j_bertolotti

Associated Professor of Physics at the University of Exeter.
Maker of typos.
Scientific visualizations (grouped under the hastag #PhysicsFactlet).
He/lui/on. All opinions are my own fault.

University Webpage: https://physics-astronomy.exeter.ac.uk/staff/jb601?sm=jb601
Google Scholar: https://scholar.google.co.uk/citations?user=wOejDGAAAAAJ&hl=en
Orcid: https://orcid.org/0000-0002-7163-6343
Website: https://jacopobertolotti.com

Public Key

npub1up9tsuaz59knsxg3al42xglqv0q89p4dlpgdwegj95nj8277qy3qr35w5m

NIP-05 Address

[email protected]

Profile Code

nprofile1qqswqj4cww32zmfcryg7l64ry0sx8srjs6kls5xhv5fz6fer400qzgspz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dsx2l5lg

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub1up9tsuaz59knsxg3al42xglqv0q89p4dlpgdwegj95nj8277qy3qr35w5m

Show more details

Last Notes

2026-05-26 08:26:32 UTC

- reply

You can do Parmigiano too, if you know how. You can't call it with the name for legal reasons, but nothing prevents you to learn how to make it.
My question is more about why. Gondolas make sense in Venice for historical reasons, but make very little sense anywhere else.
People from different places came up with different ways of doing things and different styles, which is why travelling around is interesting.

2026-05-26 08:10:07 UTC

- reply

Why is there a gondola? 🤔

2026-05-06 08:33:49 UTC

- reply

Regular reminder to self that all those scary for-loops are just iterated operators 😜

2026-04-09 10:55:24 UTC

- reply

Just think about all the implied knowledge in any cookbook. Recipes can be compressed to a page or two only because they assume you already know how to do a lot of the stuff they are talking about. If you have no idea how to cook, no amount of cookbooks will ever help you.
@npub1zcg…4xdu @npub1fdr…lvhs

2026-03-26 13:18:37 UTC

They convinced me to make a LinkedIn account.
Now I feel a bit dirty.

2026-02-28 09:01:51 UTC

- reply

I hadn't read anything from SCP in years. Probably I should start again 🙂

2026-02-25 09:41:42 UTC

- reply

Do they still produce/sell CRTs?

2026-02-13 20:13:52 UTC

- reply

I feel for the developers, but making a free-to-play life service hero shooter is equivalent to jumping in a tank full of sharks shouting "I'm tasty!"

2026-02-13 10:12:20 UTC

Random reminder that, despite what you read in "Lord of the Flies", when a real group of teenagers happened to get stranded on a inhabited island, they didn't start killing each other. On the contrary they worked together, built a commune, and survived in good health for 15 months (when they were saved by a passing ship).
Most people are good people. Even teenagers 😉
https://en.wikipedia.org/wiki/Tongan_castaways

2026-02-02 16:15:24 UTC

- reply

Counterpoint: noting in the game is photorealistic, so you might want to make the mountains less "realistic" to better fit with the general vibe.

2026-01-07 15:27:45 UTC

- reply

De gustibus non est disputandum 🙂
@npub1dxf…v5xz

2026-01-07 13:47:28 UTC

- reply

I know I am part of a minority, but I like to derive Lagrangian mechanics from newton's laws instead of postulating that the action is minimised. Which means that to me all Newton's laws are built into the fabric of Lagrangian mechanics by construction 😉
@npub1nf4…nqe4

2026-01-02 14:37:07 UTC

- reply

Exactly! And Purgatory has the same shape, but reversed to form a mountain.

2026-01-02 14:30:56 UTC

- reply

Funnily, hell is not described as spheres, but as concentric rings ("cerchi"), with the smallest one being also the deepest.

2025-12-26 19:45:26 UTC

- reply

Reminds me a lot of how Wikipedia was discussed in the media 20 years ago.
@npub1yq0…hfrz

2025-12-26 18:36:04 UTC

- reply

There are infinite cool things to learn and a finite time to learn them. You have to prioritise 😉

2025-12-26 12:31:18 UTC

- reply

I didn't mean to offend you. If I did, I apologize.
I understand this stuff on an intellectual level, but I never devoted nearly enough time to music to make this stuff less confusing at an intuitive level. There is just too much historical baggage.

