
practice, neglect of relativity does not now contribute measurably to the GPS error budget, as
the OCS software is currently configured.


So -- relativity is much more correct than the newtonian approach, it does affect GPS, and GPS has been redesigned to take relativity into account. However GPS worked adequately before they corrected for relativity.

Here is an interesting link.
http://www.ipgp.fr/~tarantola/Files/Pro ... S/GPS.html
They claim that when GPS was redesigned to correct for relativity, it was done as a galilean system with relativistic corrections. So for example they attempt to set up a single universal time, and "correct" the various moving clocks to fit that standard. The idea is that it would work better to actually do GPS with relativistic math. Each moving clock has its own time, and at any time has its own location, and you get a simpler and better system if you just use it the way it works, than if you pretend you have a newtonian system and add in relativistic fudge factors at critical points.

Well, I probably don't understand this very well, but it looks like you have these various particles that can be studied in isolation, and then you have nuclei. And the theories about what happens inside nuclei are wild extrapolations from what you know about particles in isolation, but the particular wild extrapolations that get chosen are compatible with some meager experimental evidence.

Well, you can study protons and neutrons, but the particles they are composed of can't be isolated.


Yet.

They aren't really wild extrapolations though.


They are reasonable interpretations of the limited data available. But if you make sure that your mental model doesn't go much beyond the data, then you wind up with a mental model that's mostly mush.

So it makes a certain sense to figure that there are a finite number of point particles zooming around inside the vast empty spaces inside each nucleus, and they are so sparse in there that you can imagine everything happening from pairwise interactions because it will be vanishingly rare that three of them get close enough to each other at the same time to have three-way effects. All the details make mathematical sense. It's analogous to the math for ideal gases. With ideal gases you can start out imagining that gas molecules are pointsize, but when you compare it to real results you get errors which go away when you assume that each gas molecule has a volume that it excludes other gas molecules from. So molecules are not points after all. But the math for particles inside nuclei works about as well without excluded volume as it does with it, so you might as well think of them as points.

No, the math for point particles introduces numerous issues, I don't even know where to begin listing them. Your information is several decades out of date if you think point particles are still considered reasonable approximations.


My information is spotty, some recent and some old, I've tended to look for things that make sense in my own context first. It looks to me like the basic idea of a shell model implies that particles inside a nucleus are so small that they can spend a reasonably long time between interactions. Shell models seem to get a lot of attention today. Would you settle for "very small"?

OK, so there's this theory which is compatible with current experimental data. that's what it means for it to be experimentally tested. All of the current theoretical structures and hypothetical explanations are compatible with current experimental data -- the ones that aren't have been disproved.

Experimentally tested means it made predictions which were confirmed AND is compatible with prior data.


That is arbitrary. To meet that standard a theory must be proposed at the right time, when enough of its predictions have been confirmed for it o be interesting, but when important predictions have not been tested yet. Then it must attract enough attention from experimentalists that they test its remaining predictions. You're deciding which ideas to pay attention to, based entirely on historical accident. Better to tentatively accept every hypothesis you can imagine, that is not yet disproved.

Not all hypothetical models are testable, certainly many are untestable without high energy experiments.


If you can't find a testable difference between two hypotheses, then there is no practical difference between them.

Yes, I think that's a valid interpretation. So, do you think there are protons inside a nucleus? Or is it just that when some experiment comes out compatible with a proton inside a nucleus, or an alpha particle leaves a nucleus etc, that the nucleus underwent a transition that spontaneously produced a brand new proton or alpha particle on the spot? Does it make any difference which of those is true? I can imagine physics theories which are so fuzzy that it makes no difference whether things actually exist inside nuclei or instead whether nuclei simply produce whatever you observe just in time for you to observe them. Is your theory like that?

There are blobs of stuff which correspond to quarks glued into proton-shaped bundles packed among neutron-shaped bundles with fuzzy clouds where electrons are likely to be found all around them.


Ah. Yes. As I was saying, when you make sure your mental model doesn't go beyond the data, it turns to mush.

Sufficiently developed models can produce very specific predictions of what should be found and in what quantities those results should be found in. The recent findings at the LHC of particles which correspond to Higgs decay events are further evidence that the standard model of particle physics accurately represents the real world.


Or rather, so far it has been compatible with data from the real world.

Are you saying that there have been experiments that prove there are no neutrinos inside nuclei, waiting to get out? Are you sure your interpretation is the only possible way to look at it that can be valid?

I am saying that there is no interpretation which proposes such a thing and has made testable predictions which have been experimentally confirmed, so there is little benefit in examining such an interpretation.


How will you know whether there's value in such an interpretation until you see how such an interpretation works and what it predicts? It looks to me like you are assuming your conclusion.

Yes. So imagine that there are some sort of conditions you can create that result in a lot of neutrinos produced all heading north. Then you rotate your equipment 90 degrees, and now you get a lot of neutrinos all heading east. They go one direction, the things that they split off from get pushed the opposite direction. The particles don't care which direction they go, they only care how you set up conditions and you can just as easily set up those conditions in any direction. Which conservation law does this violate?

Making particles preferentially emit neutrinos in a single direction is more complicated than you seem to think.


I have no idea how to do it. I'm using it as an example of something that might someday be found, which I hope is compatible with what we know now but goes beyond what we know now.

The bit about it probably violating conservation of momentum is a minor consideration at this point.


OK.

Yes. The neutrinos go one direction and we get the equal and opposite force. Does that violate anything?

Assuming you have a magical mirror that reflects particles which ignore everything else? I suppose not.

People keep changing their elementary particles as they get new data. Dalton thought atoms were immutable elements. The Curies etc disproved that before the controversy died down about whether atoms were real. Then we got immutable protons, electrons, and neutrons. As far as I know electrons are still supposed to be elementary particles, but that's just because the data hasn't come in yet to disprove it. Of course, electrons can disappear if you let positrons approach them, and you figure they get made from scratch every time a nucleus changes just the right way.
I think you're talking about relative stability. Each of these particles has a lot of energy locked up in it, but usually you need to spend a lot to extract what's locked up. So they're reasonably stable. Find a way to tunnel through that barrier and maybe you can harvest energy cheap.

The probability of the quarks in a proton tunneling into a different state (and collapsing into a black hole which then explodes) is low enough that it hasn't happened with any detectable rate in the last 13 billion years, and given the bit about supporting mass I mentioned, it shouldn't happen in anything less than 1040 years, as I recall.[/quote]

I'm happy that black holes are relatively rare.

You're essentially saying "if magic is real, we can use magic to get neutrinos out of atoms, and magically reflect them so we can magically fly around" at this point.


I'm saying that if we don't truly understand the laws of physics yet, then we are likely to find things we don't expect. We know some things not to expect. Most conservation laws will stay true. You can't just suck energy out of nowhere, you need a place for it to come from, and when it's gone from there it's gone. Beyond that, your common sense is utterly unreliable at telling you what unexpected results you should expect not to happen.

You can emit charged particles. You're talking some sort of definitional game here, and I don't see that it's useful. Supposing that it turns out a nucleus can release a neutrino and afterward it has 26 positive charges in its nucleus, I'm going to call it iron. You can call it something else if you want to. Currently we have a handful of stable iron isotopes and a handful of radioactive isotopes. If we find a way to make a few more, that shouldn't be such a big deal.

I'm not playing a definition game, I'm saying referring to "positive charges left in a nucleus" isn't sensible.


Conservation of charge. If a nucleus ejects a positive charge, then it has one less positive charge. If it ejects a negative charge the it has one more positive charge. If it ejects an uncharged particle then its charge is unchanged.

Additionally I should note that making a particle emit a neutrino without changing the charge or mass or spin or whatnot basically amounts to "a wizard did it", if you don't want to call it magic, that's ok, just don't pretend it's physics.


?? if it emits something that has mass then of course its mass will be changed. If it emits something that has spin then its spin will be changed. How could it be otherwise? Why would you expect me to imagine it otherwise? Well, possibly we might find that some conservation law really does get broken. I can't say we will never invent a conservation law that isn't true. But I'm real skeptical about breaking conservation laws.

Because this year, physics can get a lot of money and biology can't. It's been that way since WWII. You can say that physics has to have the money because it's the only possible way that important physics problems will ever be solved, and we have to solve those physics problems quickly. I say the same thing for ecology problems and nobody pays the least bit of attention. We don't have a national priority to solve ecology problems.

You can say with a straight face that we have to spend gibgobs of money on physics because all the cheap physics has already been done, and it's like you don't even notice the disconnect there.

I'm curious where you get the idea that ecological/environmental/climatological sciences aren't getting millions of dollars a year?

Agreed. But physics does get a big slice of money, while other disciplines mostly do not. You take it for granted that physics deserves to get it. For myself, if it's a choice between giving it to physics versus spending even more to blow stuff up, I want physics to have it. If it's a choice between physics keeping it or getting it spread more among disciplines, I want it spread. Better stlll would be more science funding all round.

Image
>.>


Interesting! I tried to look for your data and found this:
http://www.nsf.gov/statistics/nsf12318/pdf/tab19.pdf

For 2009 they give around double your 2008 numbers. I haven't found why they're so different and to me the differences tend to discredit both of them. But not enough for me to argue I could be right. I'm very happy to be wrong about this.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Pfhorrest
Posts: 5487
Joined: Fri Oct 30, 2009 6:11 am UTC
Contact:

Re: What-if 0007: Everybody Out

Postby Pfhorrest » Wed Sep 05, 2012 5:33 am UTC

J Thomas wrote:Conservation of charge. If a nucleus ejects a positive charge, then it has one less positive charge. If it ejects a negative charge the it has one more positive charge. If it ejects an uncharged particle then its charge is unchanged.

I think you really do understand this, but the way you're saying it is completely incorrect, because there is no such thing as "a charge". A particle has charge, and that charge can possibly have a range of values besides 1 or -1, but even in that case it doesn't have charges.

In this case, a nucleus has a charge equal to however many protons are in it. If one of those protons becomes a neutron, its charge decreases; but it doesn't "eject a charge". It will eject something, an antineutrino, but that is a particle that doesn't have have a charge (much less is a charge). The former proton loses some charge because it has absorbed an electron, which has negative charge. Even if the nucleus ejects a proton, its charge decreases, but it hasn't "ejected a charge" - it has ejected a proton, which has charge, and which took that charge with it, reducing the charge of the nucleus.

I think with you it's just a linguistic matter but it might belie a misunderstanding of the system we're trying to talk about, so it's important to clear it up. It's like saying "Someone gave me a money" or "I had to pay a money" - money is a quantity, different objects (like coins and bills) can carry different values of it, and you can exchange those objects, but none of those objects is "a money", and talking like that sounds like you don't understand what money is, so if you do, best to avoid giving that appearance.

By the way, we do actually have a starship drive design already in place (just hung up on political matters mostly, as I understand it) which extracts mass from atoms (reducing them to less massive atoms) and shoots high-energy particles away from them as a means of propulsion, just like you're saying. Some of those particles are even neutrinos. It's called Project Orion.
Forrest Cameranesi, Geek of All Trades
"I am Sam. Sam I am. I do not like trolls, flames, or spam."
The Codex Quaerendae (my philosophy) - The Chronicles of Quelouva (my fiction)

User avatar
PM 2Ring
Posts: 3715
Joined: Mon Jan 26, 2009 3:19 pm UTC
Location: Sydney, Australia

Re: What-if 0007: Everybody Out

Postby PM 2Ring » Wed Sep 05, 2012 8:28 am UTC

J Thomas wrote:
Well, you can study protons and neutrons, but the particles they are composed of can't be isolated.
Yet.


Well, people tried for a long time to isolate quarks, in order to prove that the quark theory was valid but they failed, and that failure almost killed quark theory. And bigger particle accelerators and fancier techniques didn't help. These days, it's considered to be impossible to isolate any particle with a color charge due to color confinement, but as Wikipedia says:

The reasons for quark confinement are somewhat complicated; no analytic proof exists that quantum chromodynamics should be confining. The current theory is that confinement is due to the force-carrying gluons having color charge. As any two electrically-charged particles separate, the electric fields between them diminish quickly, allowing (for example) electrons to become unbound from atomic nuclei. However, as two quarks separate, the gluon fields form narrow tubes (or strings) of color charge, which tend to bring the quarks together as though they were some kind of rubber band. This is quite different in behavior from electrical charge. Because of this behavior, the color force experienced by the quarks in the direction to hold them together, remains constant, regardless of their distance from each other,[3][4] at around 10,000 Newtons.

When two quarks become separated, as happens in particle accelerator collisions, at some point it is more energetically favorable for a new quark–antiquark pair to spontaneously appear, than to allow the tube to extend further. As a result of this, when quarks are produced in particle accelerators, instead of seeing the individual quarks in detectors, scientists see "jets" of many color-neutral particles (mesons and baryons), clustered together. This process is called hadronization, fragmentation, or string breaking, and is one of the least understood processes in particle physics.


So hypothetical New Physics may give us a way to isolate particles that aren't color neutral, but IMHO it's much more likely that theoretical advances will give us an analytic proof of color confinement.

J Thomas wrote:So it makes a certain sense to figure that there are a finite number of point particles zooming around inside the vast empty spaces inside each nucleus,


Not really. Nucleons are certainly not point particles, and are relatively large compared to the size of the nucleus. OTOH, the quarks that compose nucleons are point particles, but I wouldn't call a nucleon mostly empty space, since there are lots of gluons and various virtual particles in there, too.

Bear in mind that fundamental particles aren't the ultimate entities that they were once considered to be, they're really just quantized excitations of the various fundamental quantum fields that fill space. Each of these fields has various kinds of symmetry associated with it; these symmetries are ultimately the basis of the properties of the field, and are intimately related to the various conservation laws.
Quantum field theory wrote:In QFT, photons are not thought of as "little billiard balls" but are rather viewed as field quanta – necessarily chunked ripples in a field, or "excitations", that "look like" particles. Fermions, like the electron, can also be described as ripples/excitations in a field, where each kind of fermion has its own field. In summary, the classical visualisation of "everything is particles and fields", in quantum field theory, resolves into "everything is particles", which then resolves into "everything is fields". In the end, particles are regarded as excited states of a field (field quanta).

The particle approach allows us to simplify the analysis of quantum behaviour, but particles are just an approximation. Often, they are a very good approximation (eg in QED) but like all simplifying models they do have their limitations.

Anyway, I wanted to say a few word regarding neutrino production. Here's the basic Feynmann diagram of the decay of a free neutron into a proton, an electron and an antineutrino.
Image
One of the down quarks in the neutron converts into an up quark via the weak interaction, and a W- boson is released, which quickly decays into an electron and an antineutrino. The kinetic energy released in this process is around 0.7825MeV. Note that neutrinos are leptons, so lepton number is conserved. Also note that the boson's path through spacetime is not actually a wiggly line,it's just a convention of Feynmann diagrams to represent boson paths by wiggly lines. :)

At the quantum level, geometry is a bit weird due to the position-momentum uncertainty relation, but we can still use classical rules to get an idea of the probable paths of particles. So the deflection given to the neutron as it transmutes into a proton will be in the opposite direction to the path of the W boson, in accordance with conservation of momentum. The actual direction in which the boson is emitted is pretty-well random (although it may be influenced by the local magnetic field, IIRC).

The W boson has almost 100 times the mass of a neutron or proton, so conservation of momentum tells us that the speed given to the proton will be roughly 100 times that of the boson. The two leptons will be emitted roughly in the direction that the boson was emitted, but the angle between their paths varies in the usual random quantum fashion.

Another way to make neutrinos would be via spontaneous production of a neutrino - antineutrino pair, but I don't know how you'd easily induce such pair production, and anyway, I expect it'd be useless for a rocket engine without the "magic" neutrino mirror, since the two particles would be emitted with equal and opposite momenta.

Why is a neutrino mirror magical? Well, reflection requires the reflected particles to interact with the reflector, but neutrinos are rather reticent when it comes to interactions. I guess you could deflect them gravitationally, if you had a convenient black hole, but if you've got a tame black hole there are more efficient ways of using it as a rocket engine. :)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Wed Sep 05, 2012 1:07 pm UTC

Pfhorrest wrote:
J Thomas wrote:Conservation of charge. If a nucleus ejects a positive charge, then it has one less positive charge. If it ejects a negative charge the it has one more positive charge. If it ejects an uncharged particle then its charge is unchanged.

I think you really do understand this, but the way you're saying it is completely incorrect, because there is no such thing as "a charge". A particle has charge, and that charge can possibly have a range of values besides 1 or -1, but even in that case it doesn't have charges.


