Matt McIrvin

he/him

I'm a programmer who was trained in physics long ago. If on Bluesky, follow @ap.brid.gy so I can see you.

Blog on Dreamwidth:: https://mmcirvin.dreamwidth.org/
Bluesky mirror of these posts:: https://bsky.app/profile/mattmcirvin.mathstodon.xyz.ap.brid.gy

Public Key

npub12gnjepalfuzt5zjywxj0vzjf6e3kpjt8hnsqqs8l3fzv2sxvwxps6qd572

NIP-05 Address

mattmcirvin_at_mathstodon.xyz@momostr.pink

Profile Code

nprofile1qqs9yfevs7l57p96pfz8rf8kpfyavcmqe9nmecqqgrlc53x9grx8rqcpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dsdm6men

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub12gnjepalfuzt5zjywxj0vzjf6e3kpjt8hnsqqs8l3fzv2sxvwxps6qd572

Show more details

Last Notes

2026-04-03 16:25:50 UTC

- reply

...Right.
Conventionally we think of a pion as a quark-antiquark bound state, and a proton or neutron as a bound state of three quarks. But naively, from that picture, you'd expect a pion to have about 2/3 of the mass of a proton, and they're much lighter than that, more like 1/7 the mass of a proton.
The sigma-model picture is a rough way of picturing how that can be.

2026-04-01 14:20:19 UTC

- reply

I always thought of this as basically similar to the way the Higgs field gives masses to fundamental fermions, except that the condensate is a composite field (and it's not breaking a gauge symmetry).
Semi-popular accounts sometimes describe the masses of hadrons as coming mostly from strong-force binding energy, and I always thought that was confusing because binding energy is usually negative!
But when we think of it that way, we're taking the free masses of their constituents at long distance as the basis for comparison, and for quarks, there is no such thing: they only have a "free mass" in high-energy, short- distance interactions, which is the opposite.

2026-02-17 04:23:07 UTC

- reply

I was there to play some pinball at Joe's Playland but the beach had a bleak grandeur.

2026-02-17 04:14:55 UTC

Salisbury Beach in February
https://media.mathstodon.xyz/media_attachments/files/116/083/999/506/825/149/original/77053313f3f0a1f6.jpg
https://media.mathstodon.xyz/media_attachments/files/116/083/999/675/541/817/original/039ba1e3b4fce122.jpg
https://media.mathstodon.xyz/media_attachments/files/116/083/999/709/065/781/original/d4dbd422fc076f6e.jpg

2025-10-27 20:31:09 UTC

- reply

right!

2025-10-27 20:17:59 UTC

- reply

I remember a long discussion where we were going around and around on "what happens if you try to jump into the white hole?" and we kept saying "it turns into a black hole and you fall in". Which, yes, is pretty counterintuitive.

2025-09-25 09:42:42 UTC

The Rapture? Oh, it is as it was predicted when I was but a young man. The man from Mars stopped eating cars and eating bars and now he only eats guitars

2025-09-22 17:57:21 UTC

- reply

I think he lifted the word "bloog" from Dr. Seuss (with acknowledgement).
His protagonists are always these guys with really poor impulse control which makes it hard for me to empathize with them, but it does help get them into trouble.

2025-09-22 16:05:11 UTC

- reply

Rudy Rucker's weird fantasy novel "White Light" has some ideas along similar lines, though not this exactly.

2025-09-18 01:13:05 UTC

- reply

I get the impression the criticism was overblown; in this new paper they emphasize the robustness of the results across different methods--but this kind of image reconstruction is very, very difficult.

2025-09-17 10:26:06 UTC

- reply

The image released in 2019 was based on data collected in 2017.

2025-09-14 12:13:41 UTC

- reply

One of the tricks the various platforms do is to push two-factor authentication for security purposes, which is a fine idea, but then they make it so you can't use two-factor without having their own phone app installed and general notifications enabled.
Or there's YouTube's cute one: making it impossible to separately control in-browser notifications when you're on the web page, and app notifications on your phone. You opt out of one, you lose the other.