2025-12-26 12:16:53 UTC

- reply

The very fact that people decided to divide notes into "white" and "black" instead of simply say that they are ultimately the same thing is also something that will confuse me forever 🤣

2025-12-26 12:02:31 UTC

RE: https://mathstodon.xyz/@johncarlosbaez/115779744672820818
The fact that going up 7 notes is called a fifth will never cease to confuse the hell out of me.
#IAmBadAtMusic
#note1vm3…7hh3

2025-12-18 11:36:55 UTC

- reply

After not touching it for >1y, I used this occasion to deactivate it (it will take 30 days before it is deleted).

2025-12-15 12:44:32 UTC

- reply

https://www.unicodeit.net/ has saved me so many times.
@npub1whr…qc2t

2025-12-10 15:00:29 UTC

RE: https://mas.to/@kityates/115695305011695598
My new business plan:
* Code a thinly disguised random number generator.
* Name it something like "The Eye of Sauron"
* Advertise it as the most cutting edge quantum AI technology to scrape someone's social media history and extract their political inclinations.
* Sell it to the US government for a huge sum.
(Sometimes I wish I was sociopathic.)
#note1mk8…kjte

2025-12-02 14:40:17 UTC

Stuff like this is why I hate generative AI with a passion.
https://www.theguardian.com/technology/ng-interactive/2025/dec/02/jared-kaplan-artificial-intelligence-train-itself
We know for a fact that feeding a generative AI model with the output of generative AI lead to the model collapse and nothing more ( https://www.nature.com/articles/s41586-024-07566-y ), but the claim is not just factually incorrect, it is maliciously incorrect. Because they know it is false, but they will pretend otherwise to try to keep the speculative bubble afloat and the money flowing in.
And the whole article is nothing else but the "journalist" repeating the claims of someone with a very clear vested interest without ever fact-checking or even questioning them.
It is rage-inducing.
#AI #GenerativeAI #Anthropic #rant

2025-11-27 07:57:27 UTC

- reply

If I split it into 3 parts, can I then recombine them to obtain 2 spheres identical to the original?

2025-11-10 11:53:07 UTC

- reply

Either would be correct, and which one is preferred depends on the particular teacher (I expect most would want a student to write 185/43, but I haven't been in primary school for a long time).
@npub1s4h…qs84

2025-11-10 11:26:02 UTC

- reply

As the Italian native allow me to fight back: "two and a half" translates to 2+½=2.5. The plus there is very explicit.
The Fellini movie is universally called "Otto e mezzo" which translates to 8+½=8.5 and the way it is written in the movie posters is a complete anomaly in Italian (and likely done for artistic reasons).
Mixed numbers DO NOT exist in Italian. They are not taught in school, and they keep confusing me every time I see one.
Then obviously we are not stupid, and given some context we understand what they mean, and I am sure a few people have used them here and there because they felt it makes them look cosmopolitan.
@npub1s4h…qs84

2025-11-10 10:53:53 UTC

- reply

Minor correction: mixed numbers do not exist in MOST countries.
They are also an abomination that should be burnt with fire.

2025-11-04 10:35:16 UTC

- reply

I wonder if Meitner ever crossed paths with Noether. Both Jews women in Germany struggling to have their genius recognized by a close-minded University system. Both having to flee Germany. Roughly the same age.
They were interested in very different problems, but I can only imagine what they could have achieved if they ever worked together!

2025-10-18 07:40:14 UTC

There is a girl who started brining her knitting to my class, and now I wish I could spend a few lectures on knot theory and (some of) the beautiful Maths behind knitting 🧶

2025-10-13 10:49:52 UTC

#PhysicsFactlet
Sometimes you need to forgo the intuitive way to define stuff for the sake of actually being able to do anything useful with those definitions.
An example of this I always found funny is in knot theory, where a simple loop is considered to be a knot, while anything where the two extremes are dangling are not, including the common overhand knot.
This looks weird the first time you see it, but there is a very good reason to go with such a definition: you want to study what you can and can't do by manipulating the knot, and if you have the two extremes dangling, you can always untie any knot, making them all equivalent to a piece of string. In order to be able to say anything interesting about them you need to remove this trivial option, and thus accept the simple loop as a knot.
#knots
https://media.mathstodon.xyz/media_attachments/files/115/366/428/534/174/821/original/0c89383034a0cc2c.png

2025-10-07 16:50:50 UTC

- reply

A "Clifford algebra for dummies" would be nice.
(I am the dummy.)