"Give me a hit."
"Hey, man. You're sayin it wrong. You can't have a hit without a bong, or a toke or somethin. We only have a bong though. If you talk like you want a hit all by itself, people will think you don't understand."
"Yeah, OK. Whatever you say. But give me a hit."
"You gotta say it the right way or people will think your brain is fried. And you can't have a hit without a bong. A hit is a property of a bong. No wait, no a property, it's a bong method. So you have class bong, and then you have this particular instance bong, and you wanna pass a message to the bong object...."
"Gimmee."
<passes the bong>

PM 2Ring wrote:
J Thomas wrote:
Well, you can study protons and neutrons, but the particles they are composed of can't be isolated.


Yet.


Well, people tried for a long time to isolate quarks, in order to prove that the quark theory was valid but they failed, and that failure almost killed quark theory.


This is one of the things I don't understand about physics. They saw a pattern of behaviors, and they decided it implied new particles to make that pattern happen. Then when they couldn't isolate the particles they predicted, it almost killed the theory. But instead they decided the particles really are there but they're special particles that can never be isolated. Whenever you *do* isolate the special particle it quick rips new particles out of thin air to keep itself from being isolated. Literally. (Vacuum is very thin air.)
http://en.wikipedia.org/wiki/Hadronization
But isn't the observed pattern the important thing?

Bear in mind that fundamental particles aren't the ultimate entities that they were once considered to be, they're really just quantized excitations of the various fundamental quantum fields that fill space. Each of these fields has various kinds of symmetry associated with it; these symmetries are ultimately the basis of the properties of the field, and are intimately related to the various conservation laws.


Maybe this stuff would be easier to follow if you guys stopped calling them "particles". The word has a whole lot of baggage that (it looks like) no longer applies.

Why is a neutrino mirror magical? Well, reflection requires the reflected particles to interact with the reflector, but neutrinos are rather reticent when it comes to interactions. I guess you could deflect them gravitationally, if you had a convenient black hole, but if you've got a tame black hole there are more efficient ways of using it as a rocket engine. :)


Sure. It would take something that we don't know about yet. But suppose we were having this sort of discussion in 1962, just 50 years ago. Imagine that I proposed some sort of space travel that used particles which have fractional charge. Everybody would be jumping all over me because I didn't know that there is no such thing as fractional charge, charges are always integers. Every particle has either no charge, or a single positive charge, or a single negative charge. Or composite particles like alpha particles can have some multiple of integer charges. Everybody who knew anything about physics knew this. It wasn't until 1964 that reputable physicists suggested particles with fractional charge, and there was considerable resistance to the idea particularly when no such particles could be found.

I certainly don't want to imply that I think I'm right and every physicist is wrong, that I know something special which says someday neutrinos will be useful for propulsion in space. I'm only making the rather trite point that nobody knows very much about physics yet. We know a lot of things we can use to get predictable results in narrowly-defined circumstances. We don't know a lot about what's always impossible versus what may be possible sometimes.
The Law of Fives is true. I see it everywhere I look for it.

KrytenKoro
Posts: 1487
Joined: Tue Apr 05, 2011 2:58 pm UTC

Re: What-if 0007: Everybody Out

Postby KrytenKoro » Wed Sep 05, 2012 3:06 pm UTC

J Thomas wrote:"Give me a hit."
"Hey, man. You're sayin it wrong. You can't have a hit without a bong, or a toke or somethin. We only have a bong though. If you talk like you want a hit all by itself, people will think you don't understand."
"Yeah, OK. Whatever you say. But give me a hit."
"You gotta say it the right way or people will think your brain is fried. And you can't have a hit without a bong. A hit is a property of a bong. No wait, no a property, it's a bong method. So you have class bong, and then you have this particular instance bong, and you wanna pass a message to the bong object...."
"Gimmee."
<passes the bong>

Are you suggesting that physics terminology should be exactly as rigorous as toking requests? Really?

I certainly don't want to imply that I think I'm right and every physicist is wrong, that I know something special which says someday neutrinos will be useful for propulsion in space. I'm only making the rather trite point that nobody knows very much about physics yet. We know a lot of things we can use to get predictable results in narrowly-defined circumstances. We don't know a lot about what's always impossible versus what may be possible sometimes.

No, you're making the entirely fallacious claim that, because we've obtained paradigm-shifting insights in the past, we will necessarily obtain them in the future, and then going onto say that, therefore, our current understanding of physics must be disregarded as no more reliable than a fanciful story involving subatomic fairies clearly contradicting all previous observations of how physics works.

It's fine to say "It's possible we don't know everything yet." It is NOT fine to say, "We're definitely almost completely and hopelessly wrong, and should therefore sit on our asses until leprechauns show up to sort this all out for us."
From the elegant yelling of this compelling dispute comes the ghastly suspicion my opposition's a fruit.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Wed Sep 05, 2012 4:20 pm UTC

KrytenKoro wrote:
J Thomas wrote:"Give me a hit."
"Hey, man. You're sayin it wrong. You can't have a hit without a bong, or a toke or somethin. We only have a bong though. If you talk like you want a hit all by itself, people will think you don't understand."
"Yeah, OK. Whatever you say. But give me a hit."
"You gotta say it the right way or people will think your brain is fried. And you can't have a hit without a bong. A hit is a property of a bong. No wait, no a property, it's a bong method. So you have class bong, and then you have this particular instance bong, and you wanna pass a message to the bong object...."
"Gimmee."
<passes the bong>

Are you suggesting that physics terminology should be exactly as rigorous as toking requests? Really?


No, of course I'm not.

I certainly don't want to imply that I think I'm right and every physicist is wrong, that I know something special which says someday neutrinos will be useful for propulsion in space. I'm only making the rather trite point that nobody knows very much about physics yet. We know a lot of things we can use to get predictable results in narrowly-defined circumstances. We don't know a lot about what's always impossible versus what may be possible sometimes.

No, you're making the entirely fallacious claim that, because we've obtained paradigm-shifting insights in the past, we will necessarily obtain them in the future, and then going onto say that, therefore, our current understanding of physics must be disregarded as no more reliable than a fanciful story involving subatomic fairies clearly contradicting all previous observations of how physics works.


No, I'm not saying that either. It's always possible that we've already had all the paradigm-shifting insights in physics we're ever going to get. How would we estimate how likely that is? Can you think of a way? Perhaps we could look at history. Get some way to decide what's a paradigm-shifting insight, and then look at how often they have happened. If the rate has slowed down over time, maybe by now it's stopped.

It's fine to say "It's possible we don't know everything yet." It is NOT fine to say, "We're definitely almost completely and hopelessly wrong, and should therefore sit on our asses until leprechauns show up to sort this all out for us."


I'm not saying that either. I'm saying that this is probably not the best time to start a multi-quadrillion dollar engineering project which might pay off in a very long time, on the assumption that we will not find better ways.

If the USA had tried to get a network of canals that went through the mountains because we needed transport and we didn't expect ever to have better technology, we would have wasted a lot of resources. When the railroads came in a lot of canals got discarded.

So investments in space launch need to get amortized over a couple of decades at most. If they don't pay off quick they'll probably never pay off.
The Law of Fives is true. I see it everywhere I look for it.

KrytenKoro
Posts: 1487
Joined: Tue Apr 05, 2011 2:58 pm UTC

Re: What-if 0007: Everybody Out

Postby KrytenKoro » Wed Sep 05, 2012 4:40 pm UTC

J Thomas wrote:No, of course I'm not.

Then why did you invite the interpretation by comparing a request for you to use correct physics terminology to a silly request to use correct toking terminology?

No, I'm not saying that either. It's always possible that we've already had all the paradigm-shifting insights in physics we're ever going to get. How would we estimate how likely that is? Can you think of a way? Perhaps we could look at history. Get some way to decide what's a paradigm-shifting insight, and then look at how often they have happened. If the rate has slowed down over time, maybe by now it's stopped.

...why would our rate of learning that we don't fully understand reality have any actual correlation with whether we understand it now?

The argument you're making is analagous to a child calling the sky "floyd". The parent says, no, honey, actually the sky is blue. According to you, the child should expect this answer to also be overturned, and should treat the belief that the sky is blue with exactly as much accuracy as the belief that the sky is floyd.

It's fine to say "It's possible we don't know everything yet." It is NOT fine to say, "We're definitely almost completely and hopelessly wrong, and should therefore sit on our asses until leprechauns show up to sort this all out for us."


I'm not saying that either. I'm saying that this is probably not the best time to start a multi-quadrillion dollar engineering project which might pay off in a very long time, on the assumption that we will not find better ways.

And everyone is trying to inform you that:
1) We have every reason to believe it will pay off in the very short term.
2) We will not find better ways without the research and engineering projects that you are suggesting we shouldn't do.

If the USA had tried to get a network of canals that went through the mountains because we needed transport and we didn't expect ever to have better technology, we would have wasted a lot of resources. When the railroads came in a lot of canals got discarded.

And if the USA had never built canals at all, it's very likely it would never have had the resources to develop the railroads.

As an analogy: if you want to buy a computer, but keep putting it off until the next day because the price might go down further, eventually the store will just decide "well, no one's ever going to buy it" and throw it out. You can't discard a clear route to progress based on groundless guesses that you might be able to save money if you just keep putting it off. If you do have a very good reason to believe you'll save money by waiting slightly longer (for instance:

someone is suggesting that we use normal rocket fuel to launch an experimental satellite into orbit right now, but you know of an electric or nuclear rocket engine that is close to being released which studies have shown is much more efficient than normal fuels, and there is a much smaller penalty to putting off the launch until it is released than to using the normal rocket fuel

PS. This is like basic engineering here, this is what we do. It's all about optimization) then by all means put it off until the next day. But to put it off indefinitely based on the faint hope that a leprechaun will show up with the warp drive from Star Trek is ridiculous.

So investments in space launch need to get amortized over a couple of decades at most. If they don't pay off quick they'll probably never pay off.

Yeah, this is just not accurate to how engineering works on any level, and is ignorant about the quickly-impending benefits we already have good reason to believe we'll obtain.

Even for mundane helicopters and planes, this view of "RESULTS NOW OR THE TECHNOLOGY DIES!" is not accurate.
From the elegant yelling of this compelling dispute comes the ghastly suspicion my opposition's a fruit.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Wed Sep 05, 2012 5:54 pm UTC

J Thomas wrote:So -- relativity is much more correct than the newtonian approach, it does affect GPS, and GPS has been redesigned to take relativity into account. However GPS worked adequately before they corrected for relativity.


No, GPS was designed with relativity taken into account. Whoever told you otherwise was lying.

It was based on an experiment proposed to test relativity, later combined with the observation that the position of Sputnik could be determined, and thus led to wondering if one could use a satellite to determine one's own location.

Here is an interesting link.
http://www.ipgp.fr/~tarantola/Files/Pro ... S/GPS.html
They claim that when GPS was redesigned to correct for relativity, it was done as a galilean system with relativistic corrections. So for example they attempt to set up a single universal time, and "correct" the various moving clocks to fit that standard. The idea is that it would work better to actually do GPS with relativistic math. Each moving clock has its own time, and at any time has its own location, and you get a simpler and better system if you just use it the way it works, than if you pretend you have a newtonian system and add in relativistic fudge factors at critical points.

That link says no such thing.

It says that the clocks were set up with relativistic effects taken into account, but they did not know exactly how much correction was needed so they included the ability to adjust it while in flight for the first gen satellites, later satellites have an improved method of handling relativistic corrections.

Well, you can study protons and neutrons, but the particles they are composed of can't be isolated.


Yet.

As was covered before I posted, look into color charge confinement.

They aren't really wild extrapolations though.


They are reasonable interpretations of the limited data available. But if you make sure that your mental model doesn't go much beyond the data, then you wind up with a mental model that's mostly mush.

That's a fault of your model, not mine.

No, the math for point particles introduces numerous issues, I don't even know where to begin listing them. Your information is several decades out of date if you think point particles are still considered reasonable approximations.


My information is spotty, some recent and some old, I've tended to look for things that make sense in my own context first. It looks to me like the basic idea of a shell model implies that particles inside a nucleus are so small that they can spend a reasonably long time between interactions. Shell models seem to get a lot of attention today. Would you settle for "very small"?

What is this, Get Smart?

Shell models aren't very good, nor is size particularly useful when discussing particle physics, and if one wishes to use size they should use interaction cross-section, as it's the only experimental measurement remotely corresponding to particle sizes we have.

Experimentally tested means it made predictions which were confirmed AND is compatible with prior data.


That is arbitrary. To meet that standard a theory must be proposed at the right time, when enough of its predictions have been confirmed for it o be interesting, but when important predictions have not been tested yet. Then it must attract enough attention from experimentalists that they test its remaining predictions. You're deciding which ideas to pay attention to, based entirely on historical accident. Better to tentatively accept every hypothesis you can imagine, that is not yet disproved.

There is nothing whatsoever arbitrary about that, and theories aren't proposed, hypotheses are.

Nothing should be accepted without a preponderance of evidence suggesting you can not doubt it.

You are not qualified to explain my state of mind or reasons for considering certain ideas worth paying attention to.

Not all hypothetical models are testable, certainly many are untestable without high energy experiments.


If you can't find a testable difference between two hypotheses, then there is no practical difference between them.

>.>

You're kidding, right?

So you'd place say, Minimally Supersymmetric Standard Models on the same level as Magical Fairy Neutrino Fart Mirror Models?

There are blobs of stuff which correspond to quarks glued into proton-shaped bundles packed among neutron-shaped bundles with fuzzy clouds where electrons are likely to be found all around them.


Ah. Yes. As I was saying, when you make sure your mental model doesn't go beyond the data, it turns to mush.

There's nothing mush about it, just because you want the universe to be precise and sharply in focus doesn't matter if there is no reason to think it is actually so.

Sufficiently developed models can produce very specific predictions of what should be found and in what quantities those results should be found in. The recent findings at the LHC of particles which correspond to Higgs decay events are further evidence that the standard model of particle physics accurately represents the real world.


Or rather, so far it has been is compatible with data from the real world.

You included unnecessary words for no reason, I fixed the error.

I am saying that there is no interpretation which proposes such a thing and has made testable predictions which have been experimentally confirmed, so there is little benefit in examining such an interpretation.


How will you know whether there's value in such an interpretation until you see how such an interpretation works and what it predicts? It looks to me like you are assuming your conclusion.

Somewhat ironic given you stated you "look for things that make sense in my own context first", but no, I am assuming that since such models would have to reproduce prior experimental results, and there is no simple way to do so, there is no reason to think such an interpretation exists and is useful.

I have no idea how to do it. I'm using it as an example of something that might someday be found, which I hope is compatible with what we know now but goes beyond what we know now.

I get that, I'm saying that the "might" part you included is unfounded, as it is based on hope, which is worthless in regards to science.

The probability of the quarks in a proton tunneling into a different state (and collapsing into a black hole which then explodes) is low enough that it hasn't happened with any detectable rate in the last 13 billion years, and given the bit about supporting mass I mentioned, it shouldn't happen in anything less than 1040 years, as I recall.


I'm happy that black holes are relatively rare.

They aren't, but protons collapsing into black holes is so rare that it is unlikely for it to have happened in the history of the universe.

I'm saying that if we don't truly understand the laws of physics yet, then we are likely to find things we don't expect. We know some things not to expect. Most conservation laws will stay true. You can't just suck energy out of nowhere, you need a place for it to come from, and when it's gone from there it's gone. Beyond that, your common sense is utterly unreliable at telling you what unexpected results you should expect not to happen.

I don't rely on common sense in this case, I rely on accumulated knowledge, experimental evidence, and understanding.

Conservation of charge. If a nucleus ejects a positive charge, then it has one less positive charge. If it ejects a negative charge the it has one more positive charge. If it ejects an uncharged particle then its charge is unchanged.

A nucleus ejects a positively charged particle, or negatively charged particle, the definitions are very important.

?? if it emits something that has mass then of course its mass will be changed. If it emits something that has spin then its spin will be changed. How could it be otherwise? Why would you expect me to imagine it otherwise? Well, possibly we might find that some conservation law really does get broken. I can't say we will never invent a conservation law that isn't true. But I'm real skeptical about breaking conservation laws.

A neutrino has mass and spin, you don't seem to mind violation of conservation laws when arguing in favor of present ignorance and future magic mirrors.

Interesting! I tried to look for your data and found this:
http://www.nsf.gov/statistics/nsf12318/pdf/tab19.pdf

For 2009 they give around double your 2008 numbers. I haven't found why they're so different and to me the differences tend to discredit both of them. But not enough for me to argue I could be right. I'm very happy to be wrong about this.

The 2008 numbers were just the most recent of the handful that showed up (the others were 2005, 2003, etc), I said nothing about it being the same for 2009, and there is no discrediting done there. Funding is not absolutely the same per year.
mu

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Wed Sep 05, 2012 7:56 pm UTC

KrytenKoro wrote:
J Thomas wrote:No, of course I'm not.

