5 ms·

If I understood the video correctly, it seems like the new model was proposed after the experiment was run. Wouldn't that make it extremely easy to fit the theo

If I understood the video correctly, it seems like the new model was proposed after the experiment was run. Wouldn't that make it extremely easy to fit the theory to the experiment?

beecafe 5y ago
If the experiment invalidates your old model, eventually you need to come up with a new one.
naasking 5y ago
Basically yes. This hypothesis has been getting way too much attention. Science journalism really needs to improve.
kortex 5y ago
Yup. This hypothesis revolves around posited "chameleon-screened dark energy" - particles/fields whose mass and coupling with other field fluctuates based on their surrounds. While I guess this isn't impossible, it reeks of Conveniencium - a field with properties that are non-homogeneous and just happen to match your model at both microscopic and cosmic scales? Pull the other one. Xenon experiments have been undergoing a bit of a crisis of faith as of late. Each iteration is a cycle of of "we predict dark matter in this energy window" -> find nothing out of place -> "well we can't rule out the long tail" -> build a bigger, more sensitive xenon detector. Grossly oversimplified and it's been a hot minute since undergrad. But the gist is WIMPs are excluded from more and more regimes and they are running out of places to look.
- ISL 5y ago
  Chameleons are a sign that experiments are making real progress. When they first emerged, my reaction was, "seriously?!", but as I've matured, I've realized that they're a real theoretical adaptation to the experimental reality. We observe two apparently incongruous facts: 1) Something that we don't understand is operating on very long lengthscales (all of the effects accorded to dark matter and dark energy), which normally would suggest some very light particle to mediate that interaction 2) We don't see any new light particles (see inverse-square law tests of gravity, for example). Chameleons offer a strategy for simultaneously explaining both. They're far from the only option in town. Intuitively, I don't really like them, but objectively, I can't see a reason to exclude them. Self-interacting theories are general (even electromagnetism is very-weakly self-interacting); a chameleon advocate could reasonably go so far as to suggest that such things are the natural order of things...
  - kortex 5y ago
    Huh, interesting. You're absolutely right in that they definitely aren't de-factor excludable. They just feel very un-parsimonious, though I can see how they are progress nonetheless if they can form a cohesive theory. I harbor some slightly more radical opinions in that I think the whole dark energy hypothesis is broken, hell I'm not even fully convinced that the universe is accelerating in its expansion globally, and therefore the big lambda term may be way off (there are many potential other explanations for the supernovae curves out there). But I'm an engineer not a pro physicist, so my headcannon isn't worth much :).
ajross 5y ago
Uh... so? I mean, Darwin had to see the weird finches before he saw a need for a new theory to explain them. Newton had to get hit by the apocryphal apple. Plank et. al. needed to be bothered by all those funny lines in spectra. Fitting theory to observation is how science is *supposed* to work, 99% of the time. The reason we get fooled into thinking otherwise is that we tend to celebrate ab initio theories as the work of genius (really relativity specifically). But normal science is about finding something that doesn't make sense and making sense of it. Of course, after that, when you have a promising theory, you then start trying to make other predictions to show your theory isn't just an overfit model. And clearly that hasn't happened yet here, so we should apply a little salt. But I see nothing wrong with the process here at all.
- bwooceli 5y ago
  ... and Michaelson clung to the luminiferous aether until he died.

If I understood the video correctly, it seems like the new model was proposed after the experiment was run. Wouldn't that make it extremely easy to fit the theory to the experiment?

beecafe 5y ago

If the experiment invalidates your old model, eventually you need to come up with a new one.

naasking 5y ago

Basically yes. This hypothesis has been getting way too much attention. Science journalism really needs to improve.

kortex 5y ago

Yup. This hypothesis revolves around posited "chameleon-screened dark energy" - particles/fields whose mass and coupling with other field fluctuates based on their surrounds. While I guess this isn't impossible, it reeks of Conveniencium - a field with properties that are non-homogeneous and just happen to match your model at both microscopic and cosmic scales? Pull the other one. Xenon experiments have been undergoing a bit of a crisis of faith as of late. Each iteration is a cycle of of "we predict dark matter in this energy window" -> find nothing out of place -> "well we can't rule out the long tail" -> build a bigger, more sensitive xenon detector. Grossly oversimplified and it's been a hot minute since undergrad. But the gist is WIMPs are excluded from more and more regimes and they are running out of places to look.

ISL 5y ago
Chameleons are a sign that experiments are making real progress. When they first emerged, my reaction was, "seriously?!", but as I've matured, I've realized that they're a real theoretical adaptation to the experimental reality. We observe two apparently incongruous facts: 1) Something that we don't understand is operating on very long lengthscales (all of the effects accorded to dark matter and dark energy), which normally would suggest some very light particle to mediate that interaction 2) We don't see any new light particles (see inverse-square law tests of gravity, for example). Chameleons offer a strategy for simultaneously explaining both. They're far from the only option in town. Intuitively, I don't really like them, but objectively, I can't see a reason to exclude them. Self-interacting theories are general (even electromagnetism is very-weakly self-interacting); a chameleon advocate could reasonably go so far as to suggest that such things are the natural order of things...
- kortex 5y ago
  Huh, interesting. You're absolutely right in that they definitely aren't de-factor excludable. They just feel very un-parsimonious, though I can see how they are progress nonetheless if they can form a cohesive theory. I harbor some slightly more radical opinions in that I think the whole dark energy hypothesis is broken, hell I'm not even fully convinced that the universe is accelerating in its expansion globally, and therefore the big lambda term may be way off (there are many potential other explanations for the supernovae curves out there). But I'm an engineer not a pro physicist, so my headcannon isn't worth much :).

ajross 5y ago

Uh... so? I mean, Darwin had to see the weird finches before he saw a need for a new theory to explain them. Newton had to get hit by the apocryphal apple. Plank et. al. needed to be bothered by all those funny lines in spectra. Fitting theory to observation is how science is *supposed* to work, 99% of the time. The reason we get fooled into thinking otherwise is that we tend to celebrate ab initio theories as the work of genius (really relativity specifically). But normal science is about finding something that doesn't make sense and making sense of it. Of course, after that, when you have a promising theory, you then start trying to make other predictions to show your theory isn't just an overfit model. And clearly that hasn't happened yet here, so we should apply a little salt. But I see nothing wrong with the process here at all.

bwooceli 5y ago
... and Michaelson clung to the luminiferous aether until he died.