2025-09-22 15:41:26 UTC

- reply

The number of markers we can put on a ruler is limited by the fact that atoms have a finite size, so I guess in your higher-cardinality universe atoms would have a hard time existing, which would prevent us from existing and thus from doing any logic at all 🙃
@npub1p7u…kglz @npub1d60…mhwm

2025-09-11 09:21:19 UTC

- reply

I used it for a while. It is a good client for mastodon (now I use the vanilla browser version of mastodon)
That said, not all mastodon servers have the latex plugin installed.
[@victor_tokarev](https://twiukraine.com/@victor_tokarev )

2025-09-10 12:58:45 UTC

- reply

I vote for "This holds everywhere, except in a set of measure zero, which happens to contain every single point where this result would be interesting and/or useful" 🙃

2025-08-22 15:09:06 UTC

Pet peeve number 10²³: People who say Japanese is difficult because it has 3 different sets of characters.
It might surprise some, but every language using the Latin alphabet has 4 sets of characters: lowercase block, uppercase block, lowercase cursive, uppercase cursive.
(To be clear: Japanese IS difficult. But not for that reason!)

2025-08-18 13:30:24 UTC

Incoming conference!
@npub1xwn…dqus put together an amazing list of people. Then made a mistake and added my name too 😁
https://www.grc.org/image-science-conference/2026/
If you are interested in #Optics and #Imaging, you probably want to come.

2025-08-14 09:00:35 UTC

"This is why AI users often sound like they're using totally different technologies. Workers who get to decide whether and how to incorporate AI into their jobs are doubtless finding lots of utility and delight from the new tool. These workers are "centaurs" – people assisted by machines.
The workers who describe their on-the-job AI as a hellish monstrosity are being ordered to use AI, in workplaces where mass firings have terrified the survivors, who are told they must use the AI to make up for their jobless former colleagues. They are reverse-centaurs: machines assisted by human workers."
( @npub1fdr…lvhs )
https://pluralistic.net/2025/08/13/then-they-came-for-me/

2025-07-18 14:27:41 UTC

- reply

The lesson from Italian history is not one with a happy ending. When Berlusconi collapsed under the weight of his age and scandals, things didn't get "back to normal". What happened was that the bar of "normality" have been moved forever, the opposition was essentially cancelled, and the power vacuum was filled by populists.

2025-07-18 14:22:56 UTC

It is interesting/horrifying how Trump seems to be following exactly Berlusconi parabola (including a scandal involving underage girls and having people he didn't like fired from TV), but cranked up to eleven.

2025-07-14 16:36:10 UTC

- reply

#PhysicsFactlet
An improvement over the bisection method is the so-called "false position" method, where instead of dividing the bracket region in two, you cut at the point where the line through the two bracket extremes crosses zero.
#Physics #Computing #Algorithm
https://media.mathstodon.xyz/media_attachments/files/114/852/544/088/999/242/original/6963e9fe05b290ad.mp4

2025-07-11 12:57:26 UTC

#PhysicsFactlet
Finding the roots of a function is a very common problem in computational Physics, and the bisection method is a simple and effective (albeit far from optimal) way to do that.
The idea is that you start by "bracketing" your root between a value of x where the function is negative and one where it is positive. You then take the midpoint between them, check if your function there is positive or negative and update the bracket.
#Physics #Computing #Algorithm
https://media.mathstodon.xyz/media_attachments/files/114/834/690/273/158/644/original/184ee8de52e881a8.mp4

2025-07-10 08:43:53 UTC

- reply

The humanities are a foreign country to me 🙃
@npub1gt0…keck

2025-07-10 08:27:44 UTC

- reply

Preprints are also a way to make your research accessible to academics in poorer countries (and/or people doing research but without an academic affiliation, e.g. in industry) without having to pay the enormous open-access fees many journals have.
@npub1gt0…keck A good starting point when looking at preprint archives in one's field is https://en.wikipedia.org/wiki/List_of_preprint_repositories

2025-07-05 12:26:59 UTC

- reply

These are all great, but they all aim to simulate the optical system. My idea was something simple that would allow you to make 3d diagram of your setup (e.g. to use as a figure in a paper) without all the overheads of a real simulation.