Then why did you invite the interpretation by comparing a request for you to use correct physics terminology to a silly request to use correct toking terminology?


This is XKCD, not a refereed journal. I was reasonably sure that everybody would know what I meant, except for Max who I should have expected not to get it because I'm an ableist asshole.

No, I'm not saying that either. It's always possible that we've already had all the paradigm-shifting insights in physics we're ever going to get. How would we estimate how likely that is? Can you think of a way? Perhaps we could look at history. Get some way to decide what's a paradigm-shifting insight, and then look at how often they have happened. If the rate has slowed down over time, maybe by now it's stopped.

...why would our rate of learning that we don't fully understand reality have any actual correlation with whether we understand it now?


It's a statistical thing. If in 1800 there were N paradigm shifts available that would be improvements, and since that time we have gotten them at an average rate of .05/year, then maybe we'll get them all in around 20N year. But maybe as we get more and more of them, the rate will slow down. When there are 20 different great ideas available, people are more likely to find one of them than when there is only one available. And also, we might tend to find the easiest ones first. Maybe the remaining concepts that will improve our thinking are *hard* so we won't get them as fast. So if we were finding them at the rate of say 1 per year 200 years ago, but in the last 100 years we only found 1, maybe it's getting so hard to find useful new ideas that we won't get any in the next 200 years even if there are a few left to find.

So the longer we go without a paradigm shift, the more likely that we understand physics perfectly already and there just aren't any improvements left for us to find. If the MTBF is increasing, that's a positive sign, right?

I'm not sure this is the best approach to estimate the likelihood that we won't find any more revolutionary ideas, it was just the one I thought of. Do you have a better idea how to estimate whether we have everything right now, and there will not be any more big changes in physics? You seem to be saying you think this is likely. I agree that I haven't proven it can't be true.

The argument you're making is analagous to a child calling the sky "floyd". The parent says, no, honey, actually the sky is blue. According to you, the child should expect this answer to also be overturned, and should treat the belief that the sky is blue with exactly as much accuracy as the belief that the sky is floyd.


No, that's very different. Once you learn what the socially-accepted name for something is, you know the name. It will stay that way until the society starts using a different name, and you'll probably find out pretty quick when that happens. Names for things aren't right or wrong, they're just what people use. It might be that physics will be like that and we will stop changing paradigms because physicists are just ready to stop. I hope not.

Now that I think about it, I'm not sure you'll find out quickly when a social consensus happens. I've been calling for english language spelling reform for a long time, and I didn't notice that we are in fact getting widespread spelling reform. I simply did not notice that txtspeak is in fact spelling reform which is happening around me. B4 U sy no, it !care wt U thnk. hpn NyWA. I wanted the spelling reform I wanted, so I didn't notice the spelling reform that is actually taking over.

It's fine to say "It's possible we don't know everything yet." It is NOT fine to say, "We're definitely almost completely and hopelessly wrong, and should therefore sit on our asses until leprechauns show up to sort this all out for us."


I'm not saying that either. I'm saying that this is probably not the best time to start a multi-quadrillion dollar engineering project which might pay off in a very long time, on the assumption that we will not find better ways.

And everyone is trying to inform you that:
1) We have every reason to believe it will pay off in the very short term.


That's worth careful consideration. I haven't heard anybody actually present an example of a way it can pay off in the short term. I've heard a few pipe dreams, like maybe it will bring in lots of ores from the asteroid belt, or if we get a space culture going that can survive independently then we can survive an extinction-event asteroid that hits earth, or maybe we can build communications satellites in space and place them where they're supposed to go cheaper than we can do that from sealevel. Maybe we can find lots of rare earths that will be incredibly valuable on earth. Maybe we can build terawatt solar power generators and beam terawatts to the earth's surface. Was there another one I missed?

2) We will not find better ways without the research and engineering projects that you are suggesting we shouldn't do.


That doesn't follow. When we do research for space we get lots of spin-offs that help the civilian economy. Space blankets. Space pens that can write upside down. Ablative shielding. Lots of things that don't get advertised. Conversely, when we do R&D here we get lots of spin-offs that will help the space program. The longer we put it off, the cheaper it gets. On the other hand, the research results we get from space might be very useful immediately. The sooner we find out about the available ores on luna and mars etc, the better. Well, maybe not that so much, for that reason, but other things. Since it's things we don't know already, we don't know how important they will be or which ones will be important. So we definitely need space research, and it looks very valuable to look for cheaper ways to get to space. But when it comes to building extremely expensive systems to put large amounts of material in orbit, hold off on that.

If the USA had tried to get a network of canals that went through the mountains because we needed transport and we didn't expect ever to have better technology, we would have wasted a lot of resources. When the railroads came in a lot of canals got discarded.

And if the USA had never built canals at all, it's very likely it would never have had the resources to develop the railroads.


We were a nation with seaports, and small cities on navigable rivers. Places we could build cheap canals flourished by it. Expensive canals not so much. When the railroads came in even the canals which had been built cheaply and which had already had a long profitable run tended to shut down. For many purposes railroads were better. But not all -- some canals could carry massive loads for less fuel. The Erie canal successfully competed with railroads and continues to serve vacationers to this day.

As an analogy: if you want to buy a computer, but keep putting it off until the next day because the price might go down further, eventually the store will just decide "well, no one's ever going to buy it" and throw it out. You can't discard a clear route to progress based on groundless guesses that you might be able to save money if you just keep putting it off.


Sure, and when you have limited funds you do better to buy more of what you can afford and less of what you cannot. Buy the expensive stuff you don't actually need later, either when you are richer or when its prices have come down.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Wed Sep 05, 2012 8:24 pm UTC

J Thomas wrote:
KrytenKoro wrote:
J Thomas wrote:No, of course I'm not.

Then why did you invite the interpretation by comparing a request for you to use correct physics terminology to a silly request to use correct toking terminology?


This is XKCD, not a refereed journal. I was reasonably sure that everybody would know what I meant, except for Max who I should have expected not to get it because he has aspergers.

Image
mu

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26836
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: What-if 0007: Everybody Out

Postby gmalivuk » Wed Sep 05, 2012 8:33 pm UTC

J Thomas wrote:
KrytenKoro wrote:
J Thomas wrote:No, of course I'm not.
Then why did you invite the interpretation by comparing a request for you to use correct physics terminology to a silly request to use correct toking terminology?
This is XKCD, not a refereed journal. I was reasonably sure that everybody would know what I meant, except for Max who I should have expected not to get it because he has aspergers.
If you ever pull this kind of shit in one of the forums I moderate, it will be the last post you make there. As it is, I reported this post, so we'll see what the local magistrate decides to do about it.

And in addition to being a pretty shitty thing to say on its face, you're also wrong about Max's mental condition being even a little bit relevant to this. I definitely don't have Asperger's, and I still have no fucking clue what you thought you were getting at with the bong-hitting reference, except perhaps to explain to the rest of us what your mental state is as you post here.

Max's point remains that a nucleus doesn't eject "a charge", it ejects or absorbs a particle, which particle then may or may not have or carry charge. And whether its net charge goes more negative because it ejected a positively charged particle or absorbed a negatively charged particle definitely matters, at least as far as real actual physics is concerned.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Wed Sep 05, 2012 9:06 pm UTC

To be fair, I was amused at my own not-getting-it, hence the "KHAN!" response.
mu

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Wed Sep 05, 2012 9:24 pm UTC

gmalivuk wrote:
J Thomas wrote: This is XKCD, not a refereed journal. I was reasonably sure that everybody would know what I meant, except for Max who I should have expected not to get it because he has aspergers.


If you ever pull this kind of shit in one of the forums I moderate, it will be the last post you make there. As it is, I reported this post, so we'll see what the local magistrate decides to do about it.

And in addition to being a pretty shitty thing to say on its face, you're also wrong about Max's mental condition being even a little bit relevant to this. I definitely don't have Asperger's, and I still have no fucking clue what you thought you were getting at with the bong-hitting reference, except perhaps to explain to the rest of us what your mental state is as you post here.


At first I thought you were upset that I would not treat XKCD like a refereed journal. Then I figured you were bothered that I mentioned Max's Aspergers. Did it seem to you that I was disparaging him? That I intended to insult him? But Max uses Aspergers as a central part of his self-concept. It is not an insult. He doesn't think the same way other people do, and that's fine. He doesn't think worse, only different. It's just something to take into account.

Max™ wrote:
J Thomas wrote:So -- relativity is much more correct than the newtonian approach, it does affect GPS, and GPS has been redesigned to take relativity into account. However GPS worked adequately before they corrected for relativity.


No, GPS was designed with relativity taken into account. Whoever told you otherwise was lying.


I provided a link, from a US government publication.

It was based on an experiment proposed to test relativity, later combined with the observation that the position of Sputnik could be determined, and thus led to wondering if one could use a satellite to determine one's own location.


Yes. And they did not need to take relativity into account to make it work, and they in fact did not, until later they did.

Here is an interesting link.
http://www.ipgp.fr/~tarantola/Files/Pro ... S/GPS.html
They claim that when GPS was redesigned to correct for relativity, it was done as a galilean system with relativistic corrections. So for example they attempt to set up a single universal time, and "correct" the various moving clocks to fit that standard. The idea is that it would work better to actually do GPS with relativistic math. Each moving clock has its own time, and at any time has its own location, and you get a simpler and better system if you just use it the way it works, than if you pretend you have a newtonian system and add in relativistic fudge factors at critical points.

That link says no such thing.

It says that the clocks were set up with relativistic effects taken into account, but they did not know exactly how much correction was needed so they included the ability to adjust it while in flight for the first gen satellites, later satellites have an improved method of handling relativistic corrections.


The link you quoted says nothing like that, it says what I said.
The other link, which you did not quote, was describing how to change the GPS system which did not take relativity into account into one that would. I am bewildered that we could read the same text and get such different meanings.

Well, you can study protons and neutrons, but the particles they are composed of can't be isolated.


Yet.

As was covered before I posted, look into color charge confinement.


Yes? They looked hard for isolated quarks and could not find any. So then they invented color charge confinement to explain that result. If at some time they do find isolated quarks then color charge confinement will be abandoned. These are basic concepts, I don't understand why you have trouble with them.

They aren't really wild extrapolations though.


They are reasonable interpretations of the limited data available. But if you make sure that your mental model doesn't go much beyond the data, then you wind up with a mental model that's mostly mush.

That's a fault of your model, not mine.


The data is not particularly definitive. If you are careful not to go beyond it, you will have a vague model. If your model does not have that flaw then it goes beyond the data. If you disagree, send me your model and I'll try to look at it.

No, the math for point particles introduces numerous issues, I don't even know where to begin listing them. Your information is several decades out of date if you think point particles are still considered reasonable approximations.


My information is spotty, some recent and some old, I've tended to look for things that make sense in my own context first. It looks to me like the basic idea of a shell model implies that particles inside a nucleus are so small that they can spend a reasonably long time between interactions. Shell models seem to get a lot of attention today. Would you settle for "very small"?

What is this, Get Smart?

Shell models aren't very good, nor is size particularly useful when discussing particle physics, and if one wishes to use size they should use interaction cross-section, as it's the only experimental measurement remotely corresponding to particle sizes we have.


OK, give me keywords for the sort of model you prefer and I'll give it a look.

Experimentally tested means it made predictions which were confirmed AND is compatible with prior data.


That is arbitrary. To meet that standard a theory must be proposed at the right time, when enough of its predictions have been confirmed for it o be interesting, but when important predictions have not been tested yet. Then it must attract enough attention from experimentalists that they test its remaining predictions. You're deciding which ideas to pay attention to, based entirely on historical accident. Better to tentatively accept every hypothesis you can imagine, that is not yet disproved.

There is nothing whatsoever arbitrary about that, and theories aren't proposed, hypotheses are.

Nothing should be accepted without a preponderance of evidence suggesting you can not doubt it.


Nothing should be rejected until you have firm evidence that's incompatible with it. Maybe this is just a difference in attitude. You want to find something definite to believe in, and try not to think too much about what's still unknown, while I prefer to look for mistakes and hidden assumptions and try to guess where interesting possibilities might show up.

Not all hypothetical models are testable, certainly many are untestable without high energy experiments.


If you can't find a testable difference between two hypotheses, then there is no practical difference between them.

>.>

You're kidding, right?

So you'd place say, Minimally Supersymmetric Standard Models on the same level as Magical Fairy Neutrino Fart Mirror Models?


If they make the same testable predictions, sure. My guess is that magical fairy neutrino fart mirror models will not provide very specific predictions and so don't deserve the same level of consideration. But if one gives the same predictions then it's the same model. If it gives different predictions but only ones that haven't been tested yet, then we can decide which to throw out when we do the new tests that one of them fails. (Of course, it's possible with new tests they might both fail. Unless the tests are carefully designed to make sure only one of them can fail at a time.)

There are blobs of stuff which correspond to quarks glued into proton-shaped bundles packed among neutron-shaped bundles with fuzzy clouds where electrons are likely to be found all around them.


Ah. Yes. As I was saying, when you make sure your mental model doesn't go beyond the data, it turns to mush.

There's nothing mush about it, just because you want the universe to be precise and sharply in focus doesn't matter if there is no reason to think it is actually so.


Yes. If your data is mushy, models which fit the data with nothing extra will also be mushy. On the other hand you might get crispy models which assume things that are not in evidence, and those models might inspire tests which they might then fail.

Sufficiently developed models can produce very specific predictions of what should be found and in what quantities those results should be found in. The recent findings at the LHC of particles which correspond to Higgs decay events are further evidence that the standard model of particle physics accurately represents the real world.


Or rather, so far it has been is compatible with data from the real world.

You included unnecessary words for no reason, I fixed the error.


You avoided an important meaning, because you did not understand it and did not understand the reason.

Imagine that we had this conversation 50 years ago. The details of the physics would be very different. But the general thrust of the arguments would be exactly the same. The physics of 50 years ago was the best fit anybody could find for the observed data. Everything you say about today's physics was true 50 years ago of that day's physics. It was compatible with the known data from the real world. But now we have more data and better data and the newer theories are compatible with that. Your conclusions are in fact independent of the particular data and the particular theories about that data. When the time comes you will trade in the old theories for new ones, like you'd give up your trusty old computer for a newer model. Not because it is any worse than it ever was, but because you can get something that's even better.

I am saying that there is no interpretation which proposes such a thing and has made testable predictions which have been experimentally confirmed, so there is little benefit in examining such an interpretation.


How will you know whether there's value in such an interpretation until you see how such an interpretation works and what it predicts? It looks to me like you are assuming your conclusion.

Somewhat ironic given you stated you "look for things that make sense in my own context first", but no, I am assuming that since such models would have to reproduce prior experimental results, and there is no simple way to do so, there is no reason to think such an interpretation exists and is useful.


If some physicist makes a new interpretation you'll look at it, right? You won't know whether there's a simple way to match previous experimental results until you actually see the theory and see how simple it is, and whether it matches previous experimental results. If you try to make such a theory yourself and you can't find a simple way to do it, that's some evidence that there's no simple way. If you find a mathematical proof that no theory which does that can be simple, that's better but the proof itself could have some flaw.

When you assume it can't be done you are assuming your conclusion.

I have no idea how to do it. I'm using it as an example of something that might someday be found, which I hope is compatible with what we know now but goes beyond what we know now.

I get that, I'm saying that the "might" part you included is unfounded, as it is based on hope, which is worthless in regards to science.


If what I describe necessarily conflicts with existing observation, then it cannot be found to be true later. If it does not necessarily conflict with what's already known, then it might or might not be discovered later.

The probability of the quarks in a proton tunneling into a different state (and collapsing into a black hole which then explodes) is low enough that it hasn't happened with any detectable rate in the last 13 billion years, and given the bit about supporting mass I mentioned, it shouldn't happen in anything less than 1040 years, as I recall.


I'm happy that black holes are relatively rare.

They aren't, but protons collapsing into black holes is so rare that it is unlikely for it to have happened in the history of the universe.


Under the experimental conditions they have been studied, and the theoretical assumptions used to generate the probabilities. Those conditions might extrapolate well to the whole universe. I can't say they definitely don't extrapolate well to the whole universe. I dunno.

I'm saying that if we don't truly understand the laws of physics yet, then we are likely to find things we don't expect. We know some things not to expect. Most conservation laws will stay true. You can't just suck energy out of nowhere, you need a place for it to come from, and when it's gone from there it's gone. Beyond that, your common sense is utterly unreliable at telling you what unexpected results you should expect not to happen.

I don't rely on common sense in this case, I rely on accumulated knowledge, experimental evidence, and understanding.


