Hacker News Reader

Comment by isolli | Hacker News Reader

It reminds of the alternative to dark matter that is MOND, stating that the law of gravity behaves as 1/r at large distances (as opposed to 1/r2 at our scale) [0]. I personally like it because it seems less ad hoc than to posit the existence of invisible matter, and similar in my mind to the special relativity adjustment. (Do not take this endorsement too seriously, of course.)

However: "The most serious problem facing Milgrom's law is that it cannot eliminate the need for dark matter in all astrophysical systems: galaxy clusters show a residual mass discrepancy even when analyzed using MOND. The fact that some form of unseen mass must exist in these systems detracts from the adequacy of MOND as a solution to the missing mass problem, although the amount of extra mass required is a fifth that of a Newtonian analysis, and there is no requirement that the missing mass be non-baryonic." [0]

[0] https://en.wikipedia.org/wiki/Modified_Newtonian_dynamics

View on HN

The problem with dark matter is not that it is ad hoc. It is that it feels un-parsimonious to resolve a discrepancy in rotation curves by inventing five times more stuff than you had, all undetectable, and distributed just right to make all your curves come out right.

It turns out that the mass of neutrinos produced so far adds up to as much mass as all the rest of particulate matter. I guess neutrinos must be distributed about evenly throughout the universe. If we could not ever detect neutrinos, that would be awkward.

I have not heard of a mechanism by which these dark matter particles can cool and condense to clumps to seed galaxies. By contrast, baryonic matter gets to emit photons to give up kinetic energy.

Particle physics has a pretty good track record of inventing undetected particles to solve missing item problems, then finding that particle later. I'd guess we're up to hundreds of such particles like this so far, with really big finds being positrons (and all anti-partickes, needed when adding relativity to spin), neutrinos (needed to account for missing mass), all the quarks (needed for certain symmetries, among others), and the Higgs bison itself (needed to explain masses).

So don't discount adding currently undetected things as explanations for surprising measurements. It's been extremely fruitful.

I want a Higgs bison steak, please! :-) Medium rare!

Got no objections to inventing particles to take up and carry off whatever bit of stuff has to conserved: neutrinos, virtual particles, gluons, what-have-you.

I don't really even object to inventing five times the mass of the known universe.

It just feels, as I said, un-parsimonious. It feels quite a lot like inventing God to patch a "missing link" in your evolutionary succession. Maybe you'll find God there somewhere, but it doesn't seem like the first thing to try.

So, maybe dark matter really will turn out to be an ordinary axion or something detectable only if you manage to squint just right, and God just loved those so much that almost everything is them, and we are all made out of just leftover scraps. Maybe there are dozens of elementary axions, with relationships and exciting interactions we can never figure out because we can't touch them in any way but gravitationally.

But I still want to know how they cool.

I don’t think you’re giving enough credit to physicists.

Dark matter isn’t just adding a fudge/“God” parameter to the equations to make the numbers work. The “stuff” that’s thought to be missing is described with very precise properties… the fact that it’s “dark” means something, the fact that it’s “matter” means something.

Well, in this case they're not just inventing nonsense to fix one little thing. There are a lot of various pieces of evidence, all pointing to this fix. If you want to patch up a lot of little pieces of evidence, each with it's own magic, that would be even more "un-parsimonious".

The evidence for dark matter is pretty solid.

As this [1] article states, while giving 5 independent reasons scientists think dark matter exists, "No other idea explains even two of these".

Wikipedia lists eleven different places it shows up.

The research literature has more. And AFAIK there is no other unified (or even close to unified) explanation for all these observational pieces of evidence.

[1] https://medium.com/starts-with-a-bang/five-reasons-we-think-...

[2] https://en.wikipedia.org/wiki/Dark_matter

They would be more persuasive if they were seen to be working on the observations it is manifestly incompatible with.

On the flip side, why would you expect us to be equipped to observe more than 20% of matter give the narrow range of our vision and hearing? Anything that doesn't interact strongly at our scale wouldn't be useful evolutionarily.

Sadly, the Higgs bison has been extinct since the last ice age. Too bad, it would have been wonderful to see such a beast.

