How do people explore new ideas in physics?

[adam_snyder]

One of the biggest things I'm noticing about the physics community is that there are some serious barriers to entry. As a software engineer, i'm used to a much different style of community. Anyone can create and post open source code, and if it works well it gets adopted. People are generally interested in new ideas and are excited to try them out.

The exact opposite seems to be true in physics. I've tried a few times in different places to get feedback/discussion on the physics work I've done, and every time my posts get deleted and I get attacked for not being published in a peer reviewed journal. Reddit was the worst. I've reached out to people to try to get endorsed on arxiv, but that hasn't been working out either.

My paper isn't some crazy theory, it's a pure math refactoring of general relativity that reproduces the exact same results with different pedagogical understanding and has some real computational benefits. I hesitate to get into to much details for fear of being warned for self promotion.

Does anyone have any advice? Are there any physics communities that people know of that are more open to outsiders? I get that people are probably just tired of crackpots, but when people reject new things without even considering or looking at what they are, potential opportunities are lost. This seems like a major tragedy, especially for somthing as important as understanding the nature of reality itself.

 

[russ_watters]

One of the biggest things I'm noticing about the physics community is that there are some serious barriers to entry.

There's really no way around the fact that the barrier to entry (being educated in physics) is required for practical reasons. There's near zero chance that someone who isn't fully educated and up to date in the subject has anything meaningful to contribute. It would be like a non-doctor suggesting a new surgical technique. At best they'd be ignored by the medical community.

 

[.Scott]

In most cases, it's a LOT easier to test software than Physics.
If you have a better way to run a general relativity model, you don't need an endorsement, you need a collaborator - someone in the Physics community that needs to model space/time as part of their research.
If they can make good use of your work, you might be able to co-author something.

But .. I'll be interested in hearing other responses - since I am also an "outsider" - and a SW Engineer.

 

[fresh_42]

The noise-to-signal ratio is extremely high these days, where anyone can publish anything, for example, on academia.edu. Peer review is one of the ways to separate the wheat from the chaff. How should anybody know whether a read is reasonable or a waste of time?

Modern physics is highly specialized, and each branch has its own language. I recently looked up the details about an earthquake. It included technical information about its physical tensor and moments. I didn't understand a word. So if already an easily available information on the internet requires technical knowledge, how much more does a complex issue in a complex subfield of physics?

I'm not familiar with what physicists are usually confronted by laymen, only the nonsense I occasionally see here. However, I see the nonsense offered in mathematics. And you would think that mathematics is even easier: either something is false or it is true. No way. I can tell from pure sight whether it at least appears to be worth a closer look or not. People still think they have found an easy proof of a theorem with a highly sophisticated proof, only because they understood the statement. I even have a name for them: trisectionists, referring to the (provenly impossible) task of dividing an arbitrary angle into three equal parts by using straightedge and compass.

And physics is probably similar. It is, meanwhile, highly sophisticated, and you need to learn specific languages only to be able to discuss it on a scientific level. And people capable of speaking such a language are normally those who have access to scientific journals, or even arxiv.org.

 

[adam_snyder]

There's really no way around the fact that the barrier to entry (being educated in physics) is required for practical reasons. There's near zero chance that someone who isn't fully educated and up to date in the subject has anything meaningful to contribute. It would be like a non-doctor suggesting a new surgical technique. At best they'd be ignored by the medical community.

This is exactly the attitude I'm talking about. There is a "holier than thou" vibe. The idea that someone couldn't possibly have taught themselves is just objectively wrong. Some of the most important contributions in the history of physics came from people that were outsiders.

Michael Faraday had almost no formal mathematical education, worked as a bookbinder's apprentice. Self taught from books he was binding.

Einstein was a patent clerk, not in academia.

Oliver Heaviside was self taught, worked as a telegraph operator.

The patent clerk who revolutionized physics might not get past today's gatekeepers.

 

[.Scott]

My paper ... , it's a pure math refactoring of general relativity that reproduces the exact same results ... and has some real computational benefits. I hesitate to get into to much details for fear of being warned for self promotion.

Since you're a SW Engineer talking about "real computational benefits", I take that to mean reduced computer processor times.

So, don't get into full details. But do you have a specific case-in-point where you can compare processor time when computed the old way vs. processor time when computer your new way? Have you implemented that comparison, run it, and tabulated the results.