2025-09-14 02:47:06 UTC

- reply

When I was on Twitter I had to disable app notifications because even though I had only a small number of followers, occasionally something I wrote would actually get high engagement (usually just because I was riding some celebrity's coattails), and the experience of that with the notifications popping all over drove me up the wall. I don't think I want high engagement.

2025-09-14 00:04:22 UTC

- reply

It's probably going to depend on the state. I had no trouble getting one in a Massachusetts CVS, thanks to our governor essentially granting us a prescription.

2025-09-11 20:49:37 UTC

- reply

I'm half expecting "There was life on Mars, and it was from Earth" (microbes somehow transported on a meteorite).
It would be particularly epochal to find some life in our solar system that had a completely separate origin, but life is such a prolific thing once it appears, some small-scale panspermia with Earth as the source wouldn't be entirely surprising.

2025-08-05 15:25:21 UTC

- reply

yeah, I think I was mixing up the number i with the counterpart of the quaternion i, which in this scheme is different.

2025-08-05 12:52:44 UTC

- reply

I suppose that follows from expressing the state vector as, say, an SU(2) element left multiplying a fixed basis spinor like (1 0)^T.

2025-08-05 12:26:56 UTC

- reply

and they'd never touch the 2x2 matrix representation because they don't work with spin 1/2 particles, but in physics we did all the time.

2025-08-05 10:16:16 UTC

- reply

Well, I never worked with them in a graphics context myself-- when I was actually doing 3d graphics stuff I used rotation matrices. But, yes, I think in practice they work in a quaternion basis and don"t extract a matrix until they have to.

2025-08-04 20:17:40 UTC

- reply

Since my background was in particle physics I didn't call them unit quaternions, I called them elements of SU(2). (Converting to rotation matrices is then expressing them in the spin-1 or vector representation).

2025-08-04 14:35:10 UTC

- reply

Quaternions may be more popular than ever as a tool for representing rotations in 3D computer graphics, such as video games.

2025-07-29 23:01:45 UTC

- reply

it seems to me like defining them such that they're different would be an inconvenient way of representing infinitesimal differences since we can only write such a thing in the vicinity of a number with a terminating decimal expansion (or terminating in whatever base we're using). But we ought to be able to add our infinitesimals anywhere.

2025-07-29 20:39:49 UTC

- reply

I'm comfortable with saying "in the real number system (or the complex number system), 0.9999... = 1". Maybe someone can formulate some extension with infinitesimals in which it makes sense to say it's not (though my impression is that such notation is not very useful in such systems). But the real numbers don't have infinitesimals.

2025-07-15 12:20:51 UTC

- reply

Wangsness was the text we used in Marty Eckhouse's undergrad E&M class at William and Mary. One particular all-night group study session over a particularly dire homework assignment for that course (I think it was all about some kind of boundary-value problems) was the moment I started drinking coffee.

2025-07-15 11:51:31 UTC

- reply

Either one! My attitude at the time was more "the endpapers of Wangsness won't steer me wrong..." I recall Jackson's listing was not quite as extensive.

2025-07-15 10:42:08 UTC

- reply

Those vector identities are the kind of thing that were always written on the endpapers of electromagnetism textbooks--I rarely questioned their provenance...

2025-07-14 15:57:46 UTC

- reply

The Man Card is an amazing one-- people keep inventing new rules for manhood I never heard of. The last one that baffled me was "a man doesn't eat soup in public." Stated with great confidence by some Fox News guy. What? Soup? There's a soup rule now?

2025-07-11 02:50:21 UTC

Not familiar with Kali Kazoo, all I know is they did a cover of "Burning Down The House" with the Homestar Runner puppets
https://www.youtube.com/watch?v=pwCMLMktDkA

2025-07-08 18:28:34 UTC

- reply

though I've passed one hundred thousand miles I'm feeling very still, and I think my horsey knows which way to go

2025-06-20 13:08:55 UTC

- reply

also, the US agriculture giant Archer Daniels Midland is of course into biofuels, but that seems to be yet a third ADM.

2025-06-20 13:06:55 UTC

- reply

I see there is now some corporation called ADM Energy-- that makes this harder to search for!

2025-06-20 12:55:17 UTC

- reply

What you can do is define a *global* energy for a system with certain kinds of boundary conditions, like the "ADM energy". But it's a more constrained exercise.