2025-07-05 09:11:11 UTC

Hobby project I started, never finished, but I would like one day to pick up again:
* A statistics and probability manual that uses #DnD rules as both motivations and examples.
*A simple managerial videogame where you are a newly appointed academic and need to run your group, juggle teaching, admin etc
* A #GURPS-inspired ttrpg set in a world where magic is so common that everybody can use it.
* An app that allows you to make 3D diagrams for experimental setups just by drag and drop (without having to learn all the intricacies if Blender&co).
* An app that shows you the probability distribution to get a given result when rolling dice (e.g. roll 4d12 and keep the lowest 2).

2025-06-06 13:39:30 UTC

- reply

Look mom, I got a semi-viral post on Mastodon!
I don't have a soundcloud (or whatever people who regularly go viral tend to advertise), so just be nice to each other!

2025-06-06 08:50:06 UTC

"AI companies claim their tools couldn't exist without training on copyrighted material. It turns out, they could — it's just really hard. To prove it, AI researchers trained a new model that's less powerful but much more ethical. That's because the LLM's dataset uses only public domain and openly licensed material."
tl;dr: If you use public domain data (i.e. you don't steal from authors and creators) you can train a LLM just as good as what was cutting edge a couple of years ago. What makes it difficult is curating the data, but once the data has been curated once, in principle everyone can use it without having to go through the painful part.
So the whole "we have to violate copyright and steal intellectual property" is (as everybody already knew) total BS.
https://www.engadget.com/ai/it-turns-out-you-can-train-ai-models-without-copyrighted-material-174016619.html?src=rss

2025-06-04 08:45:03 UTC

- reply

Addendum: if you are writing something longer than 2-3 pages and you have more than a couple of acronyms defined in your text, seriously consider a table of symbols and acronyms somewhere so that people don't have to hunt where did you defined it to understand what you are talking about. 🙂
@npub1qhq…vwcf

2025-06-03 12:31:33 UTC

Acronyms are the worst form of jargon, as their meaning can be completely different in different fields.
(I just saw a post talking about FLT as in "Fermat Last Theorem" where my brain keeps reading it as "Faster than Light Travel", and this is one of the most benign cases.)

2025-06-02 15:16:17 UTC

#PhysicsJournalClub
"Model-free estimation of the Cramér–Rao bound for deep learning microscopy in complex media"
by I. Starshynov et al.
Nat. Photon. (2025)
https://doi.org/10.1038/s41566-025-01657-6
As everybody who ever tried to orient themselves while immersed in thick fog knows, scattering scrambles information. The question "how much information is still there?" is not particularly interesting as the answer is "essentially all of it", as elastic scattering can't destroy information. A much more interesting question is "how much information can we retrieve?" In order to even try to give an answer we need to be a bit more specific, so the authors placed a small reflective surface behind a scattering layer and asked how much information about its transverse position could be retrieved. This is a well-posed question, and the answer takes the form of a "Cramér–Rao bound" (https://en.wikipedia.org/wiki/Cram%C3%A9r%E2%80%93Rao_bound).
After estimating this upper bound, the authors investigate how well a trained neural network can do at this task, and show that a specifically built convolutional neural network can almost reach the theoretical bound.
[Conflict of interest: Ilya Starshynov (the first author) did his PhD in my group.]
#Physics #InformationTheory #Optics #MachineLearning

2025-05-23 15:54:49 UTC

#PhysicsFactlet
A Shack-Hartmann sensor is made my an array of small lenses and a camera. If the light hitting the lenses is collimated, we will get a number of equispaced foci on the camera. But if the light is not collimated, the position of the foci will change in a predictable way, so we can reconstruct where the ray were coming from.
#Optics #Physics
https://media.mathstodon.xyz/media_attachments/files/114/557/948/986/672/808/original/f79c4870825557b5.mp4

2025-05-20 17:09:27 UTC

- reply

That is plenty, thank you 🙂

2025-05-20 16:53:36 UTC

- reply

Nice article! (As always) 🙂
Just a super-naive question from a low-energy experimentalist who never properly stidied GR: givel that energy is not (globally) conserved in GR, aren't arguments like "this violates energy conservation so it can't work" a bit self-contradictory? What (likely trivial) piece did I miss?
@npub1nf4…nqe4

2025-05-19 09:40:19 UTC

- reply

Already added ;-)
(A bit small, but that page already has way too many images for the small amount of text)

2025-05-19 08:41:58 UTC

- reply

Finally managed to upload it! https://commons.wikimedia.org/wiki/File:Stone-Wales_defect_in_Fullerene.gif

2025-05-15 13:18:18 UTC

- reply

Imho video and text have very complementary advantages and disadvantages. Video is more immediate, but when you are dealing with complex stuff nothing beats having it spelled down in writing.

2025-05-15 10:20:55 UTC

Why is everybody doing science explainers in the form of youtube videos, instead of just writing stuff down (which is much more accessible and long-term stable)? 😫
#AcademicChatter #SciComm

2025-05-12 16:31:58 UTC

- reply

Travelling tomorrow, but will upload it and link it as soon as I find a bit of time 🙂

2025-05-12 14:51:50 UTC

- reply

Third time's a charm? 🙂
@npub1xuv…h8pl
https://media.mathstodon.xyz/media_attachments/files/114/495/420/743/377/392/original/98a410848799a586.mp4

2025-05-12 12:42:36 UTC

- reply

Would it make sense to always keep the red part in front and make it rotate only 180 degrees forth and back?

2025-05-12 10:40:00 UTC

- reply

Any better?
https://media.mathstodon.xyz/media_attachments/files/114/494/429/577/960/394/original/7f06130c5bef197c.mp4

2025-05-11 11:07:56 UTC

- reply

Shared on bsky with my "active particles" expert (Giovanni Volpe) to see if he has an understandable explanation.

2025-05-09 09:40:19 UTC

- reply

draft number 1. Feedback is welcome 🙂
(I know the animation will benefit from being smoother, but I won't generate a ton of frames for a draft)
https://media.mathstodon.xyz/media_attachments/files/114/477/201/989/778/967/original/032bbb7157376e68.mp4

2025-05-08 13:20:47 UTC

- reply

I am truly happy you had plenty of positive interactions with them. I didn't. I've never seen them before and this was not the best first impression.

2025-05-08 09:38:02 UTC

- reply

Have you considered asking me "why do you think blue noise is hyperuniform?" instead of presenting me with an annoyingly ill-formulated puzzles like I was a middle-schooler?

2025-05-08 09:21:25 UTC

- reply

Counter-question: what are you hoping to achieve by jumping in my mentions asking badly formulated questions?
Yes, our brain is wired for pattern-recognition and doesn't like white noise very much. That is extremely well-known.
Also: estimating the underlying distribution from just a couple hundred samples is bound to have an enormous uncertainty even if you can run a full statistical analysis.

2025-05-08 09:08:32 UTC

- reply

They are all random. They are just (randomly) sampled from different distributions.
Not all randomness is white noise.
@npub1hkl…9agh @npub1nf4…nqe4

2025-05-08 08:53:35 UTC

- reply

: is blue noise hyperuniform? (I think it is, but you are the expert)
@npub1nf4…nqe4

2025-05-06 13:35:42 UTC

- reply

"but whenever I try to explain it online, a bunch of men get offended" which is not independent from the reason why it is predominantly men 😉
@npub1ngl…2ylu

2025-05-06 08:08:08 UTC

- reply

No. It was a VERY stupid mistake on my part.

2025-05-06 07:59:50 UTC

- reply

Pity I am now locked out of my Wikipedia account (due to a stupid mistake during a password update). Will have to create a new one.

2025-05-06 07:55:54 UTC

- reply

If I find a bit of time I might take up the challenge 😁

2025-05-05 10:39:43 UTC

- reply

Depressing but not surprising.
(Deleting anything that smells of crackpottery on sight is a strategy that has served me well through the years)

2025-05-04 09:05:54 UTC

- reply

They should make him a honorary retired engineer 🤭

2025-04-23 21:01:10 UTC

- reply

@npub1wsp…rhgk might have an opinion

2025-04-01 15:49:14 UTC

- reply

All those NDAs expiring on April 1st! 😃

2025-03-20 07:50:57 UTC

- reply

One more reason to move ALL scientific conferences out of the US.

2025-03-12 16:15:57 UTC

- reply

The curse of double negatives in English. I genuinely don't know which way is the correct one 🤷‍♂️

2025-03-12 16:09:15 UTC

It truly drives me mad hearing US people claiming that, since their system clearly didn't protect them from a rich guy taking over, it is impossible to design a system that is resilient to rich guys attempting to take over.
Like similar take-overs weren't attempted in other countries before, and they didn't fail because the system in those countries was indeed resilient to such things.

2025-03-12 07:39:43 UTC

- reply

Anything bigger than a standard Russel's teapot is a moon 🙃

2025-03-12 07:33:48 UTC

- reply

Is The Asylum a sort of modern day Troma?

2025-03-05 19:09:40 UTC

- reply

She definitively knows more general relativity than most scientific journalists, and she has done a lot of good videos through the year. But (and I speak as someone that until a week ago ised to watch essentially all her videos) the frequency at which she is publishing videos that are 90% techbro apology is growing way too fast for a ton of alarm bells not to start ringing in my head.
@npub1pkc…qy0t @npub1x78…xn0m @npub17lw…5ctv @npub1nf4…nqe4 @npub1glg…tqzy

2025-03-02 18:42:38 UTC

- reply

The video is MUCH worse then it looks from the tiny piece you are quoting. @npub1nf4…nqe4 @npub1glg…tqzy

2025-03-02 17:14:12 UTC

- reply

The sad part is that she used to be very reasonable (although I never appreciated her use of "Physics" to mean "the tiny sub-sub-fieldnI used to work in")

2025-02-28 11:05:47 UTC

- reply

Isn't the very fact that 20k games were released part of the problem? Even if you are a good-willing game journalist you are unlikely to be able to try but a tiny fraction of those 20k games, and you have to decide if to devote your limited time to try a game nobody is talking about, or a game you were told by many that is a GOOD indie game.

2025-02-03 17:10:37 UTC

- reply

But at least category theory to design better software for epidemiology is a new and exciting topic! Anderson localization is fascinating, but people have banged their head on it for decades 🙂
Anyway, I can think of 3-4 people who probably work on it most of their time, but I am sure that there is a lot of people like me who have it in the back of their mind but don't do anything with it.

2025-02-03 16:57:54 UTC

- reply

I mean that nowadays there isn't a lot of people working on it. It was a very popular topic at a certain point, but it got out of fashion and people moved to other problems 🙂

2025-02-03 16:40:46 UTC

- reply

Can this be circumvented? Possibly yes. Having correlated disorder seems to be a way. Working with systems where no longitudinal component of the electric field is present is another.
This paper is the latest in this very long story, and show simulations suggesting the a systems made by conductive (non absorbing) spheres might do the trick.
Not sure how easy it is to do this experimentally (spoiler: not at all!) but interesting 🙂
3/3
https://doi.org/10.1103/PhysRevLett.134.046302