And you may not have noticed this, but you are relying on all that to predict what we will find when we look at new things we have not seen before. There's nothing wrong with extrapolating from what you know into the unknown, to guess what will happen. It's silly to have a lot of confidence in it though. "Common sense is what tells you the whole population is just like your biased sample."

Conservation of charge. If a nucleus ejects a positive charge, then it has one less positive charge. If it ejects a negative charge the it has one more positive charge. If it ejects an uncharged particle then its charge is unchanged.

A nucleus ejects a positively charged particle, or negatively charged particle, the definitions are very important.


OK, a charged particle.

?? if it emits something that has mass then of course its mass will be changed. If it emits something that has spin then its spin will be changed. How could it be otherwise? Why would you expect me to imagine it otherwise? Well, possibly we might find that some conservation law really does get broken. I can't say we will never invent a conservation law that isn't true. But I'm real skeptical about breaking conservation laws.

A neutrino has mass and spin, you don't seem to mind violation of conservation laws when arguing in favor of present ignorance and future magic mirrors.


I didn't think I was suggesting that total mass or spin would be changed. Did something I say imply that?

Interesting! I tried to look for your data and found this:
http://www.nsf.gov/statistics/nsf12318/pdf/tab19.pdf

For 2009 they give around double your 2008 numbers. I haven't found why they're so different and to me the differences tend to discredit both of them. But not enough for me to argue I could be right. I'm very happy to be wrong about this.

The 2008 numbers were just the most recent of the handful that showed up (the others were 2005, 2003, etc), I said nothing about it being the same for 2009, and there is no discrediting done there. Funding is not absolutely the same per year.


It's extremely unlikely it would change that much in one year. Many things were about doubled, some less and some more. Most likely they were counting different things. Mine probably included more funding sources and more recipients. I didn't look into it enough to really see what was going on. It's a warning that to actually use the numbers for anything important you have to look carefully at how they were measured and what they mean.

But if either version is right or close to right, I was wrong to say that physics is getting funded out of proportion. Physical sciences were funded a bit less than 7 times as much as mathematical sciences in your version, while in mine it was confused -- physical sciences get less than twice what mathematical and computer sciences do together, and a bit more than 5 times what math gets as a subset of that. And by both versions physics gets much less than biology. I was wrong. I'm glad I was wrong.
Last edited by J Thomas on Wed Sep 05, 2012 9:46 pm UTC, edited 4 times in total.
The Law of Fives is true. I see it everywhere I look for it.

KrytenKoro
Posts: 1487
Joined: Tue Apr 05, 2011 2:58 pm UTC

Re: What-if 0007: Everybody Out

Postby KrytenKoro » Wed Sep 05, 2012 9:29 pm UTC

J Thomas wrote:It's a statistical thing. If in 1800 there were N paradigm shifts available that would be improvements, and since that time we have gotten them at an average rate of .05/year, then maybe we'll get them all in around 20N year. But maybe as we get more and more of them, the rate will slow down. When there are 20 different great ideas available, people are more likely to find one of them than when there is only one available. And also, we might tend to find the easiest ones first. Maybe the remaining concepts that will improve our thinking are *hard* so we won't get them as fast. So if we were finding them at the rate of say 1 per year 200 years ago, but in the last 100 years we only found 1, maybe it's getting so hard to find useful new ideas that we won't get any in the next 200 years even if there are a few left to find.

So the longer we go without a paradigm shift, the more likely that we understand physics perfectly already and there just aren't any improvements left for us to find. If the MTBF is increasing, that's a positive sign, right?

I'm not sure this is the best approach to estimate the likelihood that we won't find any more revolutionary ideas, it was just the one I thought of. Do you have a better idea how to estimate whether we have everything right now, and there will not be any more big changes in physics? You seem to be saying you think this is likely. I agree that I haven't proven it can't be true.

My point is that approaching the accuracy of current models by analyzing how often corrections were made to past models is a fundamental misunderstanding of how truth and objective reality work. Moore's Law does not logically apply to truth.

The argument you're making is analagous to a child calling the sky "floyd". The parent says, no, honey, actually the sky is blue. According to you, the child should expect this answer to also be overturned, and should treat the belief that the sky is blue with exactly as much accuracy as the belief that the sky is floyd.


...really?

Do you really not understand the point of the analogy, or are you being deliberately obtuse? Fine, I'll (re)spell it out: Just because the child was corrected once, there is no reason to believe that any further corrections will have some kind of rate-relationship with the prior corrections. That's not how facts work. New discoveries happen when they happen, and while we can certainly encourage them to happen faster with a rational mindset...once you hit the jackpot, it's been hit.

That's worth careful consideration. I haven't heard anybody actually present an example of a way it can pay off in the short term. I've heard a few pipe dreams, like maybe it will bring in lots of ores from the asteroid belt, or if we get a space culture going that can survive independently then we can survive an extinction-event asteroid that hits earth, or maybe we can build communications satellites in space and place them where they're supposed to go cheaper than we can do that from sealevel. Maybe we can find lots of rare earths that will be incredibly valuable on earth. Maybe we can build terawatt solar power generators and beam terawatts to the earth's surface. Was there another one I missed?

Putting aside that, again, the timespan of sending probes out to the asteroids to get resources is not an extraordinary length of time in the engineering world, the research itself can provide benefits. For example, developing the carbon nanotubes to build a space elevator would teach us a lot about how the limits of manufacturing and building structures with carbon nanotubes, which we can apply to Earth issues. This is all with the benefit of providing the researchers with an obviously positive outcome of successful research (we get to go to space, yay!), which helps moral, while also being something that nations can hang some old-fashioned nationalism on to get the funding going. No, it's totally possible that there's some better way to get this whole engine funded and running, but this is still a good method.

That doesn't follow. When we do research for space we get lots of spin-offs that help the civilian economy. Space blankets. Space pens that can write upside down. Ablative shielding. Lots of things that don't get advertised. Conversely, when we do R&D here we get lots of spin-offs that will help the space program. The longer we put it off, the cheaper it gets. On the other hand, the research results we get from space might be very useful immediately. The sooner we find out about the available ores on luna and mars etc, the better. Well, maybe not that so much, for that reason, but other things. Since it's things we don't know already, we don't know how important they will be or which ones will be important. So we definitely need space research, and it looks very valuable to look for cheaper ways to get to space. But when it comes to building extremely expensive systems to put large amounts of material in orbit, hold off on that.

There's a lack of actual examples there other than the vague assertion that the flow must go both ways (although that may be true, it is not necessarily so: space travel is an exotic and possibly edge-case problem), but I think we can both agree that if your intention is to develop more advanced technology for space colonization, it is more efficient to throw the money at research into space colonization technologies, rather than throwing it at researching the social dynamics of using Twinkies as sex toys. At the very least, someone has got to do the research to make the connection at some point.

We were a nation with seaports, and small cities on navigable rivers. Places we could build cheap canals flourished by it. Expensive canals not so much. When the railroads came in even the canals which had been built cheaply and which had already had a long profitable run tended to shut down. For many purposes railroads were better. But not all -- some canals could carry massive loads for less fuel. The Erie canal successfully competed with railroads and continues to serve vacationers to this day.

This story is completely irrelevant to my most recent point on this thread, or even the original point that was (I think) something about helium, or putting research to the side.

Sure, and when you have limited funds you do better to buy more of what you can afford and less of what you cannot. Buy the expensive stuff you don't actually need later, either when you are richer or when its prices have come down.

...you really don't get analogies, it seems...fine, a direct statement: If you keep putting off purchasing something with an obvious benefit on the hope that it will be cheaper tomorrow, you will never buy that thing.

If all you require to make your decision is totally baseless wishes, you will never run out.

There's nothing wrong with extrapolating from what you know into the unknown, to guess what will happen. It's silly to have a lot of confidence in it though.

More importantly, it's intensely self-destructive to not even try to explore based on our current beliefs, in the fear that there might be a googling thundersnatch around the next bend.
From the elegant yelling of this compelling dispute comes the ghastly suspicion my opposition's a fruit.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Wed Sep 05, 2012 10:28 pm UTC

J Thomas wrote:At first I thought you were upset that I would not treat XKCD like a refereed journal. Then I figured you were bothered that I mentioned Max's Aspergers. Did it seem to you that I was disparaging him? That I intended to insult him? But Max uses Aspergers as a central part of his self-concept. It is not an insult. He doesn't think the same way other people do, and that's fine. He doesn't think worse, only different. It's just something to take into account.

I use it as an explanation because I like to be understood, but no, it didn't insult me, though I assume others may be insulted by such comments, myself I can not say I was.

Yes. And they did not need to take relativity into account to make it work, and they in fact did not, until later they did.

Like I said, the link you provided does not support this position, nor does history.

The link you quoted says nothing like that, it says what I said.
The other link, which you did not quote, was describing how to change the GPS system which did not take relativity into account into one that would. I am bewildered that we could read the same text and get such different meanings.

I was talking about the Neil Ashby paper you linked: http://www.ipgp.fr/~tarantola/Files/Pro ... ty_GPS.pdf

Yes? They looked hard for isolated quarks and could not find any. So then they invented color charge confinement to explain that result. If at some time they do find isolated quarks then color charge confinement will be abandoned. These are basic concepts, I don't understand why you have trouble with them.

I think you're getting the order backwards, and they observe hadron jets in collider experiments which should be far more than capable of isolating quarks if not for color charge confinement.

If you have a better explanation than hadron confinement, by all means, publish a paper on it.
The data is not particularly definitive. If you are careful not to go beyond it, you will have a vague model. If your model does not have that flaw then it goes beyond the data. If you disagree, send me your model and I'll try to look at it.

Things are not as vague as you present them just in the standard model, never mind any ideas I may have.

OK, give me keywords for the sort of model you prefer and I'll give it a look.

I prefer relativistic ones, first, so I'd say a relativistic mean field model would be a good starting point. The names Hartree and Fock come to mind.

Nothing should be rejected until you have firm evidence that's incompatible with it. Maybe this is just a difference in attitude. You want to find something definite to believe in, and try not to think too much about what's still unknown, while I prefer to look for mistakes and hidden assumptions and try to guess where interesting possibilities might show up.

I don't believe anything. I want to find something I can't doubt, nothing more.

If they make the same testable predictions, sure. My guess is that magical fairy neutrino fart mirror models will not provide very specific predictions and so don't deserve the same level of consideration. But if one gives the same predictions then it's the same model. If it gives different predictions but only ones that haven't been tested yet, then we can decide which to throw out when we do the new tests that one of them fails. (Of course, it's possible with new tests they might both fail. Unless the tests are carefully designed to make sure only one of them can fail at a time.)

Occam comes into play here.

Yes. If your data is mushy, models which fit the data with nothing extra will also be mushy. On the other hand you might get crispy models which assume things that are not in evidence, and those models might inspire tests which they might then fail.

Crispy models have been shown to be fairly incompatible with reality.

You avoided an important meaning, because you did not understand it and did not understand the reason.

Imagine that we had this conversation 50 years ago. The details of the physics would be very different. But the general thrust of the arguments would be exactly the same. The physics of 50 years ago was the best fit anybody could find for the observed data. Everything you say about today's physics was true 50 years ago of that day's physics. It was compatible with the known data from the real world. But now we have more data and better data and the newer theories are compatible with that. Your conclusions are in fact independent of the particular data and the particular theories about that data. When the time comes you will trade in the old theories for new ones, like you'd give up your trusty old computer for a newer model. Not because it is any worse than it ever was, but because you can get something that's even better.

The physics of 50 years ago saw the start of QED, Feynman being awesome, Relativity had long since been established, and there were signs that QED would turn into what later became QCD.

I don't buy prebuilt computers, I can make them from parts, much the same applies to theoretical models when you have experimental evidence. You don't throw away the old model and start over.

If some physicist makes a new interpretation you'll look at it, right? You won't know whether there's a simple way to match previous experimental results until you actually see the theory and see how simple it is, and whether it matches previous experimental results. If you try to make such a theory yourself and you can't find a simple way to do it, that's some evidence that there's no simple way. If you find a mathematical proof that no theory which does that can be simple, that's better but the proof itself could have some flaw.

When you assume it can't be done you are assuming your conclusion.

I will look at an interpretation which fits known physics and explains new phenomena, I've worked on this some myself, but you're not saying what I am. You're acting as though I claimed there could be no simple model, all I said was you're not going to produce a model with the same explanatory power which reproduces known models which is arbitrarily more simple. There are attempts to do so, and you know what we got? Stringy models.

Under the experimental conditions they have been studied, and the theoretical assumptions used to generate the probabilities. Those conditions might extrapolate well to the whole universe. I can't say they definitely don't extrapolate well to the whole universe. I dunno.

Well, no, we can do this because the basic observations of the rest of the universe would be very different.

You can learn a lot about fundamental physics just studying stars and galaxies.

And you may not have noticed this, but you are relying on all that to predict what we will find when we look at new things we have not seen before. There's nothing wrong with extrapolating from what you know into the unknown, to guess what will happen. It's silly to have a lot of confidence in it though. "Common sense is what tells you the whole population is just like your biased sample."

No, I'm really not doing that. If I make a hard statement it will be about things like not finding wizardry or fairy mirrors, because I operate on the assumption that the universe has certain qualities of consistency and enables reproducible experiments.

I didn't think I was suggesting that total mass or spin would be changed. Did something I say imply that?

If you emit anything, you are going to change something, if you emit a particle with mass and spin like a neutrino, and don't change them... well, magic!
mu

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Wed Sep 05, 2012 11:41 pm UTC

KrytenKoro wrote:
J Thomas wrote:It's a statistical thing. If in 1800 there were N paradigm shifts available that would be improvements, and since that time we have gotten them at an average rate of .05/year, then maybe we'll get them all in around 20N year. But maybe as we get more and more of them, the rate will slow down. When there are 20 different great ideas available, people are more likely to find one of them than when there is only one available. And also, we might tend to find the easiest ones first. Maybe the remaining concepts that will improve our thinking are *hard* so we won't get them as fast. So if we were finding them at the rate of say 1 per year 200 years ago, but in the last 100 years we only found 1, maybe it's getting so hard to find useful new ideas that we won't get any in the next 200 years even if there are a few left to find.

So the longer we go without a paradigm shift, the more likely that we understand physics perfectly already and there just aren't any improvements left for us to find. If the MTBF is increasing, that's a positive sign, right?

I'm not sure this is the best approach to estimate the likelihood that we won't find any more revolutionary ideas, it was just the one I thought of. Do you have a better idea how to estimate whether we have everything right now, and there will not be any more big changes in physics? You seem to be saying you think this is likely. I agree that I haven't proven it can't be true.

My point is that approaching the accuracy of current models by analyzing how often corrections were made to past models is a fundamental misunderstanding of how truth and objective reality work. Moore's Law does not logically apply to truth.


You seem to be saying that we should not make any model at all of the process. This is commonly done for computer program maintenance. You can measure the rate that errors are discovered, and the rate that they are "fixed". It's predictable that over time the number of errors found will decline until it approaches a minimum. At the minimum, maintainers who fix subtle problems create new subtle problems at about the same rate. So the rate of new error reports will stay near that level until it drops because the user base declines or the decision is made to end maintenance.

If it's true that physics is basicly perfect or it will be perfect after N more paradigm shifts, it's possible to make a model of the process. Of course, if it turns out that something like Godel's theorem applies and there will always be room for a new paradigm shift that will improve things, then that requires a different model.

The argument you're making is analagous to a child calling the sky "floyd". The parent says, no, honey, actually the sky is blue. According to you, the child should expect this answer to also be overturned, and should treat the belief that the sky is blue with exactly as much accuracy as the belief that the sky is floyd.


...really?

Do you really not understand the point of the analogy, or are you being deliberately obtuse? Fine, I'll (re)spell it out: Just because the child was corrected once, there is no reason to believe that any further corrections will have some kind of rate-relationship with the prior corrections. That's not how facts work. New discoveries happen when they happen, and while we can certainly encourage them to happen faster with a rational mindset...once you hit the jackpot, it's been hit.


Yes. So if there are a finite number of paradigm shifts that will bring us to the truth, that gives us one model. If the reality is different then that model will fail. Agreed, once you get a particular paradigm shift you probably won't be able to get the same one over again, that jackpot has been hit.

That's worth careful consideration. I haven't heard anybody actually present an example of a way it can pay off in the short term. I've heard a few pipe dreams, like maybe it will bring in lots of ores from the asteroid belt, or if we get a space culture going that can survive independently then we can survive an extinction-event asteroid that hits earth, or maybe we can build communications satellites in space and place them where they're supposed to go cheaper than we can do that from sealevel. Maybe we can find lots of rare earths that will be incredibly valuable on earth. Maybe we can build terawatt solar power generators and beam terawatts to the earth's surface. Was there another one I missed?