If so, I think a short summary of that would be of interest to this community. Like one paragraph on what physics problem you have chosen, another on the old way of coding a solution, another on your method, and finally a table showing the metrics - accuracy of results, processor times, and perhaps lines of code.

 

[Ibix]

There is a barrier to entry with computer code: does it compile/run and do what you claim in at least a couple of simple test cases. If you can't pass that barrier you'll be ignored.

The difference with physics is that someone else typically can't offload that basic validation of your idea on to a machine. That's why there are tests like peer review, which is basically "did you manage to convince someone your idea was at least half worthwhile". Unfortunately, as @fresh_42 has just noted, people who claim to be able to revolutionise physics are ten a penny so it's very hard to get a professional to look at an idea from a random - they know the odds of anything worthwhile coming from it are next to nothing.

Einstein was a patent clerk, not in academia.

Einstein had a doctorate under Lorentz, one of the foremost physicists of the day, and a solid publication record.

 

[PeroK]

One of the biggest things I'm noticing about the physics community is that there are some serious barriers to entry. As a software engineer, i'm used to a much different style of community. Anyone can create and post open source code, and if it works well it gets adopted. People are generally interested in new ideas and are excited to try them out.

There's no comparison between software development and the development of fundamental physical theories. The software you are posting, I imagine, is limited in scope. Anyone who is competent and motivated can develop software to do this or that.

Professional physics researchers, by and large, are working on a specific problem. Like predicting the gravitational wave pattern for a neutron star collision etc. They are generally not revolutionising physics or unifying GR with quantum mechanics or proving relativity wrong or any of that stuff. But, all the software engineers who post here (and there is no small number of them) are doing precisely that: they have a theory of everything that they've cooked up in less time than they would take to write a software system.

The exact opposite seems to be true in physics. I've tried a few times in different places to get feedback/discussion on the physics work I've done, and every time my posts get deleted and I get attacked for not being published in a peer reviewed journal. Reddit was the worst. I've reached out to people to try to get endorsed on arxiv, but that hasn't been working out either.

See above. We deleted about twenty crackpot threads a week on here. I alone report at least one a day!

My paper isn't some crazy theory, it's a pure math refactoring of general relativity that reproduces the exact same results with different pedagogical understanding and has some real computational benefits. I hesitate to get into to much details for fear of being warned for self promotion.

It's hard to believe that someone who isn't working extremely hard as a professional physicist or mathematician is going to produce something of genuine use or merit. As above, anyone can write a new piece of software, but producing a new, useful mathematical theorem or other advancement is a much greater task.

The forum has recently been discussing genuine new interpretations of QM (a thermal and a stochastic interpretation). These papers are accessible only to those with advanced knowledge and specialism.

Does anyone have any advice? Are there any physics communities that people know of that are more open to outsiders? I get that people are probably just tired of crackpots, but when people reject new things without even considering or looking at what they are, potential opportunities are lost. This seems like a major tragedy, especially for somthing as important as understanding the nature of reality itself.

There are seminars and conferences, but I'd be surprised if an amateur could follow much of the material. It's just too advanced.

Mathematics is worse. You can read a popular-science book about Andrew Wiles and Fermat's last theorem, but there are literally a handful of mathematicians around the world that can seriously assess the proof for correctness.

Mathematics and physics is a different ballgame from software development.

 

[pinball1970]

One of the biggest things I'm noticing about the physics community is that there are some serious barriers to entry. As a software engineer, i'm used to a much different style of community. Anyone can create and post open source code, and if it works well it gets adopted. People are generally interested in new ideas and are excited to try them out.

The exact opposite seems to be true in physics. I've tried a few times in different places to get feedback/discussion on the physics work I've done, and every time my posts get deleted and I get attacked for not being published in a peer reviewed journal. Reddit was the worst. I've reached out to people to try to get endorsed on arxiv, but that hasn't been working out either.

My paper isn't some crazy theory, it's a pure math refactoring of general relativity that reproduces the exact same results with different pedagogical understanding and has some real computational benefits. I hesitate to get into to much details for fear of being warned for self promotion.

Does anyone have any advice? Are there any physics communities that people know of that are more open to outsiders? I get that people are probably just tired of crackpots, but when people reject new things without even considering or looking at what they are, potential opportunities are lost. This seems like a major tragedy, especially for somthing as important as understanding the nature of reality itself.