(Lovely typo.)

> It turns out that the mass of neutrinos produced so far adds up to as much mass as all the rest of particulate matter.

I don't believe that's true? The energy density in neutrinos today is < 0.5% (baryonic matter is ten times that.)

My pet theory for gravity / general relativity (which is probably completely wrong, but fun to think about), is that space itself is created through particle interaction. Space will thus be denser and create a gravitational effect in places with lots of matter.

That means space also has to decay. And unless you want space to become non-existent far away from matter, the rate of decay must also fall off somehow as you get further away from matter.

I wonder if you could get such a space creation/decay process to match a curve that's 1/r2 at short distances and 1/r at long distances...

But aren't particles emergent property of space wave function collapse?

wave function does not have to collapse for particles to exist. Particles and wavefunctions are just different lenses for which to make sense of excitations in quantum fields

My pet theory is that let's say particles are the excitation of the field, imagine you come to a lake in a quest to understand water. The surface is field. You throw in a rock and the splashback is particle, you analyze the particle, you throw other things and analyze the other particles. Then you form a water theory based on these particles and you find out that boats shouldn't float at all.

It's because particles are exceptional things, not the normal, calm field they come from. If you base your theory on those exceptional thing it will be non representative or outright wrong.

That's why quantum physics is giving us weird results, because particles are exceptions, not the common state of the field.

I'm intrigued by this thought exercise!

If it doesn't depart too much from the analogy, what is the lake (water below the surface, under the field) and what is above the field?

Are we supposed to assume that the water below the surface is inside the field, and above the surface is outside of the field?

What is the non-exceptional form of the particle, or thing that makes up the particle? I guess this is the essence of your analogy, and your point is that we've spent so much time trying to identify this thing as a particle, we can't yet answer this question?

Or have I now dissolved the analogy by taking it too literally?

If it doesn't depart too much from the analogy, what is the lake (water below the surface, under the field) and what is above the field?

More than what it is, the question is what properties does it have. Anyway the object of interest is the surface.

What is the non-exceptional form of the particle, or thing that makes up the particle?

The surface.

The field is the "deeper truth", the particles are the second hand tidbits of information, it's like trying to understand marriage only from the jokes instead of marriage itself.

Yeah, totally get that.

Was trying to understand what was inside (water) and outside (air) the field to help me start thinking about what the field (surface) might be.

Interesting idea. Although, if particle interaction creates space, shouldnt it counteract gravity? If the Earth, full of interacting particles, is hurling space outwards, stuff around Earth should fling into infinity, not fall towards it, or what am I missing?

There's also the issue that some galaxies don't violate our pre-dark-matter models as much or do so in a different way. With dark matter that's easily explained as an abnormal dark matter halo (eg maybe displaced due to an ongoing interaction with another galaxy or absent for an unknown reason). When abnormalities like that are considered, dark matter becomes the elegant solution while theories that tweak gravity start to seem somewhat ad hoc.

Isn’t the bullet cluster the bullet in the head for MOND being a simple solution?

That's the fascinating thing to me: it doesn't matter! MOND is a successful pseudoskeptical meme despite not being a successful physical theory.

The narrative of MOND-the-meme is totally wrong but extremely compelling. That's enough to give it outsized fitness in its niche, and then that general compellingness (which comes from being really good at tricking us) is misinterpreted as scientific merit. Even by those who say they're interested in truth, not stories.

It's an excellent test-case for critical thinking because everything about it as a meme should set off alarm bells, but often it doesn't.

In the same vein, I miss cold fusion. It was such a lovely idea, and had such great promise. A shame that reality intruded...

Wait how is it pseudoskeptical? It's made independent predictions which have subsequently been confirmed, like a slope of 4 in the tully-fisher relationship and the EFE (among others). That's the definition of science. The EFE is particularly compelling because it was originally posited as a criticism of MOND but then the data came in.

I'm trying hard to be precise, but I could always do better.

MOND the physical hypothesis is perfectly good science, whether it turns out to match reality or not. My issue is with MOND the meme, the complex of beliefs among non-physicists (like me) that MOND explains "dark matter"[0] without matter which is dark, whose fitness isn't really based on the success of MOND-the-model.