Putting aside that, again, the timespan of sending probes out to the asteroids to get resources is not an extraordinary length of time


OK! If you expect it to be profitable in a short enough timeframe, get some financing and go do it. I'll do whatever I can to keep the government from stopping you, unless you intend to Orion the world.There are a lot of businesses looking hard for profitable investments. Part of the reason interest rates are so low is that there aren't enough opportunities to make a lot of money, so businesses don't borrow to do it. If you can get profits on a reasonable timescale then you don't need to persuade the public -- people like me. Just persuade a few people who have the capital.

Show them the proof that the resources are there, and that they can be brought here quickly and cheaply enough, and they'll jump at it.

in the engineering world, the research itself can provide benefits. For example, developing the carbon nanotubes to build a space elevator would teach us a lot about how the limits of manufacturing and building structures with carbon nanotubes, which we can apply to Earth issues.


Conversely, if you develop carbon nanotubes for earth issues then you can apply them to a space elevator. It works either direction.

That doesn't follow. When we do research for space we get lots of spin-offs that help the civilian economy. Space blankets. Space pens that can write upside down. Ablative shielding. Lots of things that don't get advertised. Conversely, when we do R&D here we get lots of spin-offs that will help the space program. The longer we put it off, the cheaper it gets. On the other hand, the research results we get from space might be very useful immediately. The sooner we find out about the available ores on luna and mars etc, the better. Well, maybe not that so much, for that reason, but other things. Since it's things we don't know already, we don't know how important they will be or which ones will be important. So we definitely need space research, and it looks very valuable to look for cheaper ways to get to space. But when it comes to building extremely expensive systems to put large amounts of material in orbit, hold off on that.

There's a lack of actual examples there other than the vague assertion that the flow must go both ways (although that may be true, it is not necessarily so: space travel is an exotic and possibly edge-case problem),


What sort of examples would you like? CAD/CAM has been useful for space programs, right? But it was developed largely at automobile companies (and some aerospace) and applied at NASA later.

Would you argue that the space program has affected the internet more than the internet has affected the space program? It started out with universities and the military and such, and the space program doesn't get prominently mentioned although of course their military component participated.

Electric copiers? This was a civilian development. How much harder would the space program have had it, without them?

Electric calculators? This was a civilian development. How well did space engineers do getting more than 4 or 5 digits precision with slide rules?

Computing generally? The space program has had a big effect on rad-hard components. Not so much on other things. Compare the use of workstations and personal computers on the ground for the space program. As opposed to doing everything with hand calculators....

but I think we can both agree that if your intention is to develop more advanced technology for space colonization, it is more efficient to throw the money at research into space colonization technologies, rather than throwing it at researching the social dynamics of using Twinkies as sex toys. At the very least, someone has got to do the research to make the connection at some point.


Yes, if your intention is to get into space, then you do best to persuade the government to give you money to get into space. That's beyond question. The most efficient way to get into space is to take your place at the public trough and grab as many resources as you can get. This may not get the best outcome for your earthly nation, and it might not even be the best for you in the long run. After you are in space you might benefit from a strong friendly nation on earth, and if you suck too much from the government teat early, that nation might be too weak to help you later. But it will get you into space quicker.

We were a nation with seaports, and small cities on navigable rivers. Places we could build cheap canals flourished by it. Expensive canals not so much. When the railroads came in even the canals which had been built cheaply and which had already had a long profitable run tended to shut down. For many purposes railroads were better. But not all -- some canals could carry massive loads for less fuel. The Erie canal successfully competed with railroads and continues to serve vacationers to this day.

This story is completely irrelevant to my most recent point on this thread, or even the original point that was (I think) something about helium, or putting research to the side.


If you sink too many resources into an expensive technology which goes obsolete too fast, then you lose. Not irrelevant.

Sure, and when you have limited funds you do better to buy more of what you can afford and less of what you cannot. Buy the expensive stuff you don't actually need later, either when you are richer or when its prices have come down.

...you really don't get analogies, it seems...fine, a direct statement: If you keep putting off purchasing something with an obvious benefit on the hope that it will be cheaper tomorrow, you will never buy that thing.


I agree with that. On the other hand, if you do what you can afford before you do what you can't afford, you might be able to do the expensive stuff easier later. We don't have to put colonies in the asteroids to research propulsion systems.

There's nothing wrong with extrapolating from what you know into the unknown, to guess what will happen. It's silly to have a lot of confidence in it though.

More importantly, it's intensely self-destructive to not even try to explore based on our current beliefs, in the fear that there might be a googling thundersnatch around the next bend.


I don't get where that came from. I say, space exploration is expensive now, so let's do what we can afford and look for ways to make it cheaper. When we get it cheaper then we can afford to do more of it. When you can persuade businessmen to fund it based on expected profits then it will probably turn cheaper still.
The Law of Fives is true. I see it everywhere I look for it.

KrytenKoro
Posts: 1487
Joined: Tue Apr 05, 2011 2:58 pm UTC

Re: What-if 0007: Everybody Out

Postby KrytenKoro » Thu Sep 06, 2012 12:09 am UTC

J Thomas wrote:You seem to be saying that we should not make any model at all of the process. This is commonly done for computer program maintenance. You can measure the rate that errors are discovered, and the rate that they are "fixed". It's predictable that over time the number of errors found will decline until it approaches a minimum. At the minimum, maintainers who fix subtle problems create new subtle problems at about the same rate. So the rate of new error reports will stay near that level until it drops because the user base declines or the decision is made to end maintenance.

If it's true that physics is basicly perfect or it will be perfect after N more paradigm shifts, it's possible to make a model of the process. Of course, if it turns out that something like Godel's theorem applies and there will always be room for a new paradigm shift that will improve things, then that requires a different model.

And again, I remind you that checking reality is not the same thing as checking a program for bugs.

Yes. So if there are a finite number of paradigm shifts that will bring us to the truth, that gives us one model. If the reality is different then that model will fail. Agreed, once you get a particular paradigm shift you probably won't be able to get the same one over again, that jackpot has been hit.

There are paradigm shifts solely because the initial guesses were wrong, not because the system necessitates starting with incorrect guesses and narrowing them down. The "jackpot" I am talking about here is "truth": once you discover it, by definition you will not find anything to disprove it.

OK! If you expect it to be profitable in a short enough timeframe, get some financing and go do it. I'll do whatever I can to keep the government from stopping you, unless you intend to Orion the world.There are a lot of businesses looking hard for profitable investments. Part of the reason interest rates are so low is that there aren't enough opportunities to make a lot of money, so businesses don't borrow to do it. If you can get profits on a reasonable timescale then you don't need to persuade the public -- people like me. Just persuade a few people who have the capital.

Show them the proof that the resources are there, and that they can be brought here quickly and cheaply enough, and they'll jump at it.

Me saying on the internet that I support research into space colonization, and reporting what the facts of the situation is, does not equate to me seeking your approval for starting a space colonization corporation. I don't know how you even made that leap. I've got to ask at this point, are you just trolling?

Conversely, if you develop carbon nanotubes for earth issues then you can apply them to a space elevator. It works either direction.

...and you would still need to invest into space research to do that application. At no point have I ever suggested we funnel all money into exclusively space-based research, I'm not even sure if that's logically possible.

What sort of examples would you like? CAD/CAM has been useful for space programs, right? But it was developed largely at automobile companies (and some aerospace) and applied at NASA later.

CAD/CAM is a computer technology meant to aid drafting, not a discovery that makes new forms of structures possible to build. While CAD is certainly helpful to space travel, and this point is largely a pedantic one...your example is pedantically irrelevant.

Would you argue that the space program has affected the internet more than the internet has affected the space program? It started out with universities and the military and such, and the space program doesn't get prominently mentioned although of course their military component participated.

In that satellites are vitally necessary for the modern internet and data transfer, but the internet is not vitally necessary for spaceflight? Yes, I would.

Computing generally? The space program has had a big effect on rad-hard components. Not so much on other things. Compare the use of workstations and personal computers on the ground for the space program. As opposed to doing everything with hand calculators....

See, this is a useful example...and still it avoids the actual topic of "should we wait until later to develop the technology to physically leave Earth's surface and colonize other planets.

But since I'm bored I'll go ahead and answer for you: something like star trek/red dwarf-style nanobots would probably be a major advance in space colonization that wouldn't necessarily rely on space-based research to develop. That's still not a good reason to just give space colonization the shaft until we manage to replicate science fiction.

Yes, if your intention is to get into space, then you do best to persuade the government to give you money to get into space. That's beyond question. The most efficient way to get into space is to take your place at the public trough and grab as many resources as you can get. This may not get the best outcome for your earthly nation, and it might not even be the best for you in the long run. After you are in space you might benefit from a strong friendly nation on earth, and if you suck too much from the government teat early, that nation might be too weak to help you later. But it will get you into space quicker.

Seriously, please stop bringing Gundam-style paranoia into this.

If you sink too many resources into an expensive technology which goes obsolete too fast, then you lose. Not irrelevant.

1) You didn't demonstrate, in any way, that America "lost" more than it would by not developing canals. In fact, your admission that canals are still used demonstrates the opposite.
2) You didn't demonstrate, in any way, that this analogy actually fits your summary of it, or is at all relevant to the question of whether to finance research into space colonization. All you have, currently, is an irrelevant history blurb and an irrelevant scary soundbite.

I agree with that. On the other hand, if you do what you can afford before you do what you can't afford, you might be able to do the expensive stuff easier later. We don't have to put colonies in the asteroids to research propulsion systems.

Ohmygod then what is even the argument.

I'm saying that I support research into space colonization. If you agree that the research would be beneficial then why do you keep bringing up these ultimately irrelevant scare-stories?

I don't get where that came from. I say, space exploration is expensive now, so let's do what we can afford and look for ways to make it cheaper. When we get it cheaper then we can afford to do more of it. When you can persuade businessmen to fund it based on expected profits then it will probably turn cheaper still.

It really, really isn't that expensive compared to all the other shit we do. For example, the Curiosity launch was cheaper to pull off than people spent on watching Avatar.

And we've already persuaded businessmen to fund it based on expected profits. I'm not sure if someone spelled this out for you, or if we all just assumed that you were aware of it, but yeah: there are already companies out there researching the technology to build cheaper launch systems (for communications satellites) or deepspace exploration (for mining asteroids). As I've said multiple times, these are not thousand-year pipe dreams.
From the elegant yelling of this compelling dispute comes the ghastly suspicion my opposition's a fruit.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Thu Sep 06, 2012 5:38 pm UTC

Max™ wrote:
J Thomas wrote:Yes. And they did not need to take relativity into account to make it work, and they in fact did not, until later they did.

Like I said, the link you provided does not support this position, nor does history.

The link you quoted says nothing like that, it says what I said.
The other link, which you did not quote, was describing how to change the GPS system which did not take relativity into account into one that would. I am bewildered that we could read the same text and get such different meanings.

I was talking about the Neil Ashby paper you linked: http://www.ipgp.fr/~tarantola/Files/Pro ... ty_GPS.pdf


Ah! That paper was linked to by the one I linked to. *That's* where you got it. It seems to say at least some of what you said it did, in passing.

The one I quoted, which included that as a background link, described how the relativity patches which overlay a basicly newtonian system could be improved by using a system which was fundamentally relativistic. Like, they point out that we go to a lot of trouble trying to synchronize a bunch of moving clocks. But with relativity it isn't necessary to synchronize moving clocks and it's kind of futile. I included that link because I thought it was interesting and you might likely be interested.How to do it right with relativity.

The other one appeared to say something very different. They said that the system at that time did not use relativity and did not need to, that the errors were small. They would need to correct for those small errors to improve the positioning to get the errors smaller than a few centimeters. They explained why it was that the errors were small even though if the system had been designed differently they would be so large that the system could not function.

I am not certain though that the paper means what it flat out definitely says. Perhaps *some* relativistic corrections had been made, and they are talking about *other* relativistic corrections that didn't make much difference. Well, but he discusses each relativistic correction in terms of lorentz transforms and there are none missing that would have been already corrected. Maybe the satellites already corrected some things and he was talking about extra corrections that could be done by receivers but didn't need to be? None of the explanations I come up with quite work. But GPS is not my specialty, and I know that when you look at technical stuff that you aren't familiar with, you can miss a couple of words that change the whole meaning. See what you think. They explained how to do relativistic corrections. They explained why those corrections were not needed in practice, due to details of the way the system was set up. They recommended that those corrections be done anyway, because if the Kalman filter was set differently they *would* matter, and it might be possible to get improved accuracy beyond their current needs. Also, it would be necessary to use relativistic corrections to use GPS far from the earth's surface. What they said was perfectly clear and obvious. Maybe you will see the hidden meaning that I missed.

OK, give me keywords for the sort of model you prefer and I'll give it a look.

I prefer relativistic ones, first, so I'd say a relativistic mean field model would be a good starting point. The names Hartree and Fock come to mind.


That's the other big family of models. I'm inclined toward lattice models myself. Not that I think I understand all the problems well enough to say these are the best, but I started out thinking in those terms as a possibly fruitful path and then it turned out there were people going that way.

Nothing should be rejected until you have firm evidence that's incompatible with it. Maybe this is just a difference in attitude. You want to find something definite to believe in, and try not to think too much about what's still unknown, while I prefer to look for mistakes and hidden assumptions and try to guess where interesting possibilities might show up.

I don't believe anything. I want to find something I can't doubt, nothing more.


I want a story that makes sense. See, if all you want is something you can't doubt, just make a table of all the experimental data. There it is. It really happened and got recorded. The results are reproducible within error bounds. There is not much to doubt there, unless you choose to doubt the integrity of the people who did the experiments.

If you're willing to make a little step toward doubt, you can use statistics to clean up your data. Assume that all the little anomalies are unimportant random error -- power spikes that your power supply didn't filter out etc -- and see what you get without them. The data will look sharper and smoother and it will make a bit more sense.

If you're willing to make a big step toward doubt, you can entertain ideas about what could produce that data. You can get ideas that are compatible with the data, and then go out and collect more data and see if it's still compatible, and eventually you can gain a degree of false certainty.

I am not really a physics crank, but sometime I like to pretend I am. I'll start out with a physics idea that makes sense and then try to find out why it can't be true. I'm likely to learn things about physics that way that I'd miss otherwise. One time I tried to imagine that maybe the contents of an atomic nucleus have pretty low energy. Their energy is low enough that they're stuck together, and given a whole lot of new energy something dramatic might happen. I imagined a bunch of particles inside the nucleus buzzing around in random paths, and it didn't seem all that low-energy. If they didn't have much energy maybe they'd settle into some sort of grid, where they each have a position and wiggle around in it but mostly can't get out. Like a crystal.

But there are all those positive charged particles which push against each other -- why do they stick together at all? Some sort of strong binding energy.... I imagined the quarks of all the protons and neutrons acting independently inside a nucleus. To make it easier I imagined the 1/3 fractional charges as all independent, and thought of them as units. So an alpha particle doesn't just have 2 positive charges and 4 particles. Each proton has 4 +fracs and 1 -frac, and each neutron has 2 +fracs and 2 -fracs. So that's 12 +fracs and 6 -fracs. And in my imagination it seemed a whole lot easier to hold something together that had 6 negative charges and 12 plus charges, than something that just had 6 plus charges.

So, pretending to be a physics crank, I took it farther. I wanted to take the idea far enough to be sure it was broken and then back off. What if each proton contained not just 3 quarks, but N electrons and N+1 positrons. They could somehow act together like quarks. Neutrons might have N+2 electrons and N+2 positrons. The masses don't quite work out, but then nuclear masses don't quite work out for anybody, they come up with complicated explanations why they don't. I'll leave that step for later. So our alpha particle turns into an ionic crystal like a NaCl crystal, with 4N+4 electrons and 4N+6 positrons. Two little extra charges stuck on opposite sides of the crystal. This model doesn't need a strong binding force to hold things together, it needs a strong force to hold them apart! High-energy electrons and positrons annihilate each other producing gamma rays, while these low-energy ones are supposed to stay apart despite very strong attractive forces. I liked the idea because it turned so much upside down that would have to be carefully rethought. Probably if all the details were considered carefully enough, some version of it would work as well as current approaches. So I did a quick lit search to see whether I could find anybody reputable who'd tried anything like that.

And I found people who had done quark crystals. (Of course they don't consider anything smaller than quarks.) They point out that the nuclear magic numbers are the same as the crystal magic numbers. In important ways atomic nuclei behave exactly like crystals. Their model doesn't explain everything, but then none of the other models do either. And it makes a whole lot of sense. They claim their lattices inherit the good parts of independent particle models, liquid-drop models and cluster models.