2025-06-20 12:52:57 UTC

- reply

General Relativity is different from other field theories in that, usually, you can define a localized stress-energy tensor for the field itself (up to ambiguities that don't affect conservation laws). But in GR, if you attempt it, the quantity is coordinate-dependent and cannot be made covariant. So you can make the gravitational stress-energy in any local area disappear with a choice of coordinates. It's slippery.

2025-06-20 12:48:22 UTC

- reply

For systems corresponding to our everyday intuitions (like a localized gravitating system with a flat background at long distances), it's possible to define a "gravitational energy" such that everything is conserved. But not for the general case.

2025-06-20 10:33:41 UTC

- reply

This is something that Albert Einstein got stuck on for about a year (if I recall correctly) before just giving up. Emmy Noether came up with the original formulation of Noether's Theorem to clarify the situation.

2025-06-16 19:20:14 UTC

- reply

the nation that controls magnetism controls the universe!

2025-06-15 10:27:33 UTC

- reply

This is definitely a case that's bugged me for a long time: I remember reading textbooks that started out giving the zero case explicitly every time when talking about power series because they didn't want to define 0⁰=1, and then just sort of dropped it hoping you wouldn't notice.

2025-06-13 12:32:36 UTC

- reply

this seems obvious in hindsight... I suppose it wasn't explored in detail before because orbital angular momentum isn't considered that important in a system without rotational symmetry.

2025-05-21 19:40:48 UTC

- reply

The simplified version of it that is usually used in quantum physics is very straightforward to prove: a symmetry of the Hamiltonian has some generator that commutes with it, and since time evolution is basically the exponential of -iHt, it can't affect that generator and it represents a conserved quantity. The end! But thinking about energy in a general-relativstic system is much harder.

2025-05-20 20:14:07 UTC

- reply

this kind of thing is what Emmy Noether was really thinking about in the first place, wasn't it? The theorem is ubiquitous in contexts that have nothing to do with general relativity but she started out more or less doing it on the maximum difficulty level.

2025-05-17 11:47:31 UTC

- reply

Huh.
This is so *almost* but not quite related to some of the stuff I was poking around recently--constructible regular polygons, particularly. (The odd-sided ones have numbers of sides whose prime factorizations are square-free, with unique primes of the form 2^(2^n) + 1. But those really have to be primes.)

2025-05-14 17:14:14 UTC

- reply

They're reserving the right to turn anyone into an "illegal" with the wave of a wand (you think we revoked your status illegally? Too bad, "illegals" don't get due process), so they could make those numbers higher.

2025-05-13 20:58:51 UTC

- reply

there are some bits of Project West Ford (an early experiment to blanket the Earth with a reflective orbital shell of tiny needles) still up there being tracked-- that wasn't supposed to happen, but I think some of the micro-needles clumped together and were less susceptible to drag. I think about that whenever I am in Westford, Mass., the town they were named after (because a project antenna was there, I think).

2025-05-13 16:23:51 UTC

- reply

In reality, it's probably prohibitively hard to get much past the 17-gon.

2025-05-13 16:20:10 UTC

- reply

(that should be "odd-sided regular polygon", since doubling the number of sides of a constructible polygon can always be done)

2025-05-13 16:10:21 UTC

- reply

Recently I was mildly freaked out to learn that the largest known (edit: ODD-sided) regular polygon constructible with compass and straightedge in the ancient Greek manner is the MaxUint32-gon.

2025-05-05 13:35:40 UTC

- reply

Right. The distribution of surface acceleration is going to depend on the spin of the body, which is actually different between the two cases, but the tidal stresses in operation are the same.

2025-05-05 13:24:32 UTC

- reply

it matters what we count as "tidal stretching" though?

2025-05-05 13:22:34 UTC

- reply

Either way, it's an object in free fall at the same distance (doesn't even really matter that it's in free fall, but it eliminates some confounding things). The background Riemann tensor is the same.

2025-05-03 16:42:06 UTC

- reply

I noticed you said a civilized *country* which in the UK has a pretty specific meaning...

