The h Bar

Slereah

14:00

Heraclitus apparently means "glory of Hera"

qwerty

hmmmmm

Amit

As always, "Cherchez la femme..."

Ryder Rude

Heraclitus thought everything is changing

modern physics would encode that on the phase space

Zeno thought if u pause time, things aren't moving

so nothing is moving

but phase space again resolves that

things are moving at paused time

seems like a dose of phase space is what early philosophers needed

ACuriousMind

@Slereah Every time I see this strange subculture of Hegelian category theory, I can't help but think that either these people have understood something so profound they're cursed to be unable to communicate it to ordinary mortals or they've gone completely off the rails

might be both

Slereah

@ACuriousMind why not both

Amit

14:07

ACM is a programmer! "or" always means both is not excluded ;)

Slereah

It seems to be interesting so far but not groundbreaking

Like they're just finding the cool levels of a topos

Amit

I can be a first class defense attorney

Slereah

Neat but also I'm not getting a nobel prize out of it

It makes for a nice struturation of a topos I guess?

And I guess you have the very important Semantics of Objective Logic

Ryder Rude

also, the hype about topos is niche

Slereah

I'll try to apply it to some cool topos and see if anything interesting comes out

Like a cool monad idk

Also there's a whole thing about how the Maybe Monad is fundamental for QM idk

Ironically not the "maybe" aspect of it

Despite the name it's just a monad that adds a point to things but you can make linear logic embedded in a topos

Amit

14:14

@RyderRude One of my personal commandments is "thou shalt not discourage a mathematician"... :D

Ryder Rude

@Amit u get Mozichuki this way

qwerty

what do people think of the amplituhedron stuff?

Amit

Idk what I'll get, that's basically the reason ;)

qwerty

(my qft knowledge is zilch)

Ryder Rude

@Amit no one still knows what Mozichuki got

@qwerty it is just Qm with fields

qwerty

14:17

@RyderRude yeah so?

ACuriousMind

@qwerty given that I haven't heard anything about it in a long time, not much at all :P

Ryder Rude

@qwerty it means, if u know qm, u have the basic idea of qft

recently, i finally felt that philosophy was deeper than physics

as in, physics is just models. philosophy is searching for the nature of everything

im considering learning philosophy formally

Amit

You should do it

Ryder Rude

yes

qwerty

@RyderRude I'm not in the habit of inflating my projected knowledge of a topic nor trivialising the amount of work needed to properly understand stuff

Amit

14:19

Even if only to be certain it's not for you ;)

Ryder Rude

@qwerty the statement i gave is the basic facts of qft. It is not trivialising it

@Amit i thought it is for me...

Amit

Studying something in an institution with classmates, also means you got people to bounce random ideas off of. Something I get the vague impression you're into.

Ryder Rude

@Amit yeah.. i always share when i have different ideas

Amit

Evidently

Ryder Rude

recently, i learned about the "perspective" problem of consciousness

qwerty

14:23

@ACuriousMind i think they got a (what seemed like - i dont know what the norm is) huge grant fairly recently, they were hiring postdocs and phds last year. started trying to read the work and figured I needed to learn qm and qft first

Amit

I gtg, catch you later

ACuriousMind

@qwerty ah, you mean ias.edu/news/erc-synergy-grant-universe

Slereah

While I can appreciate the Hegel stuff, no way I'm writing Aufhebung

Sublation all the way

ACuriousMind

I really don't have any opinion on the topic - Arkani-Hamed always seemed excited about it when I read or heard of it, but he never managed to convince me to particularly care about it one way or the other

qwerty

@ACuriousMind yup!

I guess tangentially I also was curious cos you said you were so disillusioned with your line of work... if you ever thought about other research paths and ideas, or maybe ideas of your own, or if you think theyre all gonna be dead ends and the "quiet period" of science is going to last a lot longer

@ACuriousMind they definitely seemed to convince some grant committee haha

ACuriousMind

14:37

I've tried reading the original amplituhedron papers several times, but I don't really get what's going on; it's an extremely strange mixture of trivialities and leaps of logic that is written in a carefree style so foreign to the mathematical and detail-oriented way I think about this kind of physics that I can't make heads or tails of it

qwerty

hmmm I see

ACuriousMind

These jokers show me a hand-drawn triangle and one page later they're talking about subspaces of Graßmannians and SYM amplitudes as if the hard part of understanding geometry is how triangles work

2

qwerty

that's disappointing

ACuriousMind

qwerty

I didn't realise it was so badly written

Slereah

14:42

@ACuriousMind that is the true wisdom

ACuriousMind

honestly, if I'd be asked to write a parody of overly self-important physics papers, it would be stuff like this: Triangulate a polygon and then say "Obviously, this is just like <at first glance completely unrelated technical topic>"

Slereah

@ACuriousMind My favorite one remains insti.physics.sunysb.edu/~siegel/parodies/sgs.html

> There's no such thing as fourth quantization, but if there were, it would be the same as the third-quantized one, due to the conformal symmetry.

qwerty

@Slereah ok that's hilarious

its also how some of my undergraduate lectures felt

Slereah

After I finish the Hegel stuff I guess I can try tackling the monster

arxiv.org/abs/1310.7930

qwerty

which only maybe probably slightly contributed to my long term fear of anything quantum

the complete non-sequiturs are perfect

Slereah

14:47

From what I can tell the topos where all the cool things happen is Smooth, and the actual topos that aren't just a mish mash of everything are slice topoi in Smooth

Ryder Rude

I hadn't heard of amplitudehedron

Amplituhedron

In mathematics and theoretical physics (especially twistor string theory), an amplituhedron is a geometric structure introduced in 2013 by Nima Arkani-Hamed and Jaroslav Trnka. It enables simplified calculation of particle interactions in some quantum field theories. In planar N = 4 supersymmetric Yang–Mills theory, also equivalent to the perturbative topological B model string theory in twistor space, an amplituhedron is defined as a mathematical space known as the positive Grassmannian. Amplituhedron theory challenges the notion that spacetime locality and unitarity are necessary components of...

qwerty

> Pythagoras, private communication.

lol

Slereah

it is a good paper

qwerty

obviously the hand drawn triangle was scanned in from a letter sent by Pythagoras himself

duh

Ryder Rude

@Slereah wtf

at least it doesn't say "obviously" :P

Slereah

14:55

> Theorem 1.56. The cohesive+elastic+solid homotopy type theory above has a faithful (i.e. non-degenerate) categorical semantics in the homotopy topos SuperFormalSmooth∞Grpd of super formal smooth infinity-groupoids.

that one also sounds like a parody but that's Schreiber

qwerty

@Slereah someone once told me that alice in wonderland's "the jabberwocky" is a perfect simulacrum of what it's like as a small child reading text above your reading level. This paper is EXACTLY what sitting through certain unlikable advanced lectures, and exiting with only your rushed scribbles for notes, felt like in the exactly the same way

Slereah

Well Lewis Carroll was a mathematician

Ryder Rude

@Slereah I'm not sure if this one is bad. things i haven't learned sound like this to me

qwerty

@Slereah yup

qwerty

15:09

ohh did you ever see this preposterousuniverse.com/blog/2006/12/12/the-string-kings

naturallyInconsistent

@qwerty we have folks here in h bar who are absolutely proud to be that, even when pointed out directly to their faces that this is the case.

Ryder Rude

what thing in physics did u all completely misunderstand initially?

i remember misunderstanding SR.... like how can each clock be slower than the other

ACuriousMind

@qwerty lol

If I had read this I forgot completely it existed

naturallyInconsistent

@ACuriousMind the string boys here had shared it multiple times...

qwerty

15:27

they won't be able to make that movie after all these years because younguns aren't interested in string theory and have never heard of lubos motl (maybe both for the best)

ACuriousMind

@naturallyInconsistent really? the transcript search does not find this link or "string kings" anywhere in chat history

naturallyInconsistent

@ACuriousMind that's more than a year ago; the transcript search does not go that far. Does it search the links too?