I haven't right off noticed any significant criticism. It seems like in the short run nobody's paying attention. That happens a lot in physics.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Fri Sep 07, 2012 12:41 am UTC

J Thomas wrote:
Max™ wrote:I was talking about the Neil Ashby paper you linked: http://www.ipgp.fr/~tarantola/Files/Pro ... ty_GPS.pdf


Ah! That paper was linked to by the one I linked to. *That's* where you got it. It seems to say at least some of what you said it did, in passing.

The one I quoted, which included that as a background link, described how the relativity patches which overlay a basicly newtonian system could be improved by using a system which was fundamentally relativistic. Like, they point out that we go to a lot of trouble trying to synchronize a bunch of moving clocks. But with relativity it isn't necessary to synchronize moving clocks and it's kind of futile. I included that link because I thought it was interesting and you might likely be interested.How to do it right with relativity.

The other one appeared to say something very different. They said that the system at that time did not use relativity and did not need to, that the errors were small. They would need to correct for those small errors to improve the positioning to get the errors smaller than a few centimeters. They explained why it was that the errors were small even though if the system had been designed differently they would be so large that the system could not function.

I am not certain though that the paper means what it flat out definitely says. Perhaps *some* relativistic corrections had been made, and they are talking about *other* relativistic corrections that didn't make much difference. Well, but he discusses each relativistic correction in terms of lorentz transforms and there are none missing that would have been already corrected. Maybe the satellites already corrected some things and he was talking about extra corrections that could be done by receivers but didn't need to be? None of the explanations I come up with quite work. But GPS is not my specialty, and I know that when you look at technical stuff that you aren't familiar with, you can miss a couple of words that change the whole meaning. See what you think. They explained how to do relativistic corrections. They explained why those corrections were not needed in practice, due to details of the way the system was set up. They recommended that those corrections be done anyway, because if the Kalman filter was set differently they *would* matter, and it might be possible to get improved accuracy beyond their current needs. Also, it would be necessary to use relativistic corrections to use GPS far from the earth's surface. What they said was perfectly clear and obvious. Maybe you will see the hidden meaning that I missed.

The reason why the corrections are needed is twofold: raising something out of a gravity well means it runs faster relative to the ground, moving something rapidly relative to the ground means it runs slower.

There are additional effects which can be mostly overlooked that are due to the different altitudes and relative acceleration on each part of an orbit, though including those enabled us to do things like produce maps of mass concentrations just based on variations in GPS satellite clocks.

The clocks themselves are atomic and the variations due to different altitudes and velocities would render the entire system inaccurate by a larger and larger amount over time without relativistic corrections.

It really is impossible to make a GPS system work with just naive corrections inserted as "a fudge factor", because it's not just a one time correction when the clocks actually run at a different rate relative to the ground.

That's the other big family of models. I'm inclined toward lattice models myself. Not that I think I understand all the problems well enough to say these are the best, but I started out thinking in those terms as a possibly fruitful path and then it turned out there were people going that way.

I usually think of lattice methods as a way of mathematically approximating things, so I'm a bit uncertain what you mean there.

I want a story that makes sense. See, if all you want is something you can't doubt, just make a table of all the experimental data. There it is. It really happened and got recorded. The results are reproducible within error bounds. There is not much to doubt there, unless you choose to doubt the integrity of the people who did the experiments.

I was speaking of models that I can't doubt, obviously there won't be such a thing, so I have to look at that as an ideal to work towards.

If you're willing to make a little step toward doubt, you can use statistics to clean up your data. Assume that all the little anomalies are unimportant random error -- power spikes that your power supply didn't filter out etc -- and see what you get without them. The data will look sharper and smoother and it will make a bit more sense.

If you're willing to make a big step toward doubt, you can entertain ideas about what could produce that data. You can get ideas that are compatible with the data, and then go out and collect more data and see if it's still compatible, and eventually you can gain a degree of false certainty.

Yeah, I prefer honest doubt.

I am not really a physics crank, but sometime I like to pretend I am. I'll start out with a physics idea that makes sense and then try to find out why it can't be true. I'm likely to learn things about physics that way that I'd miss otherwise. One time I tried to imagine that maybe the contents of an atomic nucleus have pretty low energy. Their energy is low enough that they're stuck together, and given a whole lot of new energy something dramatic might happen. I imagined a bunch of particles inside the nucleus buzzing around in random paths, and it didn't seem all that low-energy. If they didn't have much energy maybe they'd settle into some sort of grid, where they each have a position and wiggle around in it but mostly can't get out. Like a crystal.

But there are all those positive charged particles which push against each other -- why do they stick together at all? Some sort of strong binding energy.... I imagined the quarks of all the protons and neutrons acting independently inside a nucleus. To make it easier I imagined the 1/3 fractional charges as all independent, and thought of them as units. So an alpha particle doesn't just have 2 positive charges and 4 particles. Each proton has 4 +fracs and 1 -frac, and each neutron has 2 +fracs and 2 -fracs. So that's 12 +fracs and 6 -fracs. And in my imagination it seemed a whole lot easier to hold something together that had 6 negative charges and 12 plus charges, than something that just had 6 plus charges.

So, pretending to be a physics crank, I took it farther. I wanted to take the idea far enough to be sure it was broken and then back off. What if each proton contained not just 3 quarks, but N electrons and N+1 positrons. They could somehow act together like quarks. Neutrons might have N+2 electrons and N+2 positrons. The masses don't quite work out, but then nuclear masses don't quite work out for anybody, they come up with complicated explanations why they don't. I'll leave that step for later. So our alpha particle turns into an ionic crystal like a NaCl crystal, with 4N+4 electrons and 4N+6 positrons. Two little extra charges stuck on opposite sides of the crystal. This model doesn't need a strong binding force to hold things together, it needs a strong force to hold them apart! High-energy electrons and positrons annihilate each other producing gamma rays, while these low-energy ones are supposed to stay apart despite very strong attractive forces. I liked the idea because it turned so much upside down that would have to be carefully rethought. Probably if all the details were considered carefully enough, some version of it would work as well as current approaches. So I did a quick lit search to see whether I could find anybody reputable who'd tried anything like that.

And I found people who had done quark crystals. (Of course they don't consider anything smaller than quarks.) They point out that the nuclear magic numbers are the same as the crystal magic numbers. In important ways atomic nuclei behave exactly like crystals. Their model doesn't explain everything, but then none of the other models do either. And it makes a whole lot of sense. They claim their lattices inherit the good parts of independent particle models, liquid-drop models and cluster models.

I haven't right off noticed any significant criticism. It seems like in the short run nobody's paying attention. That happens a lot in physics.

Eh, I dunno how much is due to people not paying attention.
mu

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Fri Sep 07, 2012 5:02 am UTC

Max™ wrote:
J Thomas wrote:The other one appeared to say something very different. They said that the system at that time did not use relativity and did not need to, that the errors were small. They would need to correct for those small errors to improve the positioning to get the errors smaller than a few centimeters. They explained why it was that the errors were small even though if the system had been designed differently they would be so large that the system could not function.

I am not certain though that the paper means what it flat out definitely says. Perhaps *some* relativistic corrections had been made, and they are talking about *other* relativistic corrections that didn't make much difference. Well, but he discusses each relativistic correction in terms of lorentz transforms and there are none missing that would have been already corrected. Maybe the satellites already corrected some things and he was talking about extra corrections that could be done by receivers but didn't need to be? None of the explanations I come up with quite work. But GPS is not my specialty, and I know that when you look at technical stuff that you aren't familiar with, you can miss a couple of words that change the whole meaning. See what you think. They explained how to do relativistic corrections. They explained why those corrections were not needed in practice, due to details of the way the system was set up. They recommended that those corrections be done anyway, because if the Kalman filter was set differently they *would* matter, and it might be possible to get improved accuracy beyond their current needs. Also, it would be necessary to use relativistic corrections to use GPS far from the earth's surface. What they said was perfectly clear and obvious. Maybe you will see the hidden meaning that I missed.

The reason why the corrections are needed is twofold: raising something out of a gravity well means it runs faster relative to the ground, moving something rapidly relative to the ground means it runs slower.


Sure. Did you look at the paper? You get effects from different heights, different speeds, and different accelerations. They explain how to make the relativistic corrections on pages 190-192. Then they explain how it was actually done. Rather than actually make those three relativistic corrections, instead they jimmied the clocks. They had the moving clocks set to go a tiny amount slower, about -38 microseconds/day. This sort of corrects on average -- the errors will fall equally on both sides. And they had the receivers make a correction for eccentricity of orbit. That got it close enough for GPS units that are mostly stationary on earth. But to do GPS from other satellites, or high-flying planes etc, they needed the real relativistic corrections. Slowing the clocks a set amount only adjusts it for one velocity/altitude etc, -- ours.

On paged 193-194 they explain why the traditional system was adequate. If the true relativistic corrections were made instead of the half-assed way they did it, they would correct errors that amount to less than 5 millimeters. They describe an error which follows the same form as the error people claim would happen without relativistic corrections, and then they show why it is not important. The claim is that the clock-speed fudge-factor corrects the constant part of the error, but it doesn't handle the variable part at all, except on average. If the variable part is all the same direction, you could build up an error of 20 meters in an hour. But that would only happen if you kept the data and let the errors build up. In reality (as I understand it, I may still not be getting it right) you need to correct the errors for four satellites, but for that you need the time for the four satellites. You don't need your own time at all. So errors build up over perhaps half a second between one reading and the next. But they don't build up over hours or days.

So that's where I went wrong. I read about the people who thought it should be done with relativistic corrections. I read about the claims that relativistic corrections would make only small improvements. I read about the argument that sounded exactly like the argument that unacceptable errors would build up without relativistic corrections, that said the errors did not build up.

But really the dispute was not between GPS with relativity versus GPS based on classical thinking. The dispute was between people who wanted real relativistic corrections versus the people who thought the half-assed approach to relativity was good enough.

Meanwhile anybody in space who wants to use GPS has a bunch of clocks that intentionally run slow, and unless his own relative velocity and gravitation etc happen to be just right, that does him no good whatsoever. But he can adjust for that. If we tried to do it classically couldn't people adjust for clocks that ran a little fast? Only if they accepted that the clocks run fast. If they decided that the clocks had no excuse to run fast so they would not adapt to them, then they couldn't make it work. Also they might get some other answers wrong if they used classical methods.

The clocks themselves are atomic and the variations due to different altitudes and velocities would render the entire system inaccurate by a larger and larger amount over time without relativistic corrections.

It really is impossible to make a GPS system work with just naive corrections inserted as "a fudge factor", because it's not just a one time correction when the clocks actually run at a different rate relative to the ground.


It would take sophisticated fudge factors. Relativity shows what those fudge factors are. You don't have to understand the theory that says how it works to find out how to make it work. If you had to understand the source of each error before you could to statistics, you couldn't do statistics.

I want a story that makes sense. See, if all you want is something you can't doubt, just make a table of all the experimental data. There it is. It really happened and got recorded. The results are reproducible within error bounds. There is not much to doubt there, unless you choose to doubt the integrity of the people who did the experiments.

I was speaking of models that I can't doubt, obviously there won't be such a thing, so I have to look at that as an ideal to work towards.

If you're willing to make a little step toward doubt, you can use statistics to clean up your data. Assume that all the little anomalies are unimportant random error -- power spikes that your power supply didn't filter out etc -- and see what you get without them. The data will look sharper and smoother and it will make a bit more sense.

If you're willing to make a big step toward doubt, you can entertain ideas about what could produce that data. You can get ideas that are compatible with the data, and then go out and collect more data and see if it's still compatible, and eventually you can gain a degree of false certainty.

Yeah, I prefer honest doubt.


I really enjoy that feeling when I realize that I have made an assumption that is probably not true, and it's time to reconsider *everything*. A little bit like being underwater and a little bit like being drunk. A whole lot of things you thought you knew are now open to change, and the world is a stranger place than you thought. Opportunities you thought could not happen might open up for you. And things you thought were not dangerous might become so.... An exciting time. So I like to look for unwarranted assumptions, hoping I'll find some. A chance to make the world anew.

And I found people who had done quark crystals. (Of course they don't consider anything smaller than quarks.) They point out that the nuclear magic numbers are the same as the crystal magic numbers. In important ways atomic nuclei behave exactly like crystals. Their model doesn't explain everything, but then none of the other models do either. And it makes a whole lot of sense. They claim their lattices inherit the good parts of independent particle models, liquid-drop models and cluster models.

I haven't right off noticed any significant criticism. It seems like in the short run nobody's paying attention. That happens a lot in physics.

Eh, I dunno how much is due to people not paying attention.


There aren't all that many reputable physicists and they're kind of spread thin. The first time I presented a formal paper (not in physics) I was all revved up because it was great stuff, earthshaking importance, and I had it so clear anybody could see it. And it seemed like everybody was bored. I just couldn't understand it. My mentor told me, "If any of them thought that what you're doing is more interesting than what they're doing, they'd be competing with you to do what you're doing." Maybe somebody's refuted this stuff and I haven't noticed. But more likely nobody has bothered. Everybody has enough on their own plates already.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Fri Sep 07, 2012 5:53 am UTC

The paper explains that using 3 clocks for position and 1 clock for time is the most accurate, but it requires relativistic corrections.
mu

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Fri Sep 07, 2012 12:31 pm UTC

Max™ wrote:The paper explains that using 3 clocks for position and 1 clock for time is the most accurate, but it requires relativistic corrections.


The paper I've been looking at doesn't seem to mention that. It isn't important that it doesn't mention that, but I think we're looking at different papers.

The one I've been looking at is GPS and Relativity: An Engineering Overview by Fliegel and DiEsposti. They give a clear description of how to make relativistic corrections to GPS, and of what GPS did instead of make relativistic corrections.

From their description I believe I have a sense of what the GPS engineers would have done if relativity had not been invented when they did their work. It goes something like this:

"Hey, we set up our prototypes and it doesn't work right. The clocks are slow. Both of them."
"OK, what do we do about that?"
"There's something wrong. They're out of spec, and they tested out before launch."
"We knew we weren't getting the best clocks when we accepted the low bid."
"They both have the same pattern of errors. There's a constant term and then it seems like they're different when they're coming closer to us than when they're moving away. Kind of a doppler effect."
"How about we correct the frequency so the constant term is zero?"
"That would help, but I don't understand how this error could happen."
"Our job is to make a working GPS system. Will it be in spec if we adjust the frequency?"
"Yes, but--"
"Correct the frequency. Think about whether the errors are possible on your own time."
....
"That cleared up the bulk of the problem but when we correct for aberration we aren't getting quite the right directions. It's like the satellite isn't where it says it is."
"OK, what do we do about that?"
"I haven't figured out the source of the error. It could be from the ionosphere. Or maybe the orbits have shifted more than we thought. I can't correct the errors until I figure out what caused them."
"How about we get a bunch of test cases and we give them to a neural net?"
"But you never know what a neural net is *doing* to get its answer!"
"If it gives us a good enough answer, who cares? We can get our work done."
"But we won't understand it!"
"Do it."

Without relativity they would have done just about the same things they did. What relativity did was to give them a prediction ahead of time about one major source of error and several minor ones.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26836
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: What-if 0007: Everybody Out

Postby gmalivuk » Fri Sep 07, 2012 2:12 pm UTC

J Thomas wrote:I really enjoy that feeling when I realize that I have made an assumption that is probably not true, and it's time to reconsider *everything*.
That's where you go wrong, I think. Finding a contradiction in your results does indeed mean that at least one of your assumptions wasn't true, but it does not mean that you should spend equal time reconsidering *every* assumption. If my two assumptions are that energy is conserved and that there aren't any particles that could travel all the way through this block of lead without stopping, and it turns out that my conclusions from those assumptions don't match experiments, I shouldn't throw out both of them. The one I reconsider, then, is the one that doesn't have really amazingly strong theoretical underpinnings and really serious consequences for the universe if it turned out to be false. Instead, I posit that there is in fact some uncharged particle that is small enough to pass through that much lead, and we get neutrinos.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Fri Sep 07, 2012 4:06 pm UTC

gmalivuk wrote:
J Thomas wrote:I really enjoy that feeling when I realize that I have made an assumption that is probably not true, and it's time to reconsider *everything*.
That's where you go wrong, I think. Finding a contradiction in your results does indeed mean that at least one of your assumptions wasn't true, but it does not mean that you should spend equal time reconsidering *every* assumption. If my two assumptions are that energy is conserved and that there aren't any particles that could travel all the way through this block of lead without stopping, and it turns out that my conclusions from those assumptions don't match experiments, I shouldn't throw out both of them. The one I reconsider, then, is the one that doesn't have really amazingly strong theoretical underpinnings and really serious consequences for the universe if it turned out to be false. Instead, I posit that there is in fact some uncharged particle that is small enough to pass through that much lead, and we get neutrinos.


