The h Bar

02:59

@SirCumference No. That is Gibbs' entropy. Shannon was working purely theoretically about noise in transmission wires and got to the exact same (discrete) formula, and asked von Neumann what to name it. von Neumann replied that since it is the same as Gibbs' entropy and that nobody really knows what entropy fundamentally is, Shannon might as well name his quantity entropy too.

@Obliv What we call Boltzmann entropy and Boltzmann constant, are first written down by Planck. If you want to explore this history, KathyLovesPhysicsAndHistory has it.

Sir Cumference

03:16

@naturallyInconsistent I mean the difference in formulae as far as I recall is a matter of units (Gibbs using natural log and multiplying by $k$)

But theoretically aren't there still connections between them? Usually information theory can be relevant whenever you're dealing with probabilities

naturallyInconsistent

03:31

@SirCumference Correct. I am very sympathetic to von Neumann's suggestion on this. I used to be quite against it, but I'm increasingly convinced that it is correct.

Sanjana

05:30

Today I learnt about an interesting resolution to the information paradox: The information is never lost because it never goes inside --- A copy of the information of what fell inside is always available outside the black hole!

Ryder Rude

08:05

@Sanjana y wud this copy be made

Sanjana

Apparently, quantum gravity does not obey this because there are no local gauge-invariant (diffeo invariant) operators in quantum gravity---so there's no way to separate the Hilbert spaces of regions inside and outside the BH. The states inside the BH are already available in the boundary region of the particular Cauchy slice!

@RyderRude In non-gravitational QFTs and classical gravity data on a part of the Cauchy slice is independent of the data on another part. In other words the Hilbert space is factorizable into Hilbert space of different small regions. E.g. Think Birkhoff theorem---the metric outside is independent of the detailed information inside (apart from a few conserved charges like "mass".) This is called "split" property in QFTs.

@RyderRude So in a word this is because of failure of the split property of ordinary QFTs in quantum gravity.

The author also "proved" that quantum gravity should be holographic without using AdS/CFT or string theory... Any viable theory of QG must be holographic. He kinda derived holography from WdW equation which is roughly the Schrodinger equation of quantum gravity.

Triet Vo Nguyen Minh

08:51

I've read recently in a book that the light waves flip when going through a convergence point. Is that the actual phenomenon that explains why the center of the observed region in Meslin's experiment is dark?

I presume that everybody knows what I'm talking about but just to be sure:
- Meslin's experiment: cut a lens in half and move one half along its principal axis, the intersection of the ray cones is where interference happens.
- by light waves "flipping" I mean a 180° phase change.

Ryder Rude

09:15

@Sanjana so this means that quantum gravity does not have local measurements of local operators?

naturallyInconsistent

@TrietVoNguyenMinh That is a presumption you are definitely correct to doubt. Meslin's name sounds so much like a typo..., and the phenomenon is not widely taught.

Ryder Rude

@Sanjana i think this still assumes that a quantum theory of gravity exists. i dont think there is any quantum theory of gravity that obeys the Von Neumann postulates. so any proof involving the hilbert space and Schrodinger eqn cud not mean much

naturallyInconsistent

@TrietVoNguyenMinh I think that the flipping is correct, however, I think it is better considered a self-interference quite like the double-slit interference, than due to the wave flipping. I am, however, not sure of this.

Triet Vo Nguyen Minh

@naturallyInconsistent From my research, I couldn't find it online... I found it in the form of an exercise in the free chapter on interference on PhysOlymp and an esoteric French (with a stripped down Vietnamese translation) book, the book I was talking about above.

naturallyInconsistent

@TrietVoNguyenMinh Even much more reason to think that other people would NOT know what you are talking about, as opposed to it being common knowledge that you can presume that other people would know...

Triet Vo Nguyen Minh

09:29

The stripped-down version only said to admit that there is a phase change of 180° when it goes through a convergence point, and that can be derived from equations of electromagnetism

And this is the page: https://physolymp.com/chapters/interference
The setup is at problem 4.

naturallyInconsistent

09:42

@TrietVoNguyenMinh Ah, yes, that phrasing makes it exceedingly clear that to correctly account for the dark centre, you must have the phase change, so yes, you absolutely require that the flipping to explain that.

Supersymmetry

Why does a spacetime-dependent parameter e(x) appear in local symmetry transformations of a field \psi?