I like the way you have framed this, you could be a crack pot but you are an articulate crack pot if your are!
PF is very well policed on personal theory people as they tend to be bad posters 99 times out of a hundred.
This keeps the noise to signal ratio low.
As @russ_watters said it will be extremely unlikely you have a theory or just a hypothesis that is correct, novel or relevant.
Physicists do this for a living, they teach it, they experiment, they explore new ideas every day.
Having said that, you made a decent post so, do you know any faculty guys?
Anyone with a physics UG or higher? At your uni?
Talk it through with them?
Im always bugging my science guys from uni with stuff.
I am not a scientist or physicist, so if I had an idea and wrote it up, I would probably think about submitting it to a journal.
One that is not on Beals list.
One author, not associated with a uni, no previous however might just get binned.
@fresh_42 what you think?

 

[russ_watters]

This is exactly the attitude I'm talking about. There is a "holier than thou" vibe.

It's not a "holier than thou attitude", it's a true reality.

The idea that someone couldn't possibly have taught themselves is just objectively wrong.

That's not what I said. I said "near zero". It's certainly possible (not exactly zero) but it is so low as to be a waste of physicists' time searching through thousands of bad papers looking for a good one that may not exist.

Some of the most important contributions in the history of physics came from people that were outsiders.

That's a myth, particularly regarding Einstein. He was a PhD, and it's kinda mind-blowing that you didn't know that.

Also note that when your only real example (Heaviside) is 100 years old, you're just proving how rare it is. And much before that, physics was pretty immature, incomplete and simple. So Faraday didn't have much to learn.

 

[PeroK]

This is exactly the attitude I'm talking about. There is a "holier than thou" vibe. The idea that someone couldn't possibly have taught themselves is just objectively wrong. Some of the most important contributions in the history of physics came from people that were outsiders.

Michael Faraday had almost no formal mathematical education, worked as a bookbinder's apprentice. Self taught from books he was binding.

Einstein was a patent clerk, not in academia.

Oliver Heaviside was self taught, worked as a telegraph operator.

The patent clerk who revolutionized physics might not get past today's gatekeepers.

And George Green was a miller from Nottingham who did ground-breaking work in mathematical physics in the early 1800's. That was 1800, and this is 2025. The amount of physics and mathematics that has been developed in the past 225 years is enormous. The work and theorems of that time are now part of the undergraduate syllabus. Groundbreaking experiments today are multi-million dollar extravagences. Gone are the days when you can look through a telescope and find a new planet. The amount of cosmological data is vast.

Re Einstein, quite the opposite is true. He had to take a job as a patent clerk because no paid university posts were available. Today, he would probably get a post-doctorate position. Academia has opened up to people like Einstein, who were not from wealthier backgrounds and needed a salary. It was the social conditions in those days that restricted academia to those with independent wealth.

 

[martinbn]

If you have done some work, which you think is worth sharing, and have written your paper, then send it to a physics journal. You will get feedback.

 

[fresh_42]

One author, not associated with a uni, no previous however might just get binned.
@fresh_42 what you think?

Yes, but this isn't new. Fermat, Kafka, Galois have all become famous after their deaths as others published what they had left. But how are the chances of binning a new Galois?

I don't see the problem on the publishing side. As mentioned, a publication is easy. The difficulties are on the readers' side. Someone who publishes wants to be read. Now, here is the crux: they don't want to be read by other crackpots, but by serious scientists. By what justification? Every such publisher claims that his work is revolutionary, and even if you are kind and bored, you simply cannot read all of them, disregarding the fact that most of them are extremely painful to read. So, how to convince an intended reader? That is the point, not the publication. The time of a scientist is extremely valuable. You basically have only around 20 years for your own research. Too young and you don't know enough, too old and you risk becoming a tragic figure (Zuse, Atiyah, Nash), or the least, your productivity falls low. Hence, you are forced to manage your time efficiently, and I'm afraid reading other people's papers with a 99% chance of wasting time on nonsense won't make it on the list.

My mentor once told me about a dialogue with his former mentor:
"How many books (novels) do you read a year? Let's say 10. How old are you? Do the multiplication, and choose wisely!"

 

[adam_snyder]

And George Green was a miller from Nottingham who did ground-breaking work in mathematical physics in the early 1800's. That was 1800, and this is 2025. The amount of physics and mathematics that has been developed in the past 225 years is enormous. The work and theorems of that time are now part of the undergraduate syllabus. Groundbreaking experiments today are multi-million dollar extravagences. Gone are the days when you can look through a telescope and find a new planet. The amount of cosmological data is vast.