ACuriousMind

@naturallyInconsistent hmm? the transcript search absolutely covers all of chat history, I can still find this video having been posted ten years ago simply by searching for the video link

PM 2Ring

15:42

Last week, we busted the myth that electroweak gauge symmetry is broken by the Higgs mechanism. We'll also examine the existence of God and whether true love exists.

naturallyInconsistent

@ACuriousMind I dont know why; I tried to search something I knew happened in hbar before, and its logs got abruptly cut off somewhere. This isnt a singular occurrence by now

ACuriousMind

I don't know what you're doing wrong but you're doing something wrong, the search does not have a cutoff in time

at least I don't think that full transcript search is mod-only functionality

qwerty

@ACuriousMind I just checked and I could search back ten years too

More Anonymous

Bah I cant get access to this paper: journals.aps.org/pra/pdf/10.1103/PhysRevA.110.032602 any bright ideas?

Already tried scihub

qwerty

@PM2Ring will branes do? :)

More Anonymous

15:50

@PM2Ring lol

Ryder Rude

Part of the Lion Pride - Dean Schneider living with lions

►

More Anonymous

@ACuriousMind Do you have any take on Ads CFT?

Ryder Rude

the lions accept him as one of their own

@MoreAnonymous hi !

More Anonymous

@RyderRude hey!

ACuriousMind

@MoreAnonymous As far as I can tell, a perfectly valid (at physical levels of rigor) equivalence between theories; also not something I'm intimately familiar with

qwerty

15:53

@MoreAnonymous isn't it on arXiv?

More Anonymous

@qwerty it is?

qwerty

or do you need the published version

More Anonymous

nah arXivwill do

ACuriousMind

@MoreAnonymous arxiv.org/abs/2402.14412

More Anonymous

thanks

PM 2Ring

15:54

@qwerty Possibly. ;)

qwerty

@MoreAnonymous I'm on my phone so it's hard for me the check the URLs properly but it looked like it

Ryder Rude

@ACuriousMind really interesting @MoreAnonymous

More Anonymous

@RyderRude thanks

Ryder Rude

> The proposed interferometer could test the effect of gravitational redshift on quantum coherence, and implement the quantum twin paradox.

i will have to look up "quantum twin paradox"

it seems to be a hot topic. there is a recent video on it

This Could Test the Quantum Twin Paradox

►

PM 2Ring

@ACuriousMind It isn't, but chat search can be a bit flaky. And a few months ago it was particularly glitchy. IIRC, the indices got messed up, but I can't remember why. Maybe it was related to some Cloudflare stuff, or the migration away from Imgur. But (allegedly) that problem got fixed.

ACuriousMind

15:58

@PM2Ring the problem is always caching :)

PM 2Ring

Yes.

qwerty

is searching the transcript with Google not possible?

ACuriousMind

no, I think they're blocking the robots from indexing it

see chat.stackexchange.com/robots.txt

ah, wait, they allow the transcripts themselves

user430580

A photon walks into a hotel. The desk clerk says, "Welcome to our hotel. Can we help you with your luggage?" The photon says, "No thanks, I'm traveling light."

3

PM 2Ring

in Tavern on the Meta on Meta Stack Exchange Chat, Sep 20 at 13:55, by balpha

there's a delicious irony to the fact that we are getting blocked by cloudflare, too 😬

balpha wrote the SE chat software.

Amit

16:25

@user430580 Does he regret it when he finds out it's Hilbert's hotel?

Mr. Feynman

I have two index (fingers), according to whom am I a rank 2 tensor?

Relativisticcucumber

i am kind of confused about this whole laplacian of $\frac{1}{r}$ being dirac delta. i mean i follow the math presented in books (i.e. this argument math.stackexchange.com/questions/368155/…) but i feel i dont have an appreciation for the fundamental issue. i mean the derivative doesnt care if a function is undefined at one point, no? its a limit. what am i missing here?

Amit

@Mr.Feynman Ricci?? Everyone? Idk, I can't wait to find out

user430580

@Amit Haha you made the that joke even better :D

ACuriousMind

@Relativisticcucumber of course the derivatives cares - if you try to write the difference quotient at zero (of which you then want to take a limit to get the derivative), what do you get?