2025-02-03 16:40:36 UTC

- reply

What about 3D? 3D turned out to be more complicated, as the theory tells us that Anderson localization in 3D is not guaranteed, but can only happen if the system is disorder "enough" (without a well specified definition of "enough"). For (ultra)sound it took until 2008 to get an unequivocal proof that Anderson localization was happening, but nobody had any problem with it once it was published. [https://www.nature.com/articles/nphys1101]
For light the story is a LOT more complex. To the best of my knowledge the first claim of Anderson localization of light in 3D is from a 1997 Nature paper, that was received with some scepticism by the community. [https://www.nature.com/articles/37757, https://www.nature.com/articles/18347]
In 2012 the very group that put the first result into question published their own "proof" of Anderson localization of light in 3D. This was also received with scepticism. [https://www.nature.com/articles/nphoton.2012.313, https://www.nature.com/articles/nphoton.2013.210]
So, by the mid '10s the situation was that there were a few competing claims from well regarded groups but nobody in the community believed in either of them.
(Notice that nobody ever claimed the data in the papers were falsified. It has always been an argument about how to interpret those data).
Then, in 2014, Skipetrov published a theoretical paper suggesting that, under normal condition Anderson localization for light can't happen in 3D.
The problem is that at the levels of disorder necessary, the scatterers are in the near field of each other. This breaks some of the underlying necessary assumptions for Anderson localization to exist. [https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.112.023905]
2/

2025-02-03 16:40:00 UTC

#PhysicsJournalClub
"Anderson Transition for Light in a Three-Dimensional Random Medium"
by A. Yamilov, H. Cao, and S. Skipetrov
Phys. Rev. Lett. 134, 046302 (2025)
Anderson localization is a phenomenon where wave interference in a disordered scattering medium results in the wave being unable to propagate. It is a fascinating, complex, and VERY niche topic.
It is also the centre of never-ending arguments.
Anderson localization was originally proposed as a mechanism to explain why certain spin states in metals full of impurities would take longer to relax than expected (the answer was: their quantum state could not propagate due to the disorder). [https://www.nobelprize.org/uploads/2018/06/anderson-lecture-1.pdf]
It was soon realized that electron-electron interaction would dominate over a pure "Anderson localization" in most solid state systems, where a similar effect ("Mott insulation") would dominate.
The topic had a resurgence when it became clear that Anderson localization was a general wave phenomenon, and it wasn't restricted to electrons, but could in principle happen to any wave, e.g. sound or light. [https://doi.org/10.1103/PhysRevLett.58.2486]
Due to its nature, Anderson localization is easier to achieve the lower the dimensionality of the system is. Anderson localization of (ultra)sound in 1D and 2D were quickly discovered, with light coming a few years later.
1/

2025-01-28 18:40:21 UTC

- reply

Most people I know struggled with even the move from Twitter to Bsky, despite the fact that they are very similar. Moving to Mastidon requires a lot of extra learning, and moving to any other Fedi service is largely impossible if you are not already in the know, as there is no easily accessible explanation on how they work from a user standpoint.

2025-01-28 11:52:41 UTC

- reply

Do you know of any "for experimentalists/ for dummies" kind of introduction on how renormalization/ effective field theory work? 🙂
@npub1vlc…mzqq @npub1nf4…nqe4 @npub1s4u…7vpz @npub12cu…d38p

2025-01-28 07:41:06 UTC

- reply

This kind of discussions is the reason why it is worth being on social media ♥️

2025-01-26 19:37:05 UTC

- reply

In the sub-sub-field "light multiple scattering" we use Feynman diagrams a LOT! We do it for a massless field, but the formulas we get are very similar to the ones in condensed matter.
@npub1vlc…mzqq @npub12gn…d572 @npub1s4u…7vpz @npub12cu…d38p

2025-01-26 18:39:24 UTC

- reply

After working with it for 15 years I reached the conclusion that classical electrodynamics is cursed.
@npub1s4u…7vpz @npub1nf4…nqe4 @npub12cu…d38p

2025-01-26 17:18:39 UTC

- reply

Classical electrodynamics also has both advanced and retarded Green's functions. 🤔
@npub1s4u…7vpz @npub1nf4…nqe4 @npub12cu…d38p

2025-01-25 18:24:07 UTC

- reply

The trick is to learn about propagators in classical electrodynamics, where they are just as complicated but look less weird and intimidating 😉

2025-01-25 18:20:48 UTC

Feels like the right time to start planning for moving all international scientific conferences to somewhere outside the US.

2025-01-24 11:10:38 UTC

- reply

https://media.mathstodon.xyz/media_attachments/files/113/883/018/208/883/008/original/80b43cc30b979be6.png

2025-01-22 09:59:28 UTC

- reply

Reviews of old, obscure games are to me what cat pics are to a lot of people: relaxing palate cleansers from all the negativity 🙂