4 ms·

Given the lack of testable predictions and failure in experiments so far, string theory should be called string hypothesis.

onli 3y ago
If there are no testable predictions it is not a hypothesis, it's a religion.
HappyPanacea 3y ago
(I'm also not a theoretical physicist). Not entirely true, I was looking at proton decay recently and one model of string theory model apparently predicts 10^36 yr lifetime [1] which is within experimental reach. [1]: Igor R Klebanov, Edward Witten, Proton decay in intersecting D-brane models, Nuclear Physics B, Volume 664, Issues 1–2, 2003, Pages 3-20, ISSN 0550-3213, https://doi.org/10.1016/S0550-3213(03)00410-3 https://doi.org/10.1016/S0550-3213(03)00410-3. (https://www.sciencedirect.com/science/article/pii/S0550321303004103 https://www.sciencedirect.com/science/article/pii/S055032130...)
- ogogmad 3y ago
  One model? So there may be models where it doesn't happen. Or perhaps that very same model makes other, verifiably wrong predictions.
- CuriouslyC 3y ago
  If there are an infinite family of parameterized string theory models, it's basically guaranteed that you can pick one that predicts anomalous data. In other sciences you would adjust the P value for the probability that you can reject the null hypothesis given the number of experiments you're performing, which in the case of infinity is going to drop the probability of rejecting the null to zero.
- graycat 3y ago
  > 10^36 yr lifetime [1] which is within experimental reach. For this might want to assume that for each proton its lifetime is a random variable that obeys a Poisson process. That's asking a LOT, might be seen as putting some severe constraints on what is going on with a proton and its internals.
gnfargbl 3y ago
Well, String Phenomenology is a thing [1]... although more recently they appear to have to switched away from studying the frankly gargantuan set of viable vacua in favour of examining the even larger set of non-viable models [2]. And the choice to take research in that direction is just somehow just so quintessentially string theorist. [1] https://en.wikipedia.org/wiki/String_phenomenology https://en.wikipedia.org/wiki/String_phenomenology [2] https://en.wikipedia.org/wiki/Swampland_(physics) https://en.wikipedia.org/wiki/Swampland_(physics)
Nevermark 3y ago
Given the sociological aspects how about: String game theory?

onli 3y ago

If there are no testable predictions it is not a hypothesis, it's a religion.

HappyPanacea 3y ago

(I'm also not a theoretical physicist). Not entirely true, I was looking at proton decay recently and one model of string theory model apparently predicts 10^36 yr lifetime [1] which is within experimental reach. [1]: Igor R Klebanov, Edward Witten, Proton decay in intersecting D-brane models, Nuclear Physics B, Volume 664, Issues 1–2, 2003, Pages 3-20, ISSN 0550-3213, https://doi.org/10.1016/S0550-3213(03)00410-3 https://doi.org/10.1016/S0550-3213(03)00410-3. (https://www.sciencedirect.com/science/article/pii/S0550321303004103 https://www.sciencedirect.com/science/article/pii/S055032130...)

ogogmad 3y ago
One model? So there may be models where it doesn't happen. Or perhaps that very same model makes other, verifiably wrong predictions.
CuriouslyC 3y ago
If there are an infinite family of parameterized string theory models, it's basically guaranteed that you can pick one that predicts anomalous data. In other sciences you would adjust the P value for the probability that you can reject the null hypothesis given the number of experiments you're performing, which in the case of infinity is going to drop the probability of rejecting the null to zero.
graycat 3y ago
> 10^36 yr lifetime [1] which is within experimental reach. For this might want to assume that for each proton its lifetime is a random variable that obeys a Poisson process. That's asking a LOT, might be seen as putting some severe constraints on what is going on with a proton and its internals.

gnfargbl 3y ago

Well, String Phenomenology is a thing [1]... although more recently they appear to have to switched away from studying the frankly gargantuan set of viable vacua in favour of examining the even larger set of non-viable models [2]. And the choice to take research in that direction is just somehow just so quintessentially string theorist. [1] https://en.wikipedia.org/wiki/String_phenomenology https://en.wikipedia.org/wiki/String_phenomenology [2] https://en.wikipedia.org/wiki/Swampland_(physics) https://en.wikipedia.org/wiki/Swampland_(physics)

Nevermark 3y ago

Given the sociological aspects how about: String game theory?