I say MOND-the-meme is pseudoskeptical because it a) manufactures disbelief by misrepresenting a subject and then b) provides a "skeptical" alternative in order to appeal to skeptics which c) does not itself stand up to skepticism because (a) was a strawman and (b) was motivated reasoning. It survives because it makes you feel smart, not because it's sound or valid or the MOND hypothesis is true.

Accepting MOND-the-meme would still be problematic even if the MOND hypothesis was overwhelmingly confirmed tomorrow.

[0] Taking full advantage of the ambiguity between "dark matter" as the name for observations, one hypothesis explaining them, and an overall cosmological model.

Wow, no offense, but this is an insane level of mental hoops to jump through to insult people who are putting forward the MOND hypothesis. It’s almost invisible, but I think it’s there.

I guess I'm still not being clear. I'm not insulting MOND or people who like it, at least not on purpose. That's just absolutely not my goal. At all.

Rereading my first comment, I'm not happy with the tone. I wrote it quickly on a phone and didn't get it right. I was honestly just making observations without judgement, but it doesn't sound that way.

The short version is that laypeople with a skeptical bent like to say "eh I don't believe in dark matter, those silly physicists are inventing particles that aren't there, MOND solves the problem without extra particles so MOND is probably right."

It's an easy way to be "skeptical" without really knowing anything about the relative success of the two theories. Because MOND is an underdog theory, you can feel like a skeptic questioning dogma by supporting it... even if you don't really know much about the merits.

You also said:

> despite not being a successful physical theory.

I think the jury is still out on that, and, like I said, it has actually made predictions which is more than what you can say about LCDM, which as far as I can tell has only made post-hoc rationalizations.

MOND is unsuccessful in the broadest sense because the jury is still out. It hasn't failed, and does have support, but hasn't managed to convince the experts that its individual successes translate into overall success as a model.

Narrowing the meaning of "success" can lead (me, at least) into a philosophical morass about what really are predictions, what is parsimonious, how do we science, why even is anything. Sorry, maybe it's not fair, but I don't really want to go there.

You shouldn't have said successful, because that implies a finality to the judgement. There are definitely "dead" scientific theories (like phlogiston) and MOND is far from that, it's making highly competitive predictions, it just seems for some reason that the contemporary culture of science is overly resistant to it. I have no idea why.

I mean, the point of the scientific process is to come up with hypotheses. MOND might not be as "strong" of a hypothesis as Dark Matter is, but MOND does make predictions that can potentially be tested or mathematically explored.

It seems a little flippant to denounce it as a ridiculous "meme" idea. It's healthy to have different scientists pursue different theories. And valuable insight can be gleaned even when taking the time to _disprove_ a distant possibility.

I'm saying there's a scientific model called MOND and also, separately, a literal Dawkinsian meme[0] which features the acceptance of that model based in part on motivated reasoning, not science. I'm not dismissing the hypothesis. It might be right, it might be wrong. It's worth exploring either way. I totally agree that even proving something wrong improves our understanding.

I'm criticizing MOND-the-meme, I guess, but also admiring. I don't like it, but it's very good in its niche.

Like a shark. Stay over there please... but wow.

[0] https://en.wikipedia.org/wiki/Meme

No?

It's bullet in the head for dark matter as well http://backreaction.blogspot.com/2017/01/the-bullet-cluster-...

The answer is most likely MOND + dark matter.

Gotta love the downvotes by the people that don't read the article.

Tl;dr Dark matter model simulation can't explain bullet cluster, and MOND can.

> Either way, the Bullet Cluster remained a stunningly unlikely event to happen in the theory of particle dark matter. It was, in contrast, easy to accommodate in theories of modified gravity, in which collisions with high relative velocity occur much more frequently.

> The answer is most likely MOND + dark matter.

Here[0] is a toy model experiment to demonstrate that neither Dark Matter nor MOND are necessary to explain spiral galaxy rotation curves, which coincidentally is the original purpose Dark Matter was postulated to explain. Dark Matter has since been fingered to explain other phenomena since, but it is not commonly known or accepted that galaxy rotation curves no longer needs Dark Matter (or MOND, for that matter).