Mr. Feynman

16:28

@Relativisticcucumber The domain of the derivative is in general a subset of the domain of the function

For differentiable functions (I hate this word), it coincides

Amit

I didn't know that you can like tag a certain time period in this chat as "conversation"

Mr. Feynman

@ACuriousMind (hoping I've never asked here) just wondering, in German is there a word to distinguish a function that has a derivative and function that can be approximated by its differential? Of course it only makes a difference in more than one dimension.

In Italian we say "derivabile" for the former and "differenziabile" for the latter, while in English they say differentiable in both cases :P

Relativisticcucumber

$\text{lim}_{h \rightarrow 0} \frac{\frac{1}{x+h}-\frac{1}{x}}{h}$

Amit

ouch

ACuriousMind

@Mr.Feynman I'm afraid I don't really understand the distinction you're trying to make here - any function that has a derivative can be approximated by it

@Relativisticcucumber yeah, so how does that make sense at $x=0$?

Relativisticcucumber

16:34

ok i think i was muddling two things. 1) functions can have limits that are clean even if the function is weird at that point. 2) a limit can have points at which it is ill defined. these are just two separate things. bleb

Mr. Feynman

@ACuriousMind A 2D function may have partial derivatives but not be approximated by the differential

ACuriousMind

@Mr.Feynman I don't know what "not be approximated by the differential" means. Do you mean a function that is not totally differentiable?

English also has two words: Partially differentiable and totally differentiable

we usually just elide the "partially" or "totally" when it's clear from context

Amit

In fact using "derivable" as "totally differentiable" is not a crazy idea, because only the totally differentiable functions have a (singular) derivative that's special in any way, i think

Mr. Feynman

I'll use equations. In Italian we call "differenziabile" at $x_0$ a function $F:U\subset\mathbb{R}^m\to\mathbb{R}^n$ such that there exists a linear functional $L_{x_0}:\mathbb{R}^m\to\mathbb{R}^n$ (the differential) such that:
$$F(x_0+h)-F(x_0)=L_{x_0}(h)+o(h)$$

ACuriousMind

yes, that's (total) differentiability

Mr. Feynman

16:39

Oh so that's what people call it in English :P

ACuriousMind

if people need to stress that total/partial distinction, they may even call $L$ the total differential

Mr. Feynman

Oh, ok. I didn't know that. The terminology in Italian is much clearer

I mean, the distinction is only relevant when a function is differentiable but not totally differentiable

By the way, I still dislike the English name "total differential". It's not like there is a "partial" differential

Those are derivatives, damn it

PM 2Ring

@Relativisticcucumber BTW, you can write \lim\limits_{h \rightarrow 0} to get $\lim\limits_{h \rightarrow 0}$

Mr. Feynman

Or just use \$\$\$\$

Looks nicer :P

Amit

does the $$ \huge{\text{size}} $$ stuff work here too?

oh, it does. apologies

PM 2Ring

16:49

There's a sandbox chat for testing stuff: chat.stackexchange.com/rooms/1/sandbox

Amit

ah, thanks. this reminds me, i once noticed Qmechanic pastes a lot of random looking equations in comments... lol

PM 2Ring

Of course, you can also test stuff in a dummy answer on the main site. That's handy when you want to see a preview. The ChatJax script should give the same results as the MathJax on the main site, but there might be minor differences for obscure stuff.

Amit

Idk if \define macros can work here

ah sorry i meant \newcommand

ok, so that was my first use of the sandbox, and it does ;)

PM 2Ring

@Amit I think so. It does on the main sites. It used to work too well: it'd define stuff for the whole page. But that was changed a few years ago to limit its effects to the current post.

Amit

@PM2Ring Nice... wow there are some malicious scenarios coming to mind by defining stuff for the whole page, invalidating others answers etc... good they fixed it :)

PM 2Ring

16:58

Yeah. I don't think any malicious uses actually happened. But maybe a few silly ones...

ACuriousMind

if you do it in chat you can break everyone's mathjax until your renewcommand scrolls off the page

I once redefined $\uparrow$ to be $\downarrow$ in the middle of a conversation about spins

Amit

lol

you've collapsed the .... chat...

