4 ms·

I don’t often go looking in Astronomy or cosmology journals, but I’d be surprised if either scientists were trying to publish articles like that — or if they we

I don’t often go looking in Astronomy or cosmology journals, but I’d be surprised if either scientists were trying to publish articles like that — or if they were, that they could pass peer review. On the other hand, if you are working on theories that can extend into the early universe, it’s not unimportant to try and figure out where the model breaks down. Maybe t=0 isn’t possible, but t=1s? 1ns? 1ps? How much can we feasibly describe? I’d argue that not exploring the limits of models is also bad science. Knowing the limits is a fundamental part of communicating a model.

tsimionescu 4y ago
> I’d be surprised if either scientists were trying to publish articles like that — or if they were, that they could pass peer review. The article we're discussing itself even links to one such paper [0]. All of the others she mentions are also published works - Penrose's CCC [1], the ekpyrotic universe [2], Hawking's no-boundary state [3] etc. > On the other hand, if you are working on theories that can extend into the early universe, it’s not unimportant to try and figure out where the model breaks down. Maybe t=0 isn’t possible, but t=1s? 1ns? 1ps? How much can we feasibly describe? Sure, but this is a different thing. Many of these are adding elements to the existing theories, and then predict a new initial state given the modified evolution laws. [0] https://link.springer.com/article/10.1007/s10714-021-02790-7 https://link.springer.com/article/10.1007/s10714-021-02790-7 [1] https://arxiv.org/abs/1808.01740?context=astro-ph https://arxiv.org/abs/1808.01740?context=astro-ph [2] https://arxiv.org/abs/hep-th/0103239 https://arxiv.org/abs/hep-th/0103239 [3] https://arxiv.org/abs/1707.07702 https://arxiv.org/abs/1707.07702
- XorNot 4y ago
  The point of publishing those articles is...to publish them though. Like, you have an idea, you write it up, submit it and note that it can fit known data but isn't currently testable. Done, and important. Maybe it goes nowhere, maybe it inspires someone, but the point of journals is in the name: they're journals of work in the field, shared so the community can explore and benefit from them. They're not publications of "what is definitely true", they are fundamentally explorations of what could be, or the more important "this is kind of interesting where could it lead?".
  - tsimionescu 4y ago
    > note that it can fit known data but isn't currently testable. But the thing is, theories about the beginning of the universe will never be testable, they aren't just not currently testable. So, if your theory has no novel predictions about the future, but it adds extra parameters to obtain a different prediction about a past which exists beyond what can be measured, then you're wasting your time creating this theory, and wasting reviewers' and readers' time publishing it; and you're wasting money researching it. This is what Sabine usually writes and complains about - research money being spent on research that is at best unlikely to bear any fruit, and at worst navel-gazing, especially when there are very real problems in physics that are not receiving significant research. This is why she complains about people researching the beginning of the universe, or black hole entropy, or grand unified theories, or the hierarchy "problem", or looking for supersimmetry or for WIMPs in ever larger particle accelerators. Instead, she wishes more people were researching the measurement problem, non-linearity in quantum mechanics, high-energy physics through radio-telescopes instead of particle accelerators, to name a few things. Now, I don't know anywhere near enough to say that she is right, but I do believe she is not trivially wrong, like you seem to be suggesting.

tsimionescu 4y ago

> I’d be surprised if either scientists were trying to publish articles like that — or if they were, that they could pass peer review. The article we're discussing itself even links to one such paper [0]. All of the others she mentions are also published works - Penrose's CCC [1], the ekpyrotic universe [2], Hawking's no-boundary state [3] etc. > On the other hand, if you are working on theories that can extend into the early universe, it’s not unimportant to try and figure out where the model breaks down. Maybe t=0 isn’t possible, but t=1s? 1ns? 1ps? How much can we feasibly describe? Sure, but this is a different thing. Many of these are adding elements to the existing theories, and then predict a new initial state given the modified evolution laws. [0] https://link.springer.com/article/10.1007/s10714-021-02790-7 https://link.springer.com/article/10.1007/s10714-021-02790-7 [1] https://arxiv.org/abs/1808.01740?context=astro-ph https://arxiv.org/abs/1808.01740?context=astro-ph [2] https://arxiv.org/abs/hep-th/0103239 https://arxiv.org/abs/hep-th/0103239 [3] https://arxiv.org/abs/1707.07702 https://arxiv.org/abs/1707.07702

XorNot 4y ago
The point of publishing those articles is...to publish them though. Like, you have an idea, you write it up, submit it and note that it can fit known data but isn't currently testable. Done, and important. Maybe it goes nowhere, maybe it inspires someone, but the point of journals is in the name: they're journals of work in the field, shared so the community can explore and benefit from them. They're not publications of "what is definitely true", they are fundamentally explorations of what could be, or the more important "this is kind of interesting where could it lead?".
- tsimionescu 4y ago
  > note that it can fit known data but isn't currently testable. But the thing is, theories about the beginning of the universe will never be testable, they aren't just not currently testable. So, if your theory has no novel predictions about the future, but it adds extra parameters to obtain a different prediction about a past which exists beyond what can be measured, then you're wasting your time creating this theory, and wasting reviewers' and readers' time publishing it; and you're wasting money researching it. This is what Sabine usually writes and complains about - research money being spent on research that is at best unlikely to bear any fruit, and at worst navel-gazing, especially when there are very real problems in physics that are not receiving significant research. This is why she complains about people researching the beginning of the universe, or black hole entropy, or grand unified theories, or the hierarchy "problem", or looking for supersimmetry or for WIMPs in ever larger particle accelerators. Instead, she wishes more people were researching the measurement problem, non-linearity in quantum mechanics, high-energy physics through radio-telescopes instead of particle accelerators, to name a few things. Now, I don't know anywhere near enough to say that she is right, but I do believe she is not trivially wrong, like you seem to be suggesting.