Mr. Feynman

@Supersymmetry Mh? That's the meaning of local

Supersymmetry

@Mr.Feynman Yes, I am aware that we want the symmetry transformations to be dependent on our position in spacetime. However, if we define the variation as a smooth family of sections over spacetime bundle $E$ that satisfies $\hat{psi}(x, 0) = \psi(x)$, then for every $s\in I\subset \mathbb{R}$, we get a smoothly deformed field $\hat{\psi}(x, s)$. How exactly does $\epsilon(x)$ turn this to a position-dependent deformation?

Slereah

Why do you think the smoothly deformed field is $\psi(x,s)$

Automorphisms on a bundle are like uuuuh

$\psi(x, f(s,x))$

Or something

if you have your section $x \mapsto (x, e)$, the automorphisms are $x \mapsto (x,f(x, e))$

If only the value of the field changes by that transformation, that's a vertical automorphism (or "gauge")

The one you have is specifically not vertical

Your smooth deformation is just gonna be $x \mapsto (x, f_s(x,e))$

For $f_s \in \mathrm{Aut}(E)$

Sanjana

@RyderRude In pure quantum gravity, diffeo. invariance is a gauge invariance. But "local" quantities depend on coordinates. So no local gauge invariant operator exists. There are slight deformations of these "naive" arguments but I don't know much about them

@RyderRude Hmm

Ryder Rude

09:59

@Sanjana diffeo invariance is gauge invariance in classical GR too but u hav local measurements there

i think it makes sense. when the metric isnt classical, there is no meaning of "locality" becuz spacetime doesnt exist @Sanjana

Supersymmetry

@Slereah I see, thank you

Ryder Rude

local measurements can only b defined aftr constructing frames of references using the metric

Sanjana

@RyderRude Yes, even in classical GR there are no non-trivial local diffeo inv. "observables", which is subtle to define as mentioned in this answer

Ryder Rude

10:14

@Sanjana i think an observable is essentially a co-ordinate dependent quantity, as in, u hav to use the observer's frames of reference to define their observables. for e.g. energy is a co ordinate dependent quantity and is still an observable

so we shud not look for diffo inv observables maybe

or we can use another methodology which makes energy a scalar. for e.g. to define the "energy relative to an observer", u can define it as the time component of the energy momentum four-vector in that fixed basis. this energy wud b a co ordinate invariant scalar @Sanjana

Sanjana

@RyderRude Yeah, I heard of this...don't remember where.

Ryder Rude

oh

Sanjana

@RyderRude Maybe...but then we will have to see why that definition is useful. Btw the above stuff about info paradox I was talking about...it is based on Suvrat Raju's review on info paradox

arxiv.org/abs/2012.05770

Ryder Rude

@Sanjana thankss

Supersymmetry

10:52

@Slereah This might be a silly question, do we want the second factor (s) in $\hat{\psi}(x, s)$ to also depend on on position, i.e $\hat{\psi}(x, f(s,x))$ so that we associate different smooth sections over E depending on x? If f satisfies f(x, 0) = 0 for every x\in M at s = 0, then $\hat{\psi}(x, f(0,x)) = \hat{\psi}(x, 0) = \psi(x)$.

Slereah

I mean the actual formalism would be more something like $\psi_s(x) = f_s(x, \psi(x))$

Although even that is way too broad because basically no bundle has that kind of automorphisms

Most of them are given by the appropriate representations so $f_s(x, \psi(x)) = \Lambda_s(x) \psi(x)$

Slereah

13:23

Love looking in CIA files for physics stuff

Jim

@Slereah what cia person dropping what acid came up with the idea that a) flux lines are neutrinos and b) higher order (derivatives?) of flux lines are minds and thoughts??? For an intelligence agency, they seem to be lacking in it

Slereah

@Jim From what I can tell from various cases, having a crazy person in the agency wasn't usually ground for dismissal

ACuriousMind

I mean this is the agency that did MKULTRA

Slereah

I'm guessing those documents never amounted to much actions

Jim

a tulpod is an alien saucer?

