Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

He didn't discover the Higgs, but he formulated it (along with other people) in 1964. Its discovery was not until 2012.


Discovery can happen in a purely mathematical/theoretical context, too.


I think it's fair to say the 'discovery' that this boson exists came with the LHC experiments. But Higgs did discover in 1964 that the Higgs boson could explain why particles have mass. His paper couldn't say "this is definitely the way the universe is", but rather "if the universe plays by the rules we think it does, this is a relatively simple way to explain this thing we see".

And in my mind, both of those achievements are awesome.


It's definitely semantics and I'm not sure it is worth arguing if we understand what one another means. Unless we're clarifying.

But I do think it is good to discuss the role that others have played in the discovery. After receiving the 2017 Kip Thorne said (Weiss and Barish were the other two)

  It is unfortunate that, due to the statutes of the Nobel Foundation, the prize has to go to no more than three people, when our marvelous discovery is the work of more than a thousand.
Even discoveries of the past were due to the efforts of many. Einstein's shoulder's of giants, which I jokingly refer to as "3 scientists in a trench coat, all the way down". But with modern science, the problems are more difficult and the effort to make these breakthroughs more clearly depends upon the work of hundreds or thousands. It is good to promote these ideas and this recognition. I think it can also help motivate us to better work together, and not letting us think we are a lowly unimportant cog in a giant machine. Because while we may be small cogs, our work is still important (even if not as important as others).


Exactly. Give the countless engineers behind LHC some of the credit as well!


I still wouldn’t use the word “discover” in this context. Physics is not pure mathematics. There are many theories that just do not correspond to reality. Would you say that strings and extra dimensions have been discovered? No, we’d say they’ve been predicted by a certain theory, but not discovered yet — the prediction may not even be right!


Precisely, he predicted it.


[flagged]


It sounds like this comes from a misunderstanding of probability and how it applies to physics. Can you explain what you mean?



The same is true of every other boson, except arguably photons in the classical limit (e.g. you can put a multimeter on an antenna and detect them "non-probabalistically"). The Higgs detection is sound. A quick check of wikipedia says 5.9 sigma. Good enough for me, anyway.


> e.g. you can put a multimeter on an antenna and detect them "non-probabalistically"

I think I get what you mean, but we should be clear—every measurement is probabilistic, right? We can measure photons with low enough inaccuracy that we don’t bother quantifying the probability that they might not exist. But, it is actually on some philosophical level a difference of degree, not type. I’m pretty sure.


What does it mean for a measurement to be probabilistic?

I'm a math grad student, so still a layman, but here's the way I think about it: Every particle travels over every possible path as the wave spreads out. Once you measure the particle, it tells you a path.

So the measured value of the particle is probabilistic, not the measurement.


I don't know where you get 5.9 sigma but that might have been the confidence at the time it was discovered in 2012. Since then the LHC took a lot more data and the confidence is more like 20 sigma.

That said, these days it's harder to find a paper that gives a confidence for discovery, since they are all quoting the signal strength with respect to the standard model prediction. They generally measure this within about 5%, see for example

https://arxiv.org/abs/2404.05498


Any evidence of any physical phenomena is probabilistic. And with Bell's inequalities, we know that for a fact.


Any evidence of any physical phenomena is probablystic? Maybe this is true of you go deep enough into mathematical/philosophical/quantum rabbit holes, but it doesn't pass the intuitive smell test - I could off-hand name tens of physical phenomena that will perform as expected 100% of the time, deterministic physics (on the macro scale, at least) are like the entire basis for the scientific method.


You may think you have 100% probability of event but then simulation glitches and outcome is different.


Ok but that hasn't been scientifically observed, so it's more of a religion than a theory.


This doesn't follow from Bell's inequalities. There are deterministic models of quantum mechanics which are compatible with Bell's theorem.


That's why I argue with people on the internet — to be corrected and learn new things in the process!

Please tell more.


Bells theorem says that he predictions of quantum mechanics are incompatible with any "local hidden variable model". Experiments show that the real world appears to follow the predictions of quantum mechanics, so no local hidden variable model can explain what we see in these experiments.

Therefore all you need to understand is what a "local hidden variable model" is, and what the lack of one would mean.

You should think of doing measurements in quantum mechanics as something like flipping a biased coin, or rolling a biased dice. Every time you do a measurement you get some outcome (maybe heads or tails or a number from 1 to 6) but the probabilities aren't generally the same, they're biased in a way which tells you something about the system. They can even be biased so much that the outcomes are deterministic (0 and 1 are still probabilities).

When you flip a coin or roll a dice, the outcomes are random from your point of view, but there isn't really any fundamental randomness going on. There's a bunch of data about how your hand moves, and the local wind speed and pressure and the physical characteristics of the coin which if you knew them then you'd be able to predict the outcome with certainty. People have even been able to build robots which can reliably flip coins to always get a desired outcome.

This "extra information" that makes the outcome deterministic if you know it we could call hidden variables. Its "hidden" because you don't generally know it.

If Bell's theorem said that there was no hidden variables model consistent with the predictions of quantum mechanics we'd be done now. We would have decided that quantum mechanics is fundamentally probabilistic and no extra hidden information could exist to make it deterministic. But Bell emphatically doesn't do that. The theorem says there is no local hidden variable model consistent with quantum mechanics. So what does "local" mean here? For that you can imagine that instead of rolling one dice I'm rolling one here, you're rolling a second one wherever you live, our friend on Mars is rolling a third one, etc. A hidden variable is "local" if it only affects the dice in once location.

Bell's theorem says that if you want to have a hidden variable model that reproduces quantum mechanical predictions then it needs to be nom-local. That is the hidden variable has to affect the experiment here, and the one with you, and the one on Mars all at the same time. We consider that non-local hidden variables are unlikely because they're essentially magic, they need to affect things arbitrarily far away from each other all at the same time. Amongst other things this is wildly on conflict with special relativity.


> We consider that non-local hidden variables are unlikely

Doesn't that essentially mean the same as

> If Bell's theorem said that there was no hidden variables model consistent with the predictions of quantum mechanics we'd be done now.

?


Unlikely does not mean the same as impossible, so no.

There are also some "loopholes" in the story I told above. For example the many worlds interpretation of quantum mechanics is completely local and deterministic, but weird enough that the standard formulation of hidden variables models doesn't apply to it.


> We know that for a fact

Probably.


The only true facts are the useless tautologies... Sad for the theory of knowledge, truly...


"Probabilistic evidence" != "no evidence".

> The error rate is very high.

Citation needed. More citation needed to show that it's high enough that it invalidates the evidence.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: