How the ArXiv Decides What’s Science | Hacker News Reader

How the ArXiv Decides What’s Science

backreaction.blogspot.com

78 pointsyetanotheracc24 comments

The basic heuristic I use for determining what 'counts' as science relies on 3 questions

1) Are there methods or is there a description of how to observe something?

2) Are old methods being applied to a new phenomenon.

3) Are new methods being applied to an old phenomenon.

It isn't possible to do science if you don't have a viable tool (method) for measuring a phenomenon of interest or if you can't describe the phenomenon you want to study in a way that others can observe it.

Cases like Mochizuki with the ABC conjecture are harder, but at least the ABC conjecture is a know quantity.

It is reassuring that classifying papers by category of research works fairly well. Both phenomena and methodology are described using specialized vocabulary, and mismatches tend to jump out, like trying to use a hadron collider to study the blood pressure of a frog.

I almost like to believe that a crackpot that could masquerade as a scientist might actually discover something because they would be forced to actually engage with the concepts and tools of science.

The number of String Theory papers, PhD candidates, and conferences was staggering.

I have found arxiv to be a dead-end when publishing without affiliation. It's impossible to get feedback on papers, let alone the necessary endorsement to submit. That leaves SSRN as the only option. Arxiv requires you to email authors for endorsements which almost never works, since it's essentially spam.

Why not submit to a journal?

1) opposition to journal copyrighting policies

2) cost

Most journals in the fields dominating the arXiv (physics, math, CS, and probably more) allow you to submit to the arXiv concurrently. So unless you refuse out of principle, you can both get free refereeing and release the paper open access.

I'm aware that most subdisciplines of physics and CS are sympathetic to arXiv, but that is most certainly not the case in bio and chemistry, although the rise of biorxiv may start changing things.

OK, but paulpauper was considering submitting the to arXiv.

The "without affiliation" aspect is a dealbreaker in many fields. Lots of journals won't even acknowledge you if you're not at an academic institution.

Actually, I think you'll find there are sufficient number of mid or lower tier journals that will send these out to referees. I've refereed papers from non affiliated researchers before. And even Nature will send you an rejection letter.

It's not a dealbreaker in computer science, where the norm is for blind peer review. And in any case, you can just list your undergrad institution or a company.

There is vixra (vixra.org), but they also (mostly) have the "funny" stuff.

The arXiv is not necessarily a tool where everybody can publish everything. It is primarily a tool for the community, which consists mostly of people with affiliations. If you want feedback for something as an outsider, go to conferences and present a poster.

A few more nice points about arxiv: 1) peer-review can take sometime, it's nice to be able to get your work out earlier publicly for the records, which you can make a claim later on 2) even after publication, it is accessible by anyone (and it is common to update the arxiv version to match the published version, including improvements after peer review)

It's probably not very useful for getting comments. The only comments I ever get are along the lines "I see your nice manuscript, please don't forget to cite these papers of mine". To my surprise, it is mostly "leading" physicists who send such mails (which makes me doubt their citation records).

>> The “insiders” can immediately tell who is an “outsider.” Often it doesn’t take more than a sentence or two, an odd expression, a term used in the wrong context, a phrase that nobody in the field would ever use.

seems a bit like when you're on top of your game with a certain programming language and feeling bored with stackoverflow questions and mailing lists asking the same old questions. Happy that in software we can easily move on to a new language / concept / idea to keep things exciting.

Of course, you can do that in science too. I believe the intent of that quoted text is not to indicate that all questions in scientific fields represented on arxiv, or even most, are answered. The only purpose is that language easily belies who does not "belong"

Arxiv banned the physicist who actually was able to build a physical model that makes sense in every aspect. Just because the reviewer did not understand one of the conclusions, the periodic table, that was a result of the theory. Of course you don't understand a result if you don't understand the theory which requires months/years to really grasp. The principle they are using only works if you stay in your paradigm, but if your paradigm is broken, you loose any real progress.

http://archivefreedom.org/freedom/Sarg.html

Having spent the time to understand, I only laugh these days about the many papers I see every week that contradict our standard model :)

Sorry to break this to you but this is crackpot science, and there is no way such a manuscript can pass through any respectable physicist.

Obviously you're not a physicist; let me simply say that this is not how a scientific paper looks like.

It is just some philosophical text attempting to interpret some basic concepts in physics, introducing new fields no one ever seen (seriously?), and myriad of strange conjectures coming from nowhere and it goes borderline mysticism about vacuum. There is no new basic equation from which you can derive new, experimentally testable conclusions.

It's not even wrong.

Here're a few gems from his "theory":

> The energy is inseparable attribute of the matter.

> The Bohr atomic model is a correct mathematical model when assuming that the space is void.

> The intrinsic matter could never disappear.

> The vacuum is not a void space, but contains a unique grid structure. This grid structure named a Cosmic Lattice (CL) is built by two types of alternatively arranged nodes, each one containing 4 sub-elementary particles with shape of six sided prisms.