[0] https://m.youtube.com/watch?v=PL0ewiwqoTw&t=9m40s

Most physicists who can be bothered to comment do seem to think MOND is extremely ad hoc, in a sense.

My understanding is the 1/r^2 behavior comes from solving for Green's function of a PDE. Gravity is similar to Gauss's law (i.e. 1/r^2 behavior of charged particles) in the underlying PDE. So something in that equation would have to change.

Apparently it makes very accurate predictions though, so it's not just a matter of ad hoc tweaking some law to fit reality... It does seem to capture some regularity/mechanics of nature.

huh? how is MOND less ad hoc?

it introduces more parameters to fit its arbitrary conception about how gravity is supposed to scale. even worse, since basic MOND disagrees with a ton of basic evidence since it just doesn't work except for the few problems it was created for, the actual MOND derivatives that do try to make it work introduce even more parameters to fit.

It introduces ONE parameter. And TWO, for the relativistic version. To simplify, LCDM introduces at a minimum one parameter (relative dark matter:visible matter ratio) for each galaxy to get the rotation curves correct. That's at least a billion parameters.

MOND has made successful predictions:. The EFE, for example (there are others). LCDM has not, as far as I can tell -- please correct me if I'm wrong [0]

> work except for the few problems it was created for

That's actually a poor heuristic. Suppose we hadn't discovered relativity yet (but had galactic rotation curves, which only depend on good telescopes and newton, and Doppler effect), and we observed mercury's precession. Would you then use precession as an argument for dark matter on the basis that there's a dark matter entity in our solar system pulling mercury around? I would think that would not be a good idea.

[0] ok I looked it up and actually LCDM predicted no galaxies with redshift greater than 7... And thanks to jwst we now know there are galaxies with much much greater redshift, a feature explicitly predicted by MOND.

[1] I forgot about cmb polarization, that is a successful prediction of LCDM

XKCD 1758 on the subject https://xkcd.com/1758/

This is like saying that the existence of precession in mercury's orbit detracts from the adequacy of a Newtonian analysis and so there must be a dark matter blob pulling mercury's orbit around.

I always wonder if this "missing mass" is just light. As in it is so close to zero mass to be undetectable at a normal scale, but at galaxy scale, it actually has mass.

How do you square your idea with the other substantial implications of photons having mass?

Photons bend spacetime even without having mass.

Bingo, does energy not affect spacetime the same as mass? I always assumed it did. My assumption was this was already worked into the calculations and not enough to account for dark matter.

If photons had a very small but non-0 mass, it would simply mean that they actually move slightly slower than c, but it wouldn't otherwise have any significant implications. There are in fact experiments that are trying to check whether photons have a mass we can detect.

Also, just as a note, I'm using the term "mass" for what used to be called "rest mass" (as opposed to "relativistic mass"). Even if their "rest mass" is truly exactly 0, and their speed is exactly c, they still have energy and momentum, and thus they have mass in general relativity, but that is a different discussion.

Wouldn't that mean that light also has infinite energy?

This is the basis of the "ultraviolet catastrophe" and lead to the discovery of quantization of radiation. The "catastrophe" being that as classical laws for light implied a potentially unbounded amount of energy as wavelength decreased.

https://en.m.wikipedia.org/wiki/Ultraviolet_catastrophe

https://youtu.be/rCfPQLVzus4

I've wondered similarly, but I don't know why it would need mass. That just complicated things. It has energy, and we have an equation that tells us how much equivalent mass that would be. It's not missing mass, it's energy that's the cause of the extra gravity.

I also assume this has been accounted for though

General Relativity has https://en.wikipedia.org/wiki/Stress%E2%80%93energy_tensor which is the term that generates the gravitational field. It includes the energy of photons.

Is it possible that the gravitational discrepancies can be explained with dyson spheres? Or dyson spheres collected into grids (with a dyson sphere scale tech you can prob move suns by making the dyson sphere a giant fusion engine).

There is an advantage to arranging dyson spheres closer together in order to reduce latency between nodes.