Re Einstein, quite the opposite is true. He had to take a job as a patent clerk because no paid university posts were available. Today, he would probably get a post-doctorate position. Academia has opened up to people like Einstein, who were not from wealthier backgrounds and needed a salary. It was the social conditions in those days that restricted academia to those with independent wealth.

And we have much more powerful computers and tools to help us learn as well... Taught myself a crash course in quantum mechanics with 11 hour youtube videos like this:

 

[adam_snyder]

I like the way you have framed this, you could be a crack pot but you are an articulate crack pot if your are!
PF is very well policed on personal theory people as they tend to be bad posters 99 times out of a hundred.
This keeps the noise to signal ratio low.
As @russ_watters said it will be extremely unlikely you have a theory or just a hypothesis that is correct, novel or relevant.
Physicists do this for a living, they teach it, they experiment, they explore new ideas every day.
Having said that, you made a decent post so, do you know any faculty guys?
Anyone with a physics UG or higher? At your uni?
Talk it through with them?
Im always bugging my science guys from uni with stuff.
I am not a scientist or physicist, so if I had an idea and wrote it up, I would probably think about submitting it to a journal.
One that is not on Beals list.
One author, not associated with a uni, no previous however might just get binned.
@fresh_42 what you think?

LOL. I didn't go to college, so I'm totally outside of the "normal". I don't have any connections. I'm fully self taught...

Maybe I am a crackpot. Sometimes I wonder... But math doesn't really lie. I would love for someone to review my work and tell me if I should abandon my hobby!

 

[adam_snyder]

If you have done some work, which you think is worth sharing, and have written your paper, then send it to a physics journal. You will get feedback.

Oh, I can just do that? Do you have a recommendation for a journal that would be interested in a full refactoring of GR math?

 

[martinbn]

Oh, I can just do that? Do you have a recommendation for a journal that would be interested in a full refactoring of GR math?

What do you mean by refactoring? What journals are the papers you have read published in?

 

[Bandersnatch]

Einstein had a doctorate under Lorentz

I don't think that's correct. It was Kleiner.

 

[.Scott]

What do you mean by refactoring? What journals are the papers you have read published in?

Refactoring is a computer engineering term.

Physicist deal with equations. Those equations can be stated or restated in different terms to make a different point in an argument.

But when it is necessary to perform specific calculations - to implement a Physics model on a computer - then those equation transition from "declarations" to "imperatives" that direct the operations of the computer. What becomes important is not whether a point is made - but what kind and amount of processor resources will be needed to do the computations.

So, instead of changing the terms to make a point, you change the terms so that you avoid recalculating the same thing over and over again. Or you arrange things so that you can control the precision economically.

A great example of this kind of "refactoring" is the FFT. If you start with the Fourier transform and just code it up directly as written, you will not be making efficient use of your available computer resources.
J W Cooley and John Tukey refactored the DFT problem and published their method in 1965. It has made a huge difference.

 

[Ibix]

I don't think that's correct. It was Kleiner.

I stand corrected. Looks like Lorentz did teach him, but was not his PhD advisor.

 

[fresh_42]

I stand corrected. Looks like Lorentz did teach him, but was not his PhD advisor.

https://www.research-collection.eth.../fc0b3f1c-59c6-4e0c-b2cc-e9a84abc0e25/content

 

[PeroK]

And we have much more powerful computers and tools to help us learn as well... Taught myself a crash course in quantum mechanics with 11 hour youtube videos like this:

Perhaps you could pass an undergraduate QM exam, but there is a gulf between that and doing new research.

 

[.Scott]

Oh, I can just do that? Do you have a recommendation for a journal that would be interested in a full refactoring of GR math?

I have no experience in this, but this seems to be in the ball park:
https://iopscience.iop.org/journal/1347-4065/page/submission-options

It's the "Japanese Journal of Applied Physics" - but its published in English and has wide distribution.

 

[Haborix]

I think OP could have a more productive discussion by posting in the relativity subforum using the homework template. It really sounds like OP has manipulated some GR equations for a Schwarzschild black hole and then solved them numerically. The claim doesn't seem to be that the algorithm for the numerical solution is faster but just that the PDE starting point is somehow advantageous.

