What kind of and how much non-mainstream and off topic discussions should PhysicsOverflow tolerate?

Question

PhysicsOverflow is meant to be a rather professional (knowledgeable enough enthusiasts are welcome too IMHO) higher-level physics site, where graduate-level upward physics and related maths questions can be asked and answered in the Q&A section, and research-papers can be refereed and discussed in the Reviews section.

Concerning the Q&A section, it is rather clear what this means, and the level of questions can be maintained by means of community moderation.

For the Reviews section, the case is less clear cut. An important question that we probably should discuss again, is how much and what kind of non-mainstream physics we want to allow there. See for example here, and here for earlier discussions.

Related to the issue of non-mainstream (by the real world community of physicists not accepted) physics is the question of legitimate sources for submissions in the Reviews section. Is it really a good idea to (generally and unfiltered) include Vixra?

More generally, what standard or lower-level cutoff do we want to apply to the reviews section?

I personally think that we should by all means avoid that PhysicsOverflow looks like an unmoderated forum where cranks and crackpots can freely propagate and use the site as a platform to promote their personal non-mainstream theories and attack legitimate physics at the same time. Otherwise it might happen that good people already on PhysicsOverflow are driven away, and new nice knowledgeable people we would like to see here are prevented from joining.

To be honest, the fact that the "lively" but not really constructive, not really enlightening and at times even very off-topic discussions about non-mainstream physics, driven by always the same people, that pushed discussions about mainstream physics away the last few days, make me feel very uneasy, because I personally think that this is not what PhysicsOverflow is intended to and it makes the site look not so good.

How do other people think about this?

Comments

I have just started a list of types of discussions together with what I personally think about if they should be welcome on PhysicsOverflow or not for people to vote on. This is just meant to find out what the PhysicsOverflow community appreciates. Not sure if we better open a new thread to discuss what standard should be applied to the submissions we welcome.

---

This might be a good inspiration for those who want to contribute to the list ... ;-)

---

To me, pushing one's pet theory etc. into every suitable thread is too much. It won't do any good for the writer or readers. 

---

I had written an answer but it unfortunately disappeared due to some strange bug with my browser : ( I'll need to rewrite the answer but I do not remember the content of the original. 

I agree it is necessary to have some sort of a filter for papers submitted to PhysicsOverflow. It is ridiculous that someone who isn't even able to get their question remain open on PhysicsOverflow is able to submit a paper to PhysicsOverflow. It is important to ensure that the filter is completely open though, i.e. no preliminary closed reviews of submitted papers. 

I've thought of a number of possible filters but they are not really appropriate, as they appeal to the knowledge of the author etc. The finally came up with the following criterion: 

If the paper contradicts well-established physics, and is not in the scope of any mainstream physical theory, it must be supported by graduate-level mathematics. 

 I realise this (the part on graduate-level mathematics) sounds foolish, but it really sieves out a lot of bullshit. 

Answer 1

Bringing up repeatedly the same non-mainstream point of view in different threads where it is off topic, is not welcome.

Comments

I think the only censorship policy we should have here is non-duplication, so a link to the material is still ok.

---

@RonMaimon What about situations in which users try to force their research, both the content and links, into every thread they consider even remotely suitable. It's not just a non-duplication policy that applies here, also a stay-on-topic policy.

---

Deal with that problem when it comes, not proactively. Rules for hypothetical future situations are not useful.

---

Ron, the problem is already here (not sure if you have followed the discussions the last week or so) ...

---

The problem is not here, it's just in your head. The crackpot paper was dealt with fine. I am in Israel, with limited time, but I have dropped by three times a week, and I saw the stuff. It's ok, don't panic. We just need an option to order the submissions by score, or just to remove the bumping feature for comments, and list submissions by date.

---

Yes, removing the bumping feature of comments and listing submissions by date might be very helpful.

Answer 2

Claiming that legitimate physics is just another religion, philosophy, etc is not welcome.

Comments

I agree that philosophy is science too, but what I mean here is that the term should not be abused to insult legitimate physics/physicists. So people try to use maths in the same way, by claiming that legitimate physics has nothing to do with the "real world", because it is pure maths ;-)...

Answer 3

Claims that (long-time) established physics knowledge is wrong (when applied inside its intended domain of validity) without being able to back these claims up by stringent experimental or theoretical reasons, is not welcome.

Comments

This is not a good idea, as any new physics comes with a claim that established physics knowledge is wrong, with very little to back up the claim. The claim of parity violation was initially based on anomalous meson decays, a very meager result to completely overturn 300 years of automatic assumption of parity invariance. But it was correct. There is no reason to make anything at all unwelcome. This is creeping politics.

---

@RonMaimon Then the "stringent" clause can be removed. It isn't "creeping politics", everything is transparent, and public. I don't know if you've seen the inflow of submissions that claim to "present TOEs" and overturn renormalisation, and so on. It is absolutely necessary at this point to decline such submissions, so that we don't end up looking unmoderated, even if the submissions are downvoted.

---

It is not necessary at all. Just review them properly, the submissions are generally fine, and the one crackpot submission I saw is downvoted to minus infinity already. Kalitvianski's submission, while containing wacky opinions on renormalization, presents an interesting model. You must understand where Vladimir is coming from: he is worried about infrared divergences and the self-field problem, and this problem is completely different from the ultraviolet problem and the running coupling. He tends to conflate the two issues, but so what. His papers are not crackpot, and his opinions, while not mainstream, are not unreasonable. They are just wrong regarding ultraviolet renormalization (but they are correct regarding infrared issues). This is why his papers get reviewed and published in journals, and it is ridiculous that you should attack him for placing them here.

We must DECLINE NO SUBMISSIONS. Period. Declining submissions for review is mentally retarded, just review them negatively if they are bad, and let them get downvoted. The voting takes care of these issues, and so far it's ok.

---

Ron, can you then write a review for him and post it to his submission? It seems you could do it much easier and faster than I. I dont necessarily think that his physics model is bad or wrong BTW and his submission is perfectly valid.

Answer 4

Presenting consistent simplified (compared to some standard procedure) "alternative" ways of calculating quantities of interest, that work in specific cases, under well defined conditions, or in certain limitations, is welcome.

Answer 5

If a paper claims to suddenly be the final "Theory of Everything", it needs to be able to make a few predecided predictions (to be decided). 

Comments

Just put the paper up and let it get downvoted to oblivion. It is likely highly original, and the deductions are likely inaccurate.

---

@RonMaimon It can't go into "oblivion", anyone can find it on the reviews category page, and on the main page when it's bumped. It really is an instant crackpot alarm.

---

Just order the views by score, or provide the option to do so, or remove bumping on the reviews section.

---

@RonMaimon It's difficult when there are so few submissions, and the majority are crackpot.

One of the advantages of launching the reviews section early was to portray a high level to viewers, but this way, it's getting counterproductive, with these crackpot submissions, we look like a crackpot forum.

---

We don't look like a crackpot forum, we look like an open, uncensored, forum. That means that people don't have to worry about someone getting inbetween their submission and the referees.

There is a huge difference between an open forum and a crackpot forum. The majority of our submissions are perfectly legitimate, and if you are worried, just import three legitimate arxiv papers for each crackpot paper, at random. There are many many more legitimate papers than crackpot papers.

Open review makes crackpot writing a waste of time, because it is a lot of effort to write such a thing, and it takes one minute to downvote it or review it negatively. Just let the process work. Making a forum that is censorship free is difficult, precisely because pressure begins to mount to institute censorship right away. All you have to do is say "NO", and insist on it, and problem solved.

Answer 6

Discussing not (yet) established physics in the context of a submission or well defined serious (not rhetorical) question is welcome.

Comments

Is it necessary to state what is welcome? 

---

It might help alleviate accusations of bad censorhip, if they come up ...

Answer 7

Calling physicists who do by the real-world community accepted mainstream physics "the establishment" and similar names is not welcome.

Comments

While I do often appeal to "crackpot signs", I'm against "officially" discriminating based on the usage of such terms. 

---

No, it is a quite accurate crackpot alarm. I really think we should not let the site be taken over by crackpots or PhysicOverflow will soon look like a crackpot forum and we will lose or never aquire the good people we like to see here :-/

---

If we allow crackpots to dominate the site and misbehave and attack good legitimate physicists/physics as they do in not so great physics blogs, below popular news articles etc ...
 

On MathOverflow for example they do not tolerate this but maintain a professional tone and atmosphere, and this is why it is successful and are even acknowledged by the American Mathematical Society. Of course I agree that no physics discussions as such should be censored out ...

---

Ok I know that you are not omniscient either ... :-)

---

@dimension10: The question of relative transcendence asks whether value a is related to value b by a polynomial equation with integer coefficients. Most values a and b are unrelated, because there are only countably many such integer polynomials. But a model of set theory is also countable, so there are only countably many differentiable functions you can write down, so most values a and b (in the platonic sense) are not only not polynomially related, b does not equal f(a) for any real valued function f you can write down in any given model of set theory. This is the "ultimate transcendence".

For the case of generic values, it is easy to see that there is such an ultimate transcendence, but for computable values, it is quite difficult to comprehend what to take this to mean, if anything, because when you are talking about a computable real number, it is automatically in the model, because the program that computes the number is in the model. So it seems you need to consider non-computable values to define differentiation with respect to constants.

But is that really true? I am not sure. When you are talking about computing zeta(2) from a pi oracle (a quick source for digits of pi), there is a straightforward program to get zeta(2) from pi which is just squaring pi and dividing by 6. When you talk about computing zeta(3), the pi-oracle is no help. So when you ask about varying the pi-oracle at large positions, zeta(2) would change, meaning the best algorithm for computing it would depend on the large-positions of the oracle, while the value of zeta(3) wouldn't change.

It's a difficult question, and it touches on the important issues in logic. But it superficially appears crackpot. I used it as an test of how censorship works on mathoverflow (on purpose, I assumed it wouldn't be well recieved--- to my surprise it had more upvotes than downvotes when it was deleted).

Regarding the crackpot issue, I can see why you might want some sort of censorship, because there is the problem of a hypothetical insincerity attack, where people flood the site with crackpot material, simply to drown out the signal with noise. But I don't see a single submission as a problem of this sort, especially when you can easily make it go to the bottom of the list quickly.

---

Um, @RonMaimon, Isn't the meaning of the statement "zeta(2) depends on pi" just "\(\zeta(2)=R\pi^r\) where R and r are rational numbers"? Why would you ever need to "differentiate with respect to a constant"? A constant is a constant, it doesn't vary.

I can see why MathOverflow deleted the question, it isn't research-level. 

Edit: Edited the rational bit, you're right, it's not a rational multiple of pi.

---

@dimension10: you can define forcing models of set theory--- this means you introduce symbols into the model which are defined by a sequence of statements: "x is a real number", "the integer part of of x is 3", "the first digit of x after the decimal is 1", "the second digit is 4", and so on. Introducing these axioms "forces" a new symbol into your model whose numerical value is eventually pi, but the axioms do not know this. This new symbol can be substituted for pi, but the axioms will never prove the statement "x equals pi", just "x minus pi is less than .1", "x minus pi is less than .01", and so on. This symbol can be considered "infinitesimally different from pi", and you can ask whether there is some sort of automorphism which allows you to substitute pi for x, and still keep all the relations in the new model the same. This question is definitely research level, and it is exactly the question of differentiating with respect to pi. It just sounds like nonsense to someone who doesn't understand forcing.

An example of a quantity which "depends on pi" but is not a rational multiple of pi is sqrt(pi). An example of a constant which  doesn't depend on pi is exp(pi), because exp(i pi) is 1. So, because pi has a special property, namely that it is the period of the exponential function, the question is not simple--- there might be no way to do anything of the sort, because the relations that pi satisfies are too tight

But for a "generic" number, a random number added at random to the model, a traditional forcing, then it is definitely possible to differentiate with respect to the symbol, because you can introduce infinitesimally different generic numbers with no problem, and there are no special relations which the thing obeys. It's not a simple question, despite the superficial impression it gives of being crackpot, or not research level. This is always the problem with such things.

I will point out that Andreas Blass, a good logician, came out and said "No, I don't think so", while Joel Hamkins, another good logician came out and said "Maybe yes, perhaps we can jiggle this to make sense". I didn't know the answer, I was asking because I think it is important. People don't think of forcing in this way anymore, and Paul Cohen just died, and I wanted a question in the spirit of his original work (which also looked crazy at first, although of course, it was much better than my stuff).

Answer 8

Denial of established physics without mathematical or experimental justification is not welcome.

Comments

@VladimirKalitvianski Ok, your post calls well-established physics wrong without justification. Now all of the world's problems are solved, right?

---

I have edited this post a bit to make it more objective, as "personal philosophy" is a little too specific.

---

Hi Vladimir, Arnold has more than once and at length explained to you what is wrong or at least missing with your point of view. There is no point of bringing it up again and again and again at repeatedly in different posts on PhysicsOverflow ...

Below this answer, it would be off topic anyway.

---

@Dilaton No, into the section about the reviews section. I think this is not really about "unprofessional behaviour", but about using PhysicsOverflow as a platform to promote pet theories, which it is not intended as. 

---

@VladimirKalitvianski The real issue behind this thread, or at least the real issue for me, is not your papers, but Kimmo Rouvari's. Your papers are a notch above crackpot, they're unnecessary, and arise from your misunderstanding of how renormalisation works, but they're still acceptable according to me and the proposed criteria in most of these answers.

My real problem with you is your habit of repeatedly posting comments on marginally relevant threads about reofrmulation and how you are being "abused".

---

@dimension10 I'm puzzled why you consider my paper some kind of an issue. It's just a pet theory, that's all. 

---

@Dimension10 maybe the results of this thread (when equilibrated and settled) can go into the FAQ under a new title "Unprofessional behavior which is not welcome/tolerated  on PhysicsOverflow" or something similar? As you said in a comment, what is welcome is probably less important to state ...

Answer 9

No appeals to religions and other outdated philosophies. 

Comments

This is not necessary, as religion never comes up. Philosophy is pretty much settled in physics, it's positivism since Mach, and still is, so there is no dispute. Any "no philosophy" rule tends to clobber the people interested in what quantum mechanics is all about, and I think you want to hear from people like Bohm, even when he speaks in ill-defined terms of "wholeness" and "implicate order" and talks about mysticism (as he often did in later years). There is no reason for this, it is creeping politics. If it's crap, it will get downvoted, no need for a special thought-police.

---

@RonMaimon That's why I said outdated philosophies. Looking at the kind of submissions we're getting, about "overturning renormalisation", and "presenting theories of everything", we need to have a clear rejection policy so that we don't pollute the reviews section with bullshit.

---

@dimension10: According to G. Scharf's book "Finite QED", he is also "overturning renormalisation". Would you call it a bullshit?

Note, I have nothing against effective theories where renormalization is more a calculation rather than just getting rid of undesirable corrections. But I am against extrapolation of the effective field ideology to all QFTs.

---

@dimension10: don't get scared, and don't give in to temptation. Open review without censorship works, let it work. The moment you put your thumb on the scale, the scale tips over.

---

I object to call religions outdated philosophies. Far more people adhere to religions than to science.

---

Oh yes ...

The idea behind this point here is that philosophy, religions, etc should not be abused to disparage legitimate physics (to claim that it is not science etc) ...