2025-04-14 21:10:30 UTC

"Although it was known to Carl Friedrich Gauss by 1801 that the regular 65537-gon was constructible, the first explicit construction of a regular 65537-gon was given by Johann Gustav Hermes (1894). The construction is very complex; Hermes spent 10 years completing the 200-page manuscript."
https://en.wikipedia.org/wiki/65537-gon

2025-04-07 11:04:53 UTC

Seeing some buzz about the April 20th deadline for Trump's executive order on using the Insurrection Act to put down protests, and speculation that the swelling protest movement in the US is playing into his hands for totalitarian rule.
I don't *quite* agree. Here's a thing I noticed about the protests last weekend: the crowds are *way* older and whiter than I expected. It's a lot of grannies protesting about Social Security. And they're also decentralized--it's not just big rallies in the big cities, it's absolutely everywhere, it's podunk towns in red states.
Minorities and young people are staying home in part because they've already got the agent provocateur/crackdown scenario in mind. The speakers at the Boston Hands off rally already brought this April 20th date up. And what you're left with is a mass movement of people who cops do not instinctively want to beat up or shoot.
I don't think Trump expected that--he expected George Floyd redux. He's fighting the last war. He's malevolent and a fascist but he's not a chessmaster. He's also actively pissing off his billionaire base at the same time.
So I'm sure he'll *try* a lot of this. But the movement is trying hard not to play into his hands. When the army guns down some protesters, they won't be the people who Fox News viewers are scared of, they'll be 70-year-old white ladies and guys in veteran caps waving American flags on Main Street. Not the example you necessarily want to make.

2025-03-29 13:00:03 UTC

- reply

(note, I said particle *theorist*; if he'd been an experimental particle physicist responsible for building accelerators and detectors, he'd definitely care about that!)

2025-03-29 02:18:37 UTC

In, among other things, the genre of "questioning the unique genius of Richard Feynman", Kathy Loves Physics and History drops what may be her magnum opus:
https://www.youtube.com/watch?v=ipCeSTGpoLg
It's a history of the Poynting vector, which she thinks is poorly motivated... but the thing that annoys her beyond everything else was the way Feynman basically dropped the D and H fields in his treatment of Maxwell's equations.
I'm pretty sure I know why he did that: it's because he was a particle theorist, and didn't greatly care about the interaction of electromagnetic fields and bulk matter, so he just pretended everything was in vacuum for about 90% of his exposition. But that's exactly the assumption Kathy never wanted to make--and the historical development of the field definitely didn't make it, though, as she explains, they oversimplified it most of the time.
(As for the Poynting vector, it always seemed to me that the correct approach to energy flow in EM fields was through the stress-energy tensor of the electromagnetic field, and ideally you ought to be able to derive it from that... but I was never patient enough with the non-vacuum case.)

2025-03-26 11:06:48 UTC

- reply

I see a huge amount of YouTube advertising for Grammarly, and the focus of the ads seems to be on secondary and university students who speak English as a second language, to produce native-reading prose.

2025-02-27 04:27:31 UTC

@npub1h2y…3fw2
f(x) = 1/(1-x^2)

2025-02-26 01:57:50 UTC

- reply

So, it's "gee, why don't you just LEAVE your abusive boyfriend?" as an organizing principle of society.
With a bit of Neal Stephenson's Franchise-Operated Quasi-National Entities from "Snow Crash", which was not a utopian novel.

2025-01-28 16:45:27 UTC

- reply

Can't speak for the rest but to me it feels refreshing right now, people talking about something sensible and interesting instead of more doomy news.

2025-01-28 12:05:26 UTC

- reply

...This may be too specialist/inside, but I recall this 1993 review article on EFT by my old advisor Howard Georgi being pretty good, if you have a way of accessing it:
https://www.annualreviews.org/content/journals/10.1146/annurev.ns.43.120193.001233
It even mentions heavy quark EFT, which was what I actually did my doctoral thesis on!

2025-01-28 11:46:46 UTC

- reply

That's kind of what the renormalization group is, which is crucial to all this. It's how the coupling constants vary as you incorporate bigger or smaller "shells" as part of the fundamental interaction.
But what I was thinking more about is keeping the *variety* of "fundamental interactions" you have to keep track of in your model as you do this from multiplying to infinity, by ignoring things that (you think, at least!) shouldn't matter.

2025-01-28 02:40:00 UTC

- reply

Being careful about orders in some expansion parameter is also very much in the spirit of effective field theory, which is a collection of tricks for controlling the unknowns in situations where you have limited knowledge and a limited domain of application.

2025-01-28 00:22:48 UTC

- reply

That seems like it probably is the right way to do it.
Since I was going by dim memories of Jackson I hadn't even heard of the "reduction of order" approach, which sounds actually like a reasonable thing to do, and in this paper they seem to be giving it a justification on more fundamental grounds. The runaway solutions and pre-acceleration and such always seemed like they had to be artifacts, given that they were starting with a theory that was supposed to be causal and have energy conservation.

2025-01-27 22:09:12 UTC

- reply

Oh, it's along a free-fall geodesic from your current state of motion. Better time your exit just right or you'll be somewhere inside the earth.

2025-01-27 21:48:37 UTC

- reply

of course with GUTs there's still the issue that all this complexity kind of gets offloaded into the Higgs sector so you don't really have a small number of free parameters, but the way all the Standard Model coupling constants seem to converge at some high energy scale seems like it ought to be significant

2025-01-27 15:54:18 UTC

- reply

it's not right, but I have a hard time believing it wasn't in some sense on the right track. The further attempts to make a variant that worked, maybe less so.

2025-01-27 14:49:55 UTC

- reply

... there is some snippy discussion on the talk page; looks there was an edit war over the page title

2025-01-27 14:47:44 UTC

- reply

"...sometimes improperly termed a Green function"? I sense some kind of beef behind that phrase.

2025-01-27 14:04:27 UTC

- reply

someday when (if) people understand particle physics better they might look back and think of the Georgi-Glashow SU(5) GUT as another one.

2025-01-27 13:58:26 UTC

- reply

I'll have to take a look at that when I have a lot of time... this subject makes my head spin, and I think part of the problem is that most treatments display some of the paradoxical results, throw up their hands and say "this isn't real anyway, we have to go to the quantum theory," which is a really weak response. Is it just that classical electrodynamics is an inconsistent theory? If so they haven't really shown how.
I always assumed the weird stuff like runaway solutions and anti-causal influences in the Abraham-Lorentz theory were really the price paid for simplifying approximations in the calculation of the self-force, but I haven't seen a rigorous treatment of where they sneak in. And might not have the patience to fully understand it if I did.

2025-01-27 12:37:57 UTC

- reply

I don't know much about it but I'm pretty sure Feynman came up with the path integral in that context, so it had an incredibly far-reaching consequence.
(Though Feynman himself pointed out that the basic idea came from Paul Dirac, tossed out as a passage in a textbook--Dirac just never publicly developed it very quantitatively.)
I'm fascinated by these "close but no cigar" theories that often have some simplifying or illuminating characteristics compared to the one that eventually became dominant. The Bohr-Sommerfeld quantum theory that we were discussing several years ago in another social-media universe is another one.

2025-01-27 12:32:11 UTC

- reply

When I saw the movie "TENET" I remember thinking "I already read Greg Egan's version of this and he did it better."

2025-01-26 22:24:22 UTC

- reply

Certainly all the problems of self-energy and self-force that would bedevil QED first appeared there, and actually were even worse in the classical world.

2025-01-26 22:20:53 UTC

- reply

I vaguely recall Feynman saying the half-advanced-half-retarded one came up in the Wheeler-Feynman absorber theory--in many ways that was version 0.5 of Feynman's formulation of QED.

2025-01-26 18:19:30 UTC

- reply

And weirder things. Think of potentials in Coulomb gauge: The electrostatic potential component is governed by an instantaneous Green's function like in Newtonian gravity. But it's equivalent to any other gauge choice including some that make causality explicit.

2025-01-26 14:51:43 UTC

- reply

I find them pretty intuitive as long as they're constructed to focus on causality: it's "what happens if I poke the system HERE?"
And then to the extent that you can model the responses as linear, or handle nonlinearities with perturbation theory, you can derive a lot from those "response to a poke at one point" results.
But in relativistic QFT it suddenly gets weird because the most mathematically elegant form of the Green's function is one that doesn't really respect causality--it involves influences "going backward in time" as well as forward. So it can be hard to see how a causal world still emerges from that.

2025-01-25 18:27:43 UTC

- reply

I remember spending a lot of time pondering it-- classically we think of the Green's function of a field as describing casual propagation, but in the quantum theory, the propagator is a time-ordered product but casuality is associated with the commutator. Two different kinds of product.
Auto correct really does not want me to say "commutator".

2025-01-25 17:43:11 UTC

- reply

that makes some intuitive sense to me, I mean, what is the propagator at spacelike separation? It's a measure of correlation of field values at two different places in the vaccuum. The sum of three legged diagrams gives the three-point correlation, etc.

2025-01-25 16:00:46 UTC

- reply

generally speaking: no measurement you make at point A is going to affect probabilities measured at point B outside its light-cone, because observable operators commute at spacelike separation (one of the basic properties of all relativistic quantum field theories).
But those measurements *can* affect things observed later on, at events late enough that classical information can propagate... and things like Bell's inequality prove that more is going on than just a slowly propagating hidden variable.
So this is all admittedly weird. Sometimes I phrase it as something like "quantum correlations can create nonlocally correlated *noise*... but you can't control it". Because if you could, either you could send messages backward in time, or relativity would break.

2025-01-18 22:17:03 UTC

- reply

(apparently scallop pie is a Tasmanian thing. Sort of figured it was a New England thing; sounds like it ought to be)

2025-01-18 17:44:08 UTC

- reply

yes, but even so, it's a tasty one

2025-01-17 13:54:28 UTC

- reply

I think you can intuitively see that the white lines are not geodesics just from the fact that the fish fins consistently meet in a four-way vertex on one side and a three-way vertex on the other. If they slide along the line, they've got to be turning toward the three-way vertex.

2025-01-14 04:22:08 UTC

- reply

well, I posted a terrible phonecam picture for everyone

2025-01-14 02:44:29 UTC

- reply

there is a lunar occultation of Mars going on right now, btw, I'd completely forgotten about it

2025-01-14 02:32:06 UTC

- reply

mostly I have fragmentary memories that I can supplement with Google and Wikipedia. But while I'm not a very energetic amateur astronomer in the sense of having telescopes and such, I was always really into it as a kid so I absorbed a lot of lore then.

2025-01-13 13:42:45 UTC

- reply

It used to be called the Ursa Major Moving Cluster, which you can sing to the Teenage Mutant Ninja Turtles song

2025-01-13 12:24:39 UTC

- reply

Astronomers once believed that Sirius was part of the Ursa Major Moving Group, but it's too young to be physically associated with the rest of them. Menkalinan (Beta Aurigae) and Alphecca (Alpha Coronae Borealis) are in it though.

2025-01-13 11:37:31 UTC

- reply

notice that 5 of those stars seem to be moving together-- that's the Ursa Major Moving Group, a collection of stars that may have once been a bound cluster. That's the core of it, but there's a wider collection of stars apparently associated with it that are all over the sky, because we and the Sun are passing through it, though not of it.

2025-01-13 03:52:35 UTC

- reply

It says the new one is a sub-Earth rocky planet with a 3-day orbit, so, nothing like the refuted 2018 claim. Or anything like van de Kamp's planets, which were Jupiter-sized giants.

2025-01-13 03:42:07 UTC

- reply

For those who didn't click through, that picture is an artist's conception of the planet, not an actual image--there are no images at all of most deduced exoplanets, and images of a few that show them as dots.
Largely because it's so near, Barnard's Star has the fastest visible motion across the sky of any star. It's moving fast enough that if *it* were bright enough to be visible to the unaided eye (which it is not), its position in the constellation Ophiuchus would have noticeably changed over your lifetime.