PM 2Ring

Yeah, it'd be a bit tricky to fix it for ChatJax.

Amit

Why is that? Each message should have its own context similar to a post

ACuriousMind

that's not how ChatJax works

it just renders the MathJax on the page, from the top down

Amit

17:03

@ACuriousMind Yeah which is probably how the main site posts used to be rendered, right?

ACuriousMind

yes

Amit

So where is the difficulty in applying a similar fix, I don't see

Maybe a performance thing

ACuriousMind

ChatJax isn't really specific to chat, nor developed by SE

it's just a script that runs in your browser on certain sites

so in general defining a "context" for each snippet is intractable

PM 2Ring

Incidentally, the core MathJax dev is a member. stackoverflow.com/users/502334/davide-cervone

Amit

Oh right of course, that's why we all activate that script... I forget that

Slereah

17:04

I mean what programmer isn't a member of stack overflow

PM 2Ring

Sure. But it's nice that he's proactive in answering site-related MathJax questions.

Amit

They should pay him to add built in MathJax support for the chat :P which will also solve the macro issue

Slereah

Wait how much does the interpretation of mathjax go over different messages

\begin{itemize}
\item

\end{itemize}

Guess it doesn't work for scoping

Does itemize work in mathjax tho

\begin{itemize}
\item cat
\end{itemize}

it does indeed, if poorly

\iffalse
Test
\fi

PM 2Ring

@Amit Adding MathJax to chat would be very easy. But Stack Exchange (the company) don't want it.

Amit

If presumably something in the script is parsing the text and looking for either $ or $$ , the simplest solution would be to do a total cleanup/init of the macros every time before rendering the contents to LaTeX

@PM2Ring They don't want, or they want not to? :)

PM 2Ring

17:19

They're happy with the status quo. And they've done no dev work on chat for many years.

Slereah

the chat isn't getting them that sweet investor money

Amit

@Slereah To the best of their knowledge...

Claudio

@Mr.Feynman I remember other physicists kind of revered Weinberg as well, but that sentence I quoted is from Glashow, so it can be considered a trusted source. Both contributed a lot to QFT so maybe they can share that status

PM 2Ring

Exactly. They'd rather spend their dev dollars on AI stuff, because it's the latest craze.

Slereah

@Claudio I'm afraid there can only be one

Amit

17:21

Where's the big SE AI project then?

Slereah

to the Thunderdome with them

Claudio

@Slereah Then I'd say we stay with Weinberg

Slereah

@Amit stackoverflow.blog/2023/07/27/announcing-overflowai

Claudio

@PM2Ring dang that's sad :P

PM 2Ring

stackoverflow.co/teams/ai/…

Slereah

17:24

I keep seeing people complain about how the modern internet is being ruined by [all the things], but what are they doing

Just go back to the chats and forums

It's not rocket science

you are doing it to yourselves

make your own geocities website about digimon

As our ancestors did

Amit

They don't wanna admit , they like chatting with AI better

It never offends them, for one thing

lol

ACuriousMind

@Amit we killed the first one and probably prevented a large push to flood sites with LLM-generated content by not accepting their attempt to prevent moderators from suspending users posting generated garbage last year

Amit

@ACuriousMind Right right, now I recall. You kept posting updates on that one

This second attempt looks more benign

ACuriousMind

@Slereah most people on the internet are not really "internet users" in the sense someone molded in the Old Web like you understands it, they just visit the big social media sites :P

Slereah

I Say Let 'em Crash

►

PM 2Ring

17:28

The SE CEO tested the LLM detectors and discovered that they all give false positives: they all think that he sounds like a bot...

Amit

.. which means that even the percentage of people old enough to remember things were different and hence complain is shrinking :P

ACuriousMind

@PM2Ring I mean, the automated LLM detectors are highly fallible

PM 2Ring

They are. But he still sounds like a bot. :D

ACuriousMind

well...yes

the reason all the CEOs and upper management are so hyped about replacing workers with LLMs is because their job could actually be done by one

Slereah

just have to say the words

Generic Corporate Lingo In Every 80's Movie

►

Amit

17:30

@ACuriousMind that's a bit of an exaggeration isn't it?