Slereah

13:31

@ACuriousMind Reading MK Ultra you will learn a lot of interesting details

Mostly that a lot of MK Ultra documents are essentially saying "Send money to [the same company] for further research of this topic" over and over

Jim

some parts of these seem to just be "pick a word from a hat" like "Drives thrusting down"

Slereah

I am guessing that the sinister conspiracy involved is pretty pedestrian

Charles Freeman Geschickter

Charles Freeman Geschickter (8 January 1901 – 1987) was an American pathologist who made important contributions to the understanding of breast cancer and other diseases of the breast. In his later years, of which little is recorded, he apparently undertook work for the Central Intelligence Agency (CIA). == Birth and education == Charles Freeman Geschickter was born on 8 January 1901 in Washington, D.C. His father was an inventor and was involved in the fur trade and in cabinet-making. Geschickter took an undergraduate degree in engineering, then studied Educational Psychology, earning MA and...

^a lot of money sent to that guy

Jim

That first figure must be where the idea for the Geoclense came from

Slereah

Also looking into some details you can tell that the average CIA man isn't a genius rly

like you may have heard about those CIA research into psychic powers, which you would probably imagine to be some laboratory tests

But some cases I've read they were just shown those powers in some hotel room

Getting swindled by wizards like a bunch of rubes

ACuriousMind

13:47

@Sanjana I would be cautious with any arguments based on the "no local observables" argument, because the significance of "gauge-invariant" here is unclear: The "gauge symmetry" they are talking about is the diffeomorphism symmetry of GR, but we usually in GR don't consider objects that are variant under this to be irrelevant in the same sense in which gauge-variant functions in YM are unobservable

it's correct and completely obvious that there are no non-trivial functions on a manifold that are invariant under arbitrary diffeomorphism

yet for some reason in classical GR no one cares about this and we compute a bunch of observable (in the practical, not formal, sense) stuff anyway

Slereah

Actual observables in GR are pretty much nonlocal anyway

ie. measuring a length

measuring a volume

etc etc

you can't really measure pointwise the metric tensor

ACuriousMind

I mean you could say that for any physical theory

Slereah

All the more reason to not worry :p

ACuriousMind

my point is that the "gauge symmetry" in GR is usually not treated like the internal gauge symmetries of other theories: We don't really claim that only "invariant" stuff can be observed, right? If you add EM to GR, no one claims you can't measure the electric field because any non-zero vector field is not invariant under diffeomorphisms

Slereah

They are typically measured wrt some observer tho in this case, no?

idk

ACuriousMind

13:54

sure

Slereah

Diffeomorphism invariance is broken by the apparatus

ACuriousMind

interesting idea, but that would still mean the "no local observables" argument is pointless and what people should really be trying to quantize is the theory with the apparatus

Slereah

I mean I think it's worth investigating certainly, but I'm not losing any sleep about its applicability

taps sign

ACuriousMind

yeah, I think we first need to figure that one out in classical GR before we start trusting any arguments about how quantum gravity "has to be"

Slereah

I'm pretty sure the first working theory of quantum gravity we will have will not be some perfectly abstract theory anyway, it's just gonna be some index jamboree

The experimentalists aren't gonna be worrying about the diffeomorphism invariance

ACuriousMind

13:59

are the experimentalists worrying about QG at all? :P

Slereah

I mean there's some experiments for it

Actually we can check what they're actually doing

Modern searches for Lorentz violation

Modern searches for Lorentz violation are scientific studies that look for deviations from Lorentz invariance or symmetry, a set of fundamental frameworks that underpin modern science and fundamental physics in particular. These studies try to determine whether violations or exceptions might exist for well-known physical laws such as special relativity and CPT symmetry, as predicted by some variations of quantum gravity, string theory, and some alternatives to general relativity. Lorentz violations concern the fundamental predictions of special relativity, such as the principle of relativity,...

arxiv.org/abs/1303.6709

there's no diffeomorphisms in there

Mr. Feynman

@ACuriousMind yes, a guy in my uni was ordered to remove all equations from his QG thesis presentation because there were experimentalists in the committee :P

Slereah

@Mr.Feynman Did they ask him to use handpuppets to explain some of the bigger words too

ACuriousMind

@Mr.Feynman what

you sure this was because of the experimentalists and not because he did the usual theorist thing where they put 50 equations on 10 slides and think everyone will just be able to follow along? :P

Slereah

@ACuriousMind Been there

It's tough to make a powerpoint when you've been doing all theory

Mr. Feynman

14:08

@ACuriousMind the truth is somewhere in-between but still all equations were removed :P

@Slereah I guess the mood was kinda like that

To be clear, it's not like the experimentalists asked for that. It's that the thesis advisor ordered that to prevent any reaction :P

Slereah

Family Guy - Peter