2025-01-13 03:25:28 UTC

- reply

It's also news to me that Barnard's Star actually has a planet--apparently the discovery of it was announced just a few months ago! Barnard's Star has an old history of spurious exoplanet claims. Peter van de Kamp claimed he'd found evidence of some in the 1960s!
https://en.wikipedia.org/wiki/Barnard%27s_Star_b
https://en.wikipedia.org/wiki/Barnard%27s_Star

2025-01-13 03:04:26 UTC

- reply

This is a particular bugbear of the science-fiction critic James Nicoll--turns out there are a lot of science-fiction novels that blithely assume that Alpha Centauri is still the closest star to Earth millions or billions of years in the past or future. "STARS MOVE" is one of his go-to phrases.

2025-01-13 01:04:34 UTC

- reply

so we can combine the "star registry" scam with the nearly identical "Established Titles" scam--call me the Laird of HD 203291...

2025-01-12 23:36:19 UTC

- reply

I recall an old kid's book on astronomy that described Alpha Centauri as "actually three stars close together", but from a visual perspective that's not quite right: Proxima's orbit is so huge that if it were bright enough to see with the naked eye, it wouldn't even look like part of a multiple star system; it'd just be another star in Centaurus. We can't resolve it easily because it's dim.

2025-01-12 23:25:48 UTC

- reply

My reaction to this was "wait a minute, since when were 'Rigil Kentaurus' and 'Toliman' specifically the names of α Cen A and B , rather than just being two different old names for the A/B pair?"
Turns out the answer is: since 2016 and 2018 respectively, by somewhat arbitrary decision of the IAU, since they decided to start blessing proper star names. It's elegant, though perhaps not ratified by history.

2025-01-07 04:29:20 UTC

- reply

...it used to frustrate me that it was so hard to explain why "pushing off of the quantum vacuum", as one of the shifting explanations of the EmDrive claimed it was doing, wasn't a thing.
(Regardless of how many virtual particles you use to describe it, the quantum vacuum is a Lorentz symmetric state, so it has no net momentum. If you were to push off of it, you'd have to give it some momentum, which means you're changing it to a state that has some momentum. Which means there are quanta in it now. You just made a normal rocket, maybe a photon rocket. Not a thing that moves without emitting stuff.)

2025-01-07 01:51:58 UTC

- reply

I remember Gravity Probe B for how astonishingly long-delayed it was. I read about its fantastically precise gyroscopic spheres being fashioned when I was a kid in the 1970s, and they'd describe it as an experiment to be run soon when the Space Shuttle was operating. It finally launched in 2004.

2025-01-07 00:12:59 UTC

- reply

bullshit needs to be called out, regardless of who has the guns.
Physics is what I actually know something about but the "COVID lab leak hypothesis" gave me that "taking crazy pills" feeling: the way all the virologists and epidemiologists who seemed to actually understand the subject said one thing, and the cops and spies and miltiary analysts said a different thing, and the papers all assumed the second category were the ones who were right because they know secret stuff, don't they?

2025-01-06 22:42:15 UTC

- reply

Reminds me of that long stream of papers claiming positive results for reactionless space drives like the "EmDrive". If all you were reading were the headlines, you'd get the idea that there was increasingly solid consensus that this was real. If you read the actual papers, the results all contradicted each other and all they had in common was that they were right at the limit of sensitivity of the experiment, and when they improved the controls, they'd get another "positive" result that was marginal for the new controls, which would be touted as a replication.
Meanwhile, the theoretical justifications made no sense, and the discussions would spin off into wild science-fiction speculations about applications. Classic pathological science.

2025-01-06 18:00:48 UTC

- reply

like I said, the one before this was the "dark energy is hidden inside black holes" one, where my immediate reaction was "that doesn't work, does it? I mean, Alan Guth was studying what amounted to a black hole full of dark energy way back in the 80s and it doesn't do that! The no hair theorem still applies, right? From the outside, it's just a regular black hole!"
But, again, it got all this hype.

2025-01-06 17:23:56 UTC

- reply

Thanks, I hate having to bend over backwards to try to make sense of one of these stories where there's a claim that doesn't sound like it makes any sense, but everyone acts as if it does so I have to try to figure out what's going on and if there's anything to this.
The claim that every conventional cosmologist is making an elementary mistake (ignoring the effect of gravity on clocks) should have been a red flag.