Sure. Although when you know you have a contradiction, and know that you'll have to change something, you don't know what you'll need to change. Sometimes it's as simple as changing definitions. So if you insist that something is a particle, it's no big deal to make it a particle that can travel through a block of lead. Or you can make it a particle that travels every combination of paths, forward and backward in time, and self-cancels on some paths. You can make it a particle that tunnels through barriers, disappearing one place and reappearing somewhere else without passing through the space in between.

I have a conservative friend who gets upset when I ask him to consider new concepts. He says "words have meanings!" and he rejects the new concepts because they must be wrong by the definitions of his words. But physics is not like that. The meanings change with convenience. At this point it looks to me like "particle" means "something that is thought to be in the physical world that physicists want to talk about". Lightwaves are particles. Electric fields and gravitational fields are particles. Forces that hold particles together are particles. But physics theories are not particles. Distances in space-time are not particles.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Fri Sep 07, 2012 4:17 pm UTC

J Thomas wrote:Without relativity they would have done just about the same things they did. What relativity did was to give them a prediction ahead of time about one major source of error and several minor ones.

You seem to be operating under the assumption that relativity just explains a "source of error".

It isn't an error, it is a physical effect influencing the systems in question.

To obtain the correction to add to the transmitted frequency in order to obtain the received
frequency, perform the above steps and reverse the sign. To convert to the time domain,
integrate, and assume any convenient zero of time. Without a timing receiver observing four
satellites, one cannot synchronize to GPS time.

Since GPS receivers work in the time and not in the frequency domain, they handle the velocity,
gravity, and acceleration shifts differently than described above. First, each GPS space vehicle
(SV) clock is offset from its nominal rate by about -4.45xlO-'O (= -38 microseconds per day)
to allow for the relativistic offsets between the differences between the SV and the ground.
Of this -38 microseconds per day, about -45 are due to the gravitational potential difference
between the SV at its mean distance and the earth's surface, and +7 to the mean SV speed,
which is about 3.87 kmlsec. To this mean correction, each receiver must add a term due to
the eccentricity of the GPS orbit. It can be shown that this effect produces a variation in the
SV clock, as seen from the earth, of
where E? is the vector of position of the SV from the ECI, and ? the velocity vector. This is
the equation given in ICD-GPS-200. It is appropriate for users on or near the earth's surface,
but not users in space, who should apply the frequency correction equations given above, or
their integrals to transform to the time domain.


From the paper you are using, this is what is needed to compensate for relativistic effects.

It isn't a trivial effect, it isn't a random approximation which worked, it isn't a newtonian system with fudge factors.

The bit they mentioned where relativistic factors are minor is due to the relative velocity varying as the satellite comes into view, passes it's closest approach, and recedes to the horizon.

THAT portion and that portion alone is treated simply due to the nature of the corrections required being similar to that due to doppler alone.
mu

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 26836
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: What-if 0007: Everybody Out

Postby gmalivuk » Fri Sep 07, 2012 4:31 pm UTC

J Thomas wrote:
gmalivuk wrote:
J Thomas wrote:I really enjoy that feeling when I realize that I have made an assumption that is probably not true, and it's time to reconsider *everything*.
That's where you go wrong, I think. Finding a contradiction in your results does indeed mean that at least one of your assumptions wasn't true, but it does not mean that you should spend equal time reconsidering *every* assumption. If my two assumptions are that energy is conserved and that there aren't any particles that could travel all the way through this block of lead without stopping, and it turns out that my conclusions from those assumptions don't match experiments, I shouldn't throw out both of them. The one I reconsider, then, is the one that doesn't have really amazingly strong theoretical underpinnings and really serious consequences for the universe if it turned out to be false. Instead, I posit that there is in fact some uncharged particle that is small enough to pass through that much lead, and we get neutrinos.
Sure. Although when you know you have a contradiction, and know that you'll have to change something, you don't know what you'll need to change.
Perhaps not, but you most certainly know where you should start. And it's never with throwing out conservation laws.

At this point it looks to me like "particle" means "something that is thought to be in the physical world that physicists want to talk about". Lightwaves are particles. Electric fields and gravitational fields are particles. Forces that hold particles together are particles.
The thing you keep (purposefully?) missing is that there are still different *kinds* of particles, and different ways they can interact, and different sorts of things they can do. And the fact that the particles in particle physics don't behave like our Newtonian intuitions say they should just means that "particle" may not be the best word for them. It doesn't mean we don't have a very clear and precise understanding of how the things actually behave and interact.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Fri Sep 07, 2012 5:01 pm UTC

Max™ wrote:
J Thomas wrote:Without relativity they would have done just about the same things they did. What relativity did was to give them a prediction ahead of time about one major source of error and several minor ones.

You seem to be operating under the assumption that relativity just explains a "source of error".

It isn't an error, it is a physical effect influencing the systems in question.


Yes! But they *treated* it as an error that they could correct.

To obtain the correction to add to the transmitted frequency in order to obtain the received
frequency, perform the above steps and reverse the sign. To convert to the time domain,
integrate, and assume any convenient zero of time. Without a timing receiver observing four
satellites, one cannot synchronize to GPS time.

Since GPS receivers work in the time and not in the frequency domain, they handle the velocity,
gravity, and acceleration shifts differently than described above. First, each GPS space vehicle
(SV) clock is offset from its nominal rate by about -4.45xlO-'O (= -38 microseconds per day)
to allow for the relativistic offsets between the differences between the SV and the ground.
Of this -38 microseconds per day, about -45 are due to the gravitational potential difference
between the SV at its mean distance and the earth's surface, and +7 to the mean SV speed,
which is about 3.87 kmlsec. To this mean correction, each receiver must add a term due to
the eccentricity of the GPS orbit. It can be shown that this effect produces a variation in the
SV clock, as seen from the earth, of
where E? is the vector of position of the SV from the ECI, and ? the velocity vector. This is
the equation given in ICD-GPS-200. It is appropriate for users on or near the earth's surface,
but not users in space, who should apply the frequency correction equations given above, or
their integrals to transform to the time domain.


From the paper you are using, this is what is needed to compensate for relativistic effects.


Yes. Changing the moving clocks does most of the work there, but it does it as a fudge factor. It's a way to simulate simultaneity.

It isn't a trivial effect, it isn't a random approximation which worked, it isn't a newtonian system with fudge factors.


They used to treat it as a newtonian system with fudge factors. That's what the paper is talking about. And the paper describes ways to deal with many of the remaining relativistic effects on top of the existing fudges.

The bit they mentioned where relativistic factors are minor is due to the relative velocity varying as the satellite comes into view, passes it's closest approach, and recedes to the horizon.

THAT portion and that portion alone is treated simply due to the nature of the corrections required being similar to that due to doppler alone.


Yes, they had several relativistic corrections that were minor, in addition to the big ones that were already being fudged adequately. They pointed out that the system had worked OK without them, but there was no strong reason not to account for them.

The second paper claimed that it is not necessary to treat it as a newtonian system with relativistic fudge factors. You have four satellites, all at roughly equivalent heights. They have relative motion and acceleration which leaves each of them recording the others' clocks as slow. There is no need to "correct" orbiting clocks, or run any clocks fast or slow, just solve the relativistic space-time problem and you're done. (Apart from all the *other* error corrections which must still be performed.) You would use the same solution for GPS on earth that you have to do anyway for GPS solutions everywhere else.

I think the obvious problem with this is that we have a great big installed base of GPS machines that work adequately the way they are. The new system would be better in theory, and it might be better in practice, but it would be expensive to switch and what we have now works well enough.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Pfhorrest
Posts: 5487
Joined: Fri Oct 30, 2009 6:11 am UTC
Contact:

Re: What-if 0007: Everybody Out

Postby Pfhorrest » Fri Sep 07, 2012 7:09 pm UTC

J Thomas wrote:At this point it looks to me like "particle" means "something that is thought to be in the physical world that physicists want to talk about". Lightwaves are particles. Electric fields and gravitational fields are particles. Forces that hold particles together are particles.

Not exactly. And the issue here isn't so much one of redefining "particle" to mean something radically different from what was previously meant by it -- it's more of discovering that particles are radically different from what we thought they were.

A particle is a "small part or division of a whole". Particles are the little bits of stuff that things are made of. And we've discovered that things are made of very different stuff than we once thought. They're not made of tiny indestructible perfectly elastic billiard balls all knocking around. They're not made of infinitesimal points pushing and pulling on each other via action at a distance. They're made of perturbations in energy fields, which affect each other in much more complicated ways than just bouncing off of or pushing and pulling on each other. That has all kinds of counter-intuitive implications. It means that two things of different kinds can interact and result in one thing of a third kind which is not made of the original two things, in the sense that they're not still inside it somewhere, they don't exist anymore, and instead this new third thing exists (which conserves various properties of the first two) -- or it can result in two new things of two new kinds, taking a proton and an electron and turning them into a neutron and neutrino. It means that if the underlying energy field is perturbed, new particles can spring out of nowhere, or disappear into nowhere -- so long as all the important properties of the energy field are still conserved. And it means that interactions between particles at a distance, being an exchange of various properties (like say momentum, for a simple example of two particles attracting or repelling), looks exactly like the emission and absorption of other particles between them, and so can be accurately described as such.

But they're still the discrete little bits that things are made out of, so they're still particles. And it's not like there are still things which match our old understanding of particles out there, and we're begun misapplying the word to a broader variety of things. It just turns out that all the things that we called particles behave differently than we thought they did. That doesn't mean we have to start calling them something else, or that we've redefined the word "particle".
Forrest Cameranesi, Geek of All Trades
"I am Sam. Sam I am. I do not like trolls, flames, or spam."
The Codex Quaerendae (my philosophy) - The Chronicles of Quelouva (my fiction)

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Fri Sep 07, 2012 7:42 pm UTC

J Thomas wrote:
Max™ wrote:
J Thomas wrote:Without relativity they would have done just about the same things they did. What relativity did was to give them a prediction ahead of time about one major source of error and several minor ones.

You seem to be operating under the assumption that relativity just explains a "source of error".

It isn't an error, it is a physical effect influencing the systems in question.


Yes! But they *treated* it as an error that they could correct.

It's not an error, it's a known effect which has to be compensated for.

Error usually implies unexpected or unknown, no one was surprised that clocks ran faster out of a gravity well and slower when moving rapidly relative to an observer.

Yes. Changing the moving clocks does most of the work there, but it does it as a fudge factor. It's a way to simulate simultaneity.

Nowhere did I see any mention of simultaneity, nor a fudge factor.

It's known that the clocks run at a different rate due to relativistic effects, thus it is compensated for so receivers on the ground don't have to perform additional calculations on multiple time signals.

They used to treat it as a newtonian system with fudge factors. That's what the paper is talking about. And the paper describes ways to deal with many of the remaining relativistic effects on top of the existing fudges.

There is no "used to", the newtonian "fudge factor" portion you mention is ONLY speaking of the variation in relative velocity as the satellite traces an arc across the sky relative to an observer. It turns out that the error from not including one portion of relativistic effects in this one case is canceled out by the error from including another portion, so you can just call it close enough and apply a general adjustment.


There are three things being discussed here:

1. Clocks raised out of a gravity well run faster than those deeper inside the well.

2. Clocks moving rapidly relative to an observer will be seen as running slower than the observers clock.

3. Satellites follow a curved path across the sky relative to each observer, this introduces relativistic effects which can for the most part be left out by using a newtonian assumption plus a general correction, this is only due to the fact that the effects across the entire path balance out almost exactly, so there is less of a need to use relativistic corrections in this one instance. Ideally it would still be done, but it would not degrade the accuracy as badly as the first two effects would if relativity were ignored.

Yes, they had several relativistic corrections that were minor, in addition to the big ones that were already being fudged adequately. They pointed out that the system had worked OK without them, but there was no strong reason not to account for them.

The second paper claimed that it is not necessary to treat it as a newtonian system with relativistic fudge factors. You have four satellites, all at roughly equivalent heights. They have relative motion and acceleration which leaves each of them recording the others' clocks as slow. There is no need to "correct" orbiting clocks, or run any clocks fast or slow, just solve the relativistic space-time problem and you're done. (Apart from all the *other* error corrections which must still be performed.) You would use the same solution for GPS on earth that you have to do anyway for GPS solutions everywhere else.

No, you do need to adjust the clocks to account for the relativistic effects, otherwise the system becomes increasingly inaccurate.

You can omit CERTAIN effects which are strictly due to the curved trajectory the satellite follows across the sky from horizon to horizon, relative to your position.

I think the obvious problem with this is that we have a great big installed base of GPS machines that work adequately the way they are. The new system would be better in theory, and it might be better in practice, but it would be expensive to switch and what we have now works well enough.

No, GPS systems all use relativistic corrections for the distance out of the gravity well and the relative velocity, and any GPS system with those corrections will work adequately without including the other corrections, as they only incur a small error which mostly cancels out.
mu

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Sat Sep 08, 2012 12:44 am UTC

Pfhorrest wrote:
J Thomas wrote:At this point it looks to me like "particle" means "something that is thought to be in the physical world that physicists want to talk about". Lightwaves are particles. Electric fields and gravitational fields are particles. Forces that hold particles together are particles.

Not exactly. And the issue here isn't so much one of redefining "particle" to mean something radically different from what was previously meant by it -- it's more of discovering that particles are radically different from what we thought they were.


That's a peculiar semantic interpretation. We had meanings for the word, and it turned out the meanings did not fit the reality. So we redefined the word to mean something radically different from what we meant before -- better meanings that fit our current understanding -- but you say that the word hasn't been redefined to mean something radically different from the wrong thing it previously meant. Strange.

Perhaps we should have gone on using the word"phlogiston" but changed the meaning to fit reality, rather than give up the word when it turned out not to fit....

A particle is a "small part or division of a whole". Particles are the little bits of stuff that things are made of. And we've discovered that things are made of very different stuff than we once thought. They're not made of tiny indestructible perfectly elastic billiard balls all knocking around. They're not made of infinitesimal points pushing and pulling on each other via action at a distance. They're made of perturbations in energy fields, which affect each other in much more complicated ways than just bouncing off of or pushing and pulling on each other. That has all kinds of counter-intuitive implications. It means that two things of different kinds can interact and result in one thing of a third kind which is not made of the original two things, in the sense that they're not still inside it somewhere, they don't exist anymore, and instead this new third thing exists (which conserves various properties of the first two) -- or it can result in two new things of two new kinds, taking a proton and an electron and turning them into a neutron and neutrino.


This is another semantic thing. What would it mean for the proton and electron to continue to exist inside the neutron? Whatever it means, how would you show that it doesn't happen?

The view I have of all this stuff is that you can describe it mathematically, and if the math fits the current experimental reality, then the math is tentatively correct. Then any explanation which fits the math, is equivalent to any other explanation which fits the math.

It means that if the underlying energy field is perturbed, new particles can spring out of nowhere, or disappear into nowhere -- so long as all the important properties of the energy field are still conserved. And it means that interactions between particles at a distance, being an exchange of various properties (like say momentum, for a simple example of two particles attracting or repelling), looks exactly like the emission and absorption of other particles between them, and so can be accurately described as such.


Once you've redefined your particles this way, what would it take for something not to be a particle? You have described a limited sampling of the things particles can do. What can particles not do?

But they're still the discrete little bits that things are made out of, so they're still particles.


Except they aren't exactly discrete. Or maybe they are, who knows? How would you tell whether particles are discrete or not? If a particular molecule loses a particular quanta of energy, and sometime later one or two or three other particles each absorb roughly the same quanta of energy, how do you decide whether one of them got the whole thing from the first molecule? Could they have gotten fractional quanta from various sources?

And it's not like there are still things which match our old understanding of particles out there, and we're begun misapplying the word to a broader variety of things. It just turns out that all the things that we called particles behave differently than we thought they did. That doesn't mean we have to start calling them something else, or that we've redefined the word "particle".


If we called modern heat things "phlogiston" when it turns out that heat behaves differently from what we thought it did when we called it phlogiston, would you say that we had not redefined the word "phlogiston"?

But all this is not very important. It's just about how we use the words, and the words don't matter much to physics. What's far more important is how the math works. I run into problems with that a lot with laymen who think they know physics. (I'm one of those myself, except I don't have much illusion about the depth of my understanding.) Like, one time I came up with a way to describe light polarization in terms of quaternions. You can have an axis of rotation in any direction, and depending on the direction of the axis you get linear or circular or elliptical polarization. But when I showed it to these people they insisted that elliptical polarization *is* a linear combination of two linear polarizations. No other description is correct because this is the way it is. To my way of thinking, if the math comes out the same then the explanations are equivalent and you can use whichever helps your intuition. But they were convinced that one way was right and all other equivalent descriptions had to be wrong.