If your dyson spheres were optimized enough, could absorb all known spectrums and recycle them. Wouldn’t that appear completely dark save for gravity?

I mean, isn’t it possible entire galaxies have been converted in this way? Seems like it could be a much simpler explanation within the confines of current understanding.

But I’m sure this has been thought about and accounted for.

My layman understanding is that the dyson spheres would occlude objects behind them if they happen to be in line of sight to us. Given that things in the universe are not static, other (visible) bodies would move in and out of occlusion, allowing us to detect the DS.

Nothing an LCD and some cameras can't solve

Only if they would want to fool us specifically. Those LCDs would have to display a different image for every different point of view, so they can choose only one planet in the universe to fool.

And they chose us. Just our luck!

Sure, but occams razor would probably prefer some gas cloud.

> If your dyson spheres were optimized enough, could absorb all known spectrums and recycle them.

A really cool outcome of statistical mechanics and thermodynamics is that what you suggest is impossible. The dyson sphere can be as amazing as you want, using every single trickery permitted by physics, but at some point it will enter thermal equilibrium with its host start (or pipe that heat somewhere else where you would still be able to see its thermal radiation). Inherently, if they are optimized to be good absorbers, they will have to also be thermal emitters. There is no such thing as "recycle all spectrum".

Tangentially related to the fact that a "perfectly black body", i.e. a perfect absorber, is also the perfect emitter of thermal radiation of its own.

Couldn't it be a good emitter, but redirect the heat in a focused way? (beam of RF)

Nope. That would violate the 2nd Law of Thermodynamics.

Why?

Couldn’t one build a dyson sphere around a blackhole? Wouldn’t the blackhole have properties of a Bose–Einstein condensate given the matter at the center cannot vibrate?

It would be the coldest temperature possible. Heat moves downhill, you deposit heat into a singularity and have a shell around it.

That aside, at even a scale of a single dyson sphere, something tells me when a civilization has the power to bend spacetime, the rules of possibilities begin to change. I think it is unwise to operate with so much certainty in the face of possibilities.

The sibling already mentioned the axioms of thermodynamics that forbid it. Another way to see that this is forbidden is Etendue https://en.wikipedia.org/wiki/Etendue#Conservation_of_etendu...

Any thoughts on this? (discussion on temperature of a black hole)

https://astronomy.stackexchange.com/a/26429

It's very similar to artificial MACHOs. The natural version uses neutron starts and other difficult to detect big natural things instead of Dyson spheres. https://en.wikipedia.org/wiki/Massive_compact_halo_object

For this to match observations, all galaxies visible outside the milky way would have to be Dyson sphered through their whole extent. If they weren't, it would be obviously visible, as an expanding civilization that had only spread throughout half a galaxy would be a totally lopsided galaxy. And for some reason, all of those civilizations must have decided to leave a certain percentage of stars un-sphered even though they have the capacity to sphere them all.

And the Milky way would have to be un-dyson sphered, because if it were, we would be able to notice these spheres close up. e.g. if dark matter is 5x regular matter, we would expect there to be 5 Dyson spheres of stars the size of the alpha centauri system all within 4ly of Earth. This would be easy to detect through occlusion. Also, gravitational interactions between sphered stars and unsphered stars would be easily observable in our own galaxy. So for this to work, at minimum, the whole observable universe must by Dyson sphered, except for our own galaxy.

As gravity is a signed integer it overflows when it gets big enough and turns into a negative number.

What is NaN+1?

NaN1

Transfinite[1] number.

[1] https://en.m.wikipedia.org/wiki/Transfinite_number

how would NaN apply to signed integers?

Then you have a signed NaN and instead of eating them at your local indian food restaurant, you put it in a collection.

Watman!

(Reference https://www.destroyallsoftware.com/talks/wat)

What ever happened to Gary Bernhardt? I miss his funny insight and those fantastic screencasts of his, they've shaped my way of thinking about code architecture

Highly recommend his new thing: https://www.executeprogram.com/

Very happy to see him doing big things.

Now he can go back to stealth meta sarcasm talks.

You're thinking of temperature.

Reply on news.ycombinator.com