 

[.Scott]

I think OP could have a more productive discussion by posting in the relativity subforum using the homework template. It really sounds like OP has manipulated some GR equations for a Schwarzschild black hole and then solved them numerically. The claim doesn't seem to be that the algorithm for the numerical solution is faster but just that the PDE starting point is somehow advantageous.

Yes to the "manipulated part". No to the "faster" part - that is, his algorithm is faster.

He is claiming that of the two software implementations of this problem, the second method is hugely faster than the "traditional" method. And, although I have not tried to run his code, a code inspection suggests that it most certainly should be faster.

So, the remaining questions are:
1) Is anyone actually using the "traditional method".
2) Are the two methods really (or practically) equivalent?

He can demonstrate the answer to that second question with either changes to his test procedure and how he describes it (as I explained in an earlier post). Although, it would be good if one of our Forum Physicists could verify it - because they might have better in sight into whether this could be applied to rotating BH and such.

 

[Haborix]

I'm not really convinced that what is being called the traditional method isn't starting from a different set of PDEs. In other words, I think the numerics are getting confused with the equation selection and manipulation.

[EDIT] To try and make my point clearer. Say I come up with two formulations of the same physical problem and apply Runge-Kutta to numerically solve each. If one formulation gets solved more quickly than the other, I don't attribute that difference to the numerics---both formulations use the same numerical method.

 

[jbergman]

One of the biggest things I'm noticing about the physics community is that there are some serious barriers to entry. As a software engineer, i'm used to a much different style of community. Anyone can create and post open source code, and if it works well it gets adopted. People are generally interested in new ideas and are excited to try them out.

The exact opposite seems to be true in physics. I've tried a few times in different places to get feedback/discussion on the physics work I've done, and every time my posts get deleted and I get attacked for not being published in a peer reviewed journal. Reddit was the worst. I've reached out to people to try to get endorsed on arxiv, but that hasn't been working out either.

My paper isn't some crazy theory, it's a pure math refactoring of general relativity that reproduces the exact same results with different pedagogical understanding and has some real computational benefits. I hesitate to get into to much details for fear of being warned for self promotion.

Does anyone have any advice? Are there any physics communities that people know of that are more open to outsiders? I get that people are probably just tired of crackpots, but when people reject new things without even considering or looking at what they are, potential opportunities are lost. This seems like a major tragedy, especially for somthing as important as understanding the nature of reality itself.

I don't understand. Why can't you just post your work on a blog or github or YouTube? I have done this before and have good results. What specifically are you looking for?

 

[PAllen]

This is exactly the attitude I'm talking about. There is a "holier than thou" vibe. The idea that someone couldn't possibly have taught themselves is just objectively wrong. Some of the most important contributions in the history of physics came from people that were outsiders.

Michael Faraday had almost no formal mathematical education, worked as a bookbinder's apprentice. Self taught from books he was binding.

Wrong. He did a long apprenticeship with Sir Humphrey Davy (president of the Royal Society, world famous chemist) before his work on EM. Also, the field was young and undeveloped compared to physics fields today.

Einstein was a patent clerk, not in academia.

Wrong again. He graduated from a top physics school, and was working in a patent office temporarily while finalizing his doctoral thesis. He was already affiliated with Annalen der Physik, a leading journal of the time.

Oliver Heaviside was self taught, worked as a telegraph operator.

Don't know this history, but your track record above leaves me thinking there is a lot more to the story.

The patent clerk who revolutionized physics might not get past today's gatekeepers.

 

[Nugatory]

Einstein was a patent clerk, not in academia.

However he also had been through the most rigorous physics program available at the the time, had completed his PhD, was a regular correspondent with many of the top physicists of the era and was familiar with the work that had already been done by Lorentz, Fitzgerald, Poincaré towards the problem that special relativity finally solved.

Einstein was not a lone genius slaving away in isolation to find a truth that had escaped the academic community. Instead, his early career confirmed that you cannot advance the frontiers of knowledge unless you know where the frontier is and have gotten yourself there.

 

[PeterDonis]

I would love for someone to review my work and tell me if I should abandon my hobby!

Unfortunately we don't do that here. Please note that your post showing some of your work, and a number of posts in response to it, have been deleted. They're off topic, both because we don't review personal research here, and because this is the GD forum, not one of the physics forums, which is where actual physics questions belong.