> The Newton’s gravitation (universal gravitational law) is a propagation of the IG field in CL space

What?!? I'm sorry but I can't believe I wasted ~10 minutes reading this crap. Yes, sorry about the language, but this is bullshit pseudoscience and any sane respectable physicist will reject this and blacklist this guy.

Here's how new theories in physics work: you put very little and very very basic assumptions into your theory (which is essentially an equation, that is compatible with experiments and hence is compatible with the old theory within a certain limited domain) and new stuff you derive from your basic equation comes out it in the form of experimentally testable predictions. You have to get a lot more stuff than you put into building your theory.

Special relativity is a very nice example, you just assume Galilean invariance and constantancy of speed of light and out comes time-dilation, length contraction, energy-mass relation. If you add equivalence principle you get general relativity which gives you spacetime and new theory of gravity, blackholes, etc etc. And old Newtonian physics is happily reproduced when speeds and energy densities are small. And most importantly, predictions of new theory agrees with observations and experiments in the domain where the old theory failed.

Some intermediate mathematical stuff between your basic equation and experimentally testable measurable predictions might be difficult to interpret (wavefunction and entanglement with Schordinger equation, "holes" with Dirac equation, virtual particles, etc.) but you don't dwell on this stuff like a philosopher because neither your current theory nor experiments doesn't give you a deeper understanding to it anyway. So instead of going mystic, you sweep the non-testable stuff under the rug, hope that future physicist will have a better theory eventually and just "Shut up and calculate".

What you don't do is, you don't pull a myriad of new philosophical stuff out of nowhere where your assumptions are all that comes out and call it physics.

Even if you're not a physicist, you can see what kind of a "scientist" he is by checking his publication history (0 papers in any respectable physics journal).

The requirement of testable predictions would rule out an entire field of current theoretical physics. Not that I disagree with you (quite the contrary), but (almost) no paper on String theory would be accepted under that rule. Also, papers on pilot wave theory vs. Kopenhagen interpretation would come to mind.

The BSM model is testable in every aspect. That the basic building blocks are just fucking small compared to a neutrino for example, is a physical problem. The model does not prevent you from measuring it, quite the contrary compared to many other unified models currently discussed.

You could for example find another plateau in FQHE and the model would be gone.

Science isn't a computer program based on some simple algorithm. I understand that it can be difficult to spot crackpot to a layman.

1. String theory doesn't assume 20 crazy things and its axioms aren't its only predictions 2. String theory does have a basic equation from which you draw many results, leading to falsifiable predictions that is not covered by standard model. Some of them may not practical with today's experimental tools, but that's not all of them. I take you haven't seen any string theory papers. I'm saying we're suddenly going to see superparticles this year at LHC, but it is entirely plausible that if they exist, LHC should eventually detect them. 3. You're probably talking about this http://advances.sciencemag.org/content/2/2/e1501466 recently posted on HN. In case you missed, that is an experimental paper.

You pulling most out of context. The theory is here:

http://www.amazon.com/Basic-Structures-Matter-Supergravitati...

Most of your points are taken out of context. If you don't understand the model, they don't make much sense.

First this are the basic assumptions:

- 3 dimensional eucledian space without physical properties

- 2 types of very small fundamental particels. Basically balls that are indestructible and vibrate

- 1 law of attraction valid in empty space. F_sg = G_0 * (m1*m2)/r^3. Which makes logical sense in our 3 dimensions. - classical logic

>> The energy is inseparable attribute of the matter.

You don't understand the difference between Newtonian mass and matter in this model. Newtownian mass is a derived effect of all high level structures that are impassible by CL space - eg. a certain size.

>> The Bohr atomic model is a correct mathematical model when assuming that the space is void.

After certain constants from the standard model are derived, eg. chapter 2,3,5,6 you can put up a mathematical model that is like the Bohr one, but this is not what the BSM model actually shows. Once you understand the geometric structure of protons/neutrons/electrons you will understand that even quantum mechanics is very close to the Bohr model compared to the BSM one.

>> The intrinsic matter could never disappear.

Yes. It would violate the energy conservation and it has been shown that this is the golden rule. As said mass != matter !!!!

>> The vacuum is not a void space, but contains a unique grid structure. This grid structure named a Cosmic Lattice (CL) is built by two types of alternatively arranged nodes, each one containing 4 sub-elementary particles with shape of six sided prisms.

It becomes obvious once a super massive black hole is understood, how a crystalization process will first create those complex formed prisms with it's substructures and later high level structures like protons/neutrons/electrons. Unfortunately apart from one type of GRBs as the last signature of the process, it is mostly hidden from observation due physical reasons.

> The Newton’s gravitation (universal gravitational law) is a propagation of the IG field in CL space

So, you neigher will be able to understand CL space nor the IG field by now and the first thing you cry is crackpottery.

> Even if you're not a physicist, you can see what kind of a "scientist" he is by checking his publication history (0 papers in any respectable physics journal).

https://www.researchgate.net/profile/Stoyan_Sargoytchev/publ...

you are very bad at searching I have to say.