When Xx(YxZ) = Y(X.Z ) - Z(X.Y) then you don't say that Xx(YxZ) is right but Y(X.Z ) - Z(X.Y) is wrong, do you? Well yes, people do. Mostly people who don't understand math or physics, though.
The Law of Fives is true. I see it everywhere I look for it.

User avatar
PM 2Ring
Posts: 3715
Joined: Mon Jan 26, 2009 3:19 pm UTC
Location: Sydney, Australia

Re: What-if 0007: Everybody Out

Postby PM 2Ring » Sat Sep 08, 2012 9:27 am UTC

J Thomas wrote:
PM 2Ring wrote:Well, people tried for a long time to isolate quarks, in order to prove that the quark theory was valid but they failed, and that failure almost killed quark theory.

This is one of the things I don't understand about physics. They saw a pattern of behaviors, and they decided it implied new particles to make that pattern happen. Then when they couldn't isolate the particles they predicted, it almost killed the theory. But instead they decided the particles really are there but they're special particles that can never be isolated. Whenever you *do* isolate the special particle it quick rips new particles out of thin air to keep itself from being isolated. Literally. (Vacuum is very thin air.)
http://en.wikipedia.org/wiki/Hadronization
But isn't the observed pattern the important thing?


Sure, observation has to be the basis, so the theory ought to predict the observed patterns and it oughtn't predict stuff that's never observed.

Before quark theory was born it was pretty obvious that protons and neutrons are composite particles, since free neutrons spontaneously decay (so they can't be truly fundamental), and from the POV of the strong force protons and neutrons are extremely similar to each other. It gradually became clear from the symmetries that arise in various collider experiments that the proton has 3 charge centres. (IIRC, this was first done at SLAC, firing high speed electrons at proton targets; I don't know when similar experiments were first done bombarding neutrons with electrons). So although quarks can't exist in isolation, we can still detect their presence in composite particles. Note that these observations demolish your idea mentioned later in this thread of nucleons containing 4 components of fractional charge.

J Thomas wrote:
PM 2Ring wrote:Bear in mind that fundamental particles aren't the ultimate entities that they were once considered to be, they're really just quantized excitations of the various fundamental quantum fields that fill space. Each of these fields has various kinds of symmetry associated with it; these symmetries are ultimately the basis of the properties of the field, and are intimately related to the various conservation laws.

Maybe this stuff would be easier to follow if you guys stopped calling them "particles". The word has a whole lot of baggage that (it looks like) no longer applies.


Maybe. In fact one physics professor who contributes to another science forum I visit prefers the term "quantum thingy". :) OTOH, I have no problem with the approach espoused by Pfhorrest. And these quantum thingies do behave a bit like traditional particles in certain circumstances. So it's reasonable (IMHO) to continue to call them particles, as long as you bear in mind that in other circumstances their wave-like nature will predominate.

Most interpretations of QM say that if you ask "is light (or an electron or other quantum thingy) really a wave or really a particle" then you're asking the wrong question. What you get depends on what you're looking for. If you want to determine the exact location of a quantum thingy you can do so, provided that you allow its momentum state to become undetermined. In such a state, you can measure an exact location for the quantum thingy and it will thus look like a particle. Conversely, if you want to determine the exact momentum of a quantum thingy, then you have to allow its position state to become undetermined, and it ends up looking like a wave.

Mathematically, this happens because the momentum state is a Fourier transform of the position state. Physically, we call it the Heisenberg Uncertainty Principle. Heisenberg himself originally thought that this uncertainty relationship was due to the interaction between the measuring device and the quantum thingy; the modern approach is to see that it's an unavoidable consequence of the mathematics of the Schrödinger wave equation that we use to model the quantum behaviour of the field. Mathematically, we could drop the notion of particle entirely, and only talk about various combinations of waves, but some of those wave combinations are rather particle-like, so the particle approach retains its usefulness. But ultimately it's the symmetries underlying the fields' behaviours that are the truly important thing.

J Thomas wrote:
PM 2Ring wrote:Why is a neutrino mirror magical? Well, reflection requires the reflected particles to interact with the reflector, but neutrinos are rather reticent when it comes to interactions. I guess you could deflect them gravitationally, if you had a convenient black hole, but if you've got a tame black hole there are more efficient ways of using it as a rocket engine. :)

Sure. It would take something that we don't know about yet. But suppose we were having this sort of discussion in 1962, just 50 years ago. Imagine that I proposed some sort of space travel that used particles which have fractional charge. Everybody would be jumping all over me because I didn't know that there is no such thing as fractional charge, charges are always integers. Every particle has either no charge, or a single positive charge, or a single negative charge. Or composite particles like alpha particles can have some multiple of integer charges. Everybody who knew anything about physics knew this. It wasn't until 1964 that reputable physicists suggested particles with fractional charge, and there was considerable resistance to the idea particularly when no such particles could be found.

Fair comment.

J Thomas wrote:I certainly don't want to imply that I think I'm right and every physicist is wrong, that I know something special which says someday neutrinos will be useful for propulsion in space. I'm only making the rather trite point that nobody knows very much about physics yet. We know a lot of things we can use to get predictable results in narrowly-defined circumstances. We don't know a lot about what's always impossible versus what may be possible sometimes.


I disagree. We currently know quite a bit of physics: current theory covers the behaviour of normal matter under all but the most extreme circumstances. High energy particle collider experiments allow us to confidently model what happens in supernova explosions, and the upper layers of neutron stars. We're not totally sure about what goes on in the interior of neutron stars or inside the event horizon of black holes, since we need some kind of quantum gravity theory to handle those situations properly. OTOH, we are fairly ignorant about the nature of dark matter, and that might seem to be a major failure of current theory, given that DM is such a large component of the universe by mass. But DM, by its very nature, is rather boring stuff. :) If DM did interesting things apart from exerting its gravitational influence, then it wouldn't be so dark.

J Thomas
Everyone's a jerk. You. Me. This Jerk.^
Posts: 1190
Joined: Fri Sep 23, 2011 3:18 pm UTC

Re: What-if 0007: Everybody Out

Postby J Thomas » Sat Sep 08, 2012 2:13 pm UTC

PM 2Ring wrote:
J Thomas wrote:
PM 2Ring wrote:Well, people tried for a long time to isolate quarks, in order to prove that the quark theory was valid but they failed, and that failure almost killed quark theory.

This is one of the things I don't understand about physics. They saw a pattern of behaviors, and they decided it implied new particles to make that pattern happen. Then when they couldn't isolate the particles they predicted, it almost killed the theory. But instead they decided the particles really are there but they're special particles that can never be isolated. Whenever you *do* isolate the special particle it quick rips new particles out of thin air to keep itself from being isolated. Literally. (Vacuum is very thin air.)
http://en.wikipedia.org/wiki/Hadronization
But isn't the observed pattern the important thing?


Sure, observation has to be the basis, so the theory ought to predict the observed patterns and it oughtn't predict stuff that's never observed.


OK. I say it's OK to have theories which predict things that are never observed and vice versa, just don't try to apply them to the real world as currently observed. ;)

Before quark theory was born it was pretty obvious that protons and neutrons are composite particles, since free neutrons spontaneously decay (so they can't be truly fundamental), and from the POV of the strong force protons and neutrons are extremely similar to each other. It gradually became clear from the symmetries that arise in various collider experiments that the proton has 3 charge centres. (IIRC, this was first done at SLAC, firing high speed electrons at proton targets; I don't know when similar experiments were first done bombarding neutrons with electrons). So although quarks can't exist in isolation, we can still detect their presence in composite particles. Note that these observations demolish your idea mentioned later in this thread of nucleons containing 4 components of fractional charge.


I don't remember an idea that would require a proton not to appear to have three charge centers when hit by a high speed electron. Quarks have charges with size 1/3 or 2/3, and it isn't impossible that they might separate out into pieces each with 1/3 charge sometimes, and bind into 2/3 charge pieces other times. I wouldn't want to claim it's true unless there was some reason that a model like that was better, and I don't have such a reason except it was easier for me to think about when I was heading in a particular direction. If that approach were to get very promising results, then I would go back and see whether I could do it with some parts that had 2/3 charge, or whether I should look at ways it could do what I wanted and still change to fit the known data when hit by a high-energy electron. But if there's no big advantage to thinking that way in the first place, I can put it aside without bothering about those details.

Similarly, I'd like to consider the idea of quarks each containing (composed of) a whole bunch of electrons and positrons. And perhaps a nucleus in a low-energy state where the quarks smooth out to so the whole nucleus becomes a sort of crystal of electrons and positrons. If that approach explains something useful, then it's worth seeing whether it can be compatible with high-energy data. But if there's no advantage to it in the first place, put it aside.

J Thomas wrote:
PM 2Ring wrote:Bear in mind that fundamental particles aren't the ultimate entities that they were once considered to be, they're really just quantized excitations of the various fundamental quantum fields that fill space. Each of these fields has various kinds of symmetry associated with it; these symmetries are ultimately the basis of the properties of the field, and are intimately related to the various conservation laws.

Maybe this stuff would be easier to follow if you guys stopped calling them "particles". The word has a whole lot of baggage that (it looks like) no longer applies.


Maybe. In fact one physics professor who contributes to another science forum I visit prefers the term "quantum thingy". :) OTOH, I have no problem with the approach espoused by Pfhorrest. And these quantum thingies do behave a bit like traditional particles in certain circumstances. So it's reasonable (IMHO) to continue to call them particles, as long as you bear in mind that in other circumstances their wave-like nature will predominate.

Most interpretations of QM say that if you ask "is light (or an electron or other quantum thingy) really a wave or really a particle" then you're asking the wrong question. What you get depends on what you're looking for. If you want to determine the exact location of a quantum thingy you can do so, provided that you allow its momentum state to become undetermined. In such a state, you can measure an exact location for the quantum thingy and it will thus look like a particle. Conversely, if you want to determine the exact momentum of a quantum thingy, then you have to allow its position state to become undetermined, and it ends up looking like a wave.

Mathematically, this happens because the momentum state is a Fourier transform of the position state. Physically, we call it the Heisenberg Uncertainty Principle. Heisenberg himself originally thought that this uncertainty relationship was due to the interaction between the measuring device and the quantum thingy; the modern approach is to see that it's an unavoidable consequence of the mathematics of the Schrödinger wave equation that we use to model the quantum behaviour of the field. Mathematically, we could drop the notion of particle entirely, and only talk about various combinations of waves, but some of those wave combinations are rather particle-like, so the particle approach retains its usefulness. But ultimately it's the symmetries underlying the fields' behaviours that are the truly important thing.


I'd like to ask a serious question about this, since I've never done the math myself. You say that the momentum state is a Fourier transform of the position state, and so you can't be clear about both at the same time. How does this translate experimentally? Can you take the same data, and interpret it one way to get position but the momentum is unknown, or interpret it a different way to get momentum but the position is unknown, or maybe somewhere inbetween so you're fuzzy about the position and momentum both but you have a moderately good idea of both?

Or is it that you can measure position while randomizing the momentum, or measure momentum while randomizing position?

The first way, position and momentum are mathematical constructs that maybe represent parts of something real that you can get a fuzzy measurement for. The second way, they could both be real and separate, distinct, but your interaction with them messes them up.

Is one of these interpretations right? Or both, or neither?

J Thomas wrote:
PM 2Ring wrote:Why is a neutrino mirror magical? Well, reflection requires the reflected particles to interact with the reflector, but neutrinos are rather reticent when it comes to interactions. I guess you could deflect them gravitationally, if you had a convenient black hole, but if you've got a tame black hole there are more efficient ways of using it as a rocket engine. :)

Sure. It would take something that we don't know about yet. ....

Fair comment.

J Thomas wrote:I certainly don't want to imply that I think I'm right and every physicist is wrong, that I know something special which says someday neutrinos will be useful for propulsion in space. I'm only making the rather trite point that nobody knows very much about physics yet. We know a lot of things we can use to get predictable results in narrowly-defined circumstances. We don't know a lot about what's always impossible versus what may be possible sometimes.


I disagree. We currently know quite a bit of physics: current theory covers the behaviour of normal matter under all but the most extreme circumstances. High energy particle collider experiments allow us to confidently model what happens in supernova explosions, and the upper layers of neutron stars. We're not totally sure about what goes on in the interior of neutron stars or inside the event horizon of black holes, since we need some kind of quantum gravity theory to handle those situations properly. OTOH, we are fairly ignorant about the nature of dark matter, and that might seem to be a major failure of current theory, given that DM is such a large component of the universe by mass. But DM, by its very nature, is rather boring stuff. :) If DM did interesting things apart from exerting its gravitational influence, then it wouldn't be so dark.
[/quote][/quote]

Yes, those are all great accomplishments. There is a lot that people know. And a whole lot of what people observe does fit the theory well enough.

If you say "Physics observes that X happens" then that's fine. But if you say "Physics observes that X is impossible and cannot ever happen" then I will doubt you unless you have X defined in a way that makes it impossible by definition.

All the things you point out that physicists are uncertain about, are things that are hard to observe. Of course it's difficult to get good testable theories when it's difficult to get data. And until you do get that data, you don't know what it will be. That won't matter a whole lot while the hard-to-get data stays hard-to-get. As long as the inside of neutron stars and the inside of black holes have very little influence on things outside, we can afford to be ignorant of them. They won't make much difference to what we *can* measure.

So the possibility for really interesting new stuff comes from the chance we can find ways to manipulate stuff in ways that rarely happen in nature. Or that rarely happen in a coherent large-scale way that's easy to observe. Once we had nuclear reactors we found natural reactors where fissionables had been concentrated by geology. Once we had lasers we found lasers in the sky. Etc. There might be interesting behaviors available to us that we haven't noticed yet, that would allow engineering we can hardly imagine now. Some of those behaviors might depend on combinations of known fundamental properties. Some of them might allow refined theories. Things that are currently unknown, that are ignored or assumed to be unknowable by definition, might become knowable in some circumstances.

So for example, dark matter might turn out to have some very interesting properties, but we won't find out until they are actually observed.

"You certainly usually find something, if you look, but it is not always quite the something you were after.” Bilbo Baggins
The Law of Fives is true. I see it everywhere I look for it.

User avatar
Max™
Posts: 1792
Joined: Thu Jun 21, 2012 4:21 am UTC
Location: mu

Re: What-if 0007: Everybody Out

Postby Max™ » Sun Sep 09, 2012 6:08 am UTC

http://www.sciencedaily.com/releases/20 ... 125154.htm

"We designed an apparatus to measure a property -- the polarization -- of a single photon. We then needed to measure how much that apparatus disturbed that photon," says Lee Rozema, a Ph.D. candidate in Professor Aephraim Steinberg's quantum optics research group at U of T, and lead author of a study published this week in Physical Review Letters.

"To do this, we would need to measure the photon before the apparatus but that measurement would also disturb the photon," Rozema says.

In order to overcome this hurdle, Rozema and his colleagues employed a technique known as weak measurement wherein the action of a measuring device is weak enough to have an imperceptible impact on what is being measured. Before each photon was sent to the measurement apparatus, the researchers measured it weakly and then measured it again afterwards, comparing the results. They found that the disturbance induced by the measurement is less than Heisenberg's precision-disturbance relation would require.

"Each shot only gave us a tiny bit of information about the disturbance, but by repeating the experiment many times we were able to get a very good idea about how much the photon was disturbed," says Rozema.

The findings build on recent challenges to Heisenberg's principle by scientists the world over. Nagoya University physicist Masanao Ozawa suggested in 2003 that Heisenberg's uncertainty principle does not apply to measurement, but could only suggest an indirect way to confirm his predictions. A validation of the sort he proposed was carried out last year by Yuji Hasegawa's group at the Vienna University of Technology. In 2010, Griffith University scientists Austin Lund and Howard Wiseman showed that weak measurements could be used to characterize the process of measuring a quantum system. However, there were still hurdles to clear as their idea effectively required a small quantum computer, which is difficult to build.

"In the past, we have worked experimentally both on implementing weak measurements, and using a technique called 'cluster state quantum computing' to simplify building quantum computers. The combination of these two ideas led to the realization that there was a way to implement Lund and Wiseman's ideas in the lab," says Rozema.

It is often assumed that Heisenberg's uncertainty principle applies to both the intrinsic uncertainty that a quantum system must possess, as well as to measurements. These results show that this is not the case and demonstrate the degree of precision that can be achieved with weak-measurement techniques.

"The results force us to adjust our view of exactly what limits quantum mechanics places on measurement," says Rozema. "These limits are important to fundamental quantum mechanics and also central in developing 'quantum cryptography' technology, which relies on the uncertainty principle to guarantee that any eavesdropper would be detected due to the disturbance caused by her measurements."

"The quantum world is still full of uncertainty, but at least our attempts to look at it don't have to add as much uncertainty as we used to think!"
mu


Return to “What If?”

Who is online

Users browsing this forum: No registered users and 4 guests