PM 2Ring

LLMs are great at organising information. Not so great at being genuinely creative.

Amit

a lot of upper management job is also about the dubious area of "human relations"

ACuriousMind

@Amit I feel little reason to be charitable towards people who spew meaningless nonsense about technology they do not understand and base decisions that affect lots of people around it

Yes, it's an exaggeration, but they started it by making absolutely absurd claims about what parts of my job could be replaced by an LLM; if they dish it out they also have to take it in return

Amit

@ACuriousMind Yeah I agree about that. I'm just being nitpicky about the generalization -- some don't understand it and as you say are employing the technique of "the best defense is offense" (if I understood you rightly). While others that don't understand it, may sincerely think it's a good opportunity to cut expanses...

PM 2Ring

@Slereah Behind the scenes, the modern internet is being ruined by a vast plague of bots scraping any site worth scraping, consuming a ridiculous percentage of the bandwidth.

Amit

17:34

@ACuriousMind That's fair

Slereah

@PM2Ring Who cares

just make your own space

It's not terribly hard to keep bots out

You can even make it invite only if you wish to thwart scrapers

Relativisticcucumber

@PM2Ring $\lim\limits_{\text{me} \rightarrow \text{yippee}}$

yay thanks

ACuriousMind

(also I'm not really convinced that many companies would note any adverse effects in the short term if their CEO just vanished and stopped responding to emails)

Slereah

But who will execute

PM 2Ring

@Slereah Sites like the SE network don't really have that option. We get a lot of bots on this network. It's an on-going battle.

Amit

17:37

lol. I think that's actually a legitimate technique suggested by many business advisors to CEOs... improve productivity by taking unexpected vacations, I'm sure I read it somewhere

Slereah

@PM2Ring Part of the problem

What are people doing going to websites with 50 million users

Amit

IIRC, IRC used to be a thing :P

Slereah

It still is

I'm there

Amit

:O

I used to visit EFNet a lot

ACuriousMind

IRC is just a protocol, you just need to find a weirdo still running a server

Amit

17:40

Yes that's the point.. how many servers / people on them actually active

Slereah

Plenty

Amit

But not comparable to the social media monsters

PM 2Ring

Hey, I'm a big fan of old-school bare-bones sites. Eg leapsecond.com & poynton.ca/notes/colour_and_gamma/ColorFAQ.html & machination.eclipse.co.uk

Slereah

@Amit I mean yeah

why would you want a social media monster

Claudio

@Slereah comment section under this one is made up of pure comedy hahaha

Slereah

17:42

Have you seen how those sites are

Me I still salute the anthem

We Didn't Start This Website (YTMND)

►

🫡

Amit

It's interesting to listen to someone like Ted Nelson, he claims to have invented the "back button" on browsers, and he had all kind of ideas for how the web should look like that didn't come into fruition

ACuriousMind

I like The Web That Never Was as an alternate history exercise

Amit

cool, I should watch that

Slereah

Let's make our own internet

PM 2Ring

18:10

With carrier pigeons.

IP over Avian Carriers

In computer networking, IP over Avian Carriers (IPoAC) is a joke proposal to carry Internet Protocol (IP) traffic by birds such as homing pigeons. IP over Avian Carriers was initially described in RFC 1149 issued by the Internet Engineering Task Force, written by David Waitzman, and released on April 1, 1990. It is one of several April Fools' Day Request for Comments. Waitzman described an improvement of his protocol in RFC 2549, IP over Avian Carriers with Quality of Service (1 April 1999). Later, in RFC 6214—released on 1 April 2011, and 13 years after the introduction of IPv6—Brian Carpenter...

Amit

@Slereah There should be a social media site with that name.. 🤣

user430580

@PM2Ring I am lost for words, the fact that not only someone came up with such an idea, but actually implemented it in real life and tested for latency and packet loss? Pure insanity in the most positive and marvelous way, and yet they still say that engineers have no sense of humor!

Slereah

@ACuriousMind If you take a sheaf of local rings but the ring is just the trivial ring {0}, is that just isomorphic to a manifold