Journal of Atmospheric and Solar-Terrestrial Physics

Annales Geophysicae

Indian Journal of Physics

Canadian Journal of Physics

Applied Optics

Physics Essays

Advances in Space Research

Proceedings of SPIE - The International Society for Optical Engineering

...

To deduce a complete new model, you need to have a look in all areas of physics/chemistry and not just at one, especially theoretical ones. And you need to read a lot of papers and especially the ones that contradict the status quo.

And yes, the model explains much more then the standard model through fewer assumptions. Most of your "constants" are just derived.

I said a respectable physics journal, not no-name journals with impact factor <1. APS journals, Nature group journals or NJP are examples of respectable journals. Indian Journal of Physics, Physics Essays or Canadian Journal of Physics are not. Furthermore, geophysics and atmospheric physics aren't even remotely related to high energy physics, so I don't understand why you're talking about those papers anyway: neither high energy physics nor cosmology are his fields.

Let's assume you don't know any physics. You can go with this: if a new model is any good, if it is indeed a groundbreaking new theory, it appears on Physical Review Letters. It does't appear just on arxiv (only to be removed 3 times and gets the author banned) and definitely not on the crackpot hive vixra.

He wasn't banned from axiv because he is a hidden super-genius misunderstood by his stupid narrow-minded peers and prevail after his death. He was banned because, for god's sake, he is a crackpot scientist.

Apparently, I need to spell this out for you: his publication is a disaster, no respectable physics department will ever even consider hiring him.

I'm not even gonna waste time on your "theoretical" ramblings about his crackpot manuscript, which is 100% bullshit.

I checked your post history: either you're that crackpot or you're a disciple of his cult (which might be expected given that he's written several books based on his "theory). I'd bet on the former, that you're Stoyan Sarg Sargoytchev; your very first post here, as well as many of your other posts are about his "theory", I can't image someone getting hooked up it, promote and defend a no-name pseudoscientist for so long.

I've read the paper that's available on Arxiv and agree it's quackpottery [1]. The paper does not introduce concepts in a scientific way. It seems to me like a series of ramblings or ideas about how things might work, with no results or justifications, indistinguishable from hypothetical future technology in science fiction. There's no actual scientific content, no presentation of a new theory and working out results that derive from it.

Be advised, poelzi seems like an apologist for Stoyan Sarg. See previous comment history where they mention the theory and link to the book numerous times. poelzi seems to mention and link to M. Sarg's book at every opportunity. E.g. [2], while also saying that "The papers helped me to get into the theory, but real deep understanding is not possible from them." This alone should be enough to tell you that it's garbage science. The papers reference material published in M. Sarg's books and do not stand alone.

poelzi, if you believe this is science then you've been led astray. It's not. It's pseudo-scientific make-believe. There are no scientific theories, explanations, or results in M. Sarg's papers. Credible scientists write papers describing their results to the world, that are self-contained and justify the theory and results that they allege. M. Sarg's papers and/or books may contain assertions that his theory actually explains things and makes predictions, but there's no substance in the papers whatsoever that I can tell. For [3] to be a credible paper it would need to extensively document the mathematical modeling that yielded these structures (or cite another paper that does).

Perhaps the degree to which these papers are lacking will be more apparent if you review some respected scientific papers introducing new theories, such as "A Dynamical Theory of the Electromagnetic Field" [4]. If you are a layperson without scientific training, then the introductory paragraphs might not be obviously distinguishable from M. Sarg's writings. However, the paper is concrete and lays out explanations alongside mathematical results. For example, observe page 10 in the PDF where the real mathematics begins, as Maxwell lays out his results with a series of equations interspersed with commentary about what they mean and how they're derived. This continues for the rest of the paper, and is the justification by which Maxwell's results are asserted and explained. Anyone thinking that they can credibly evaluate a unified theory of everything should feel comfortable working with Maxwell's equations on a white board, and using them to predict the behavior of physical systems; his equations are now introductory level for students of physics, and indeed Maxwell's equations are now taught to high school students taking AP Physics C (Electricity and Magnetism).

M. Sarg's papers are absent any kind of explanation or justifications; they are absent mathematics and seem to be bare assertions or "mentions" of ideas in his books. This is not how credible science is published to the world. Anyone can whip up diagrams and data plots alleging that their theory explains something, but without the substance behind it for others to review in a detailed mathematical way, there's no science there.

[1] http://lanl.arxiv.org/abs/physics/0205052

[2] https://news.ycombinator.com/item?id=11166770 https://news.ycombinator.com/item?id=10781247 https://news.ycombinator.com/item?id=10770899 https://news.ycombinator.com/item?id=10626579 etc.

[3] I came across another paper "Graphical 3D Modeling and Analysis of Molecules and Nanostructures with the BSM-SG Atomic Models", which has a lot of pictures of hypothetical atomic structures, and no mathematics describing the derivation of these structures. https://www.researchgate.net/publication/270572848_Graphical...

[4] http://rstl.royalsocietypublishing.org/content/155/459

Reply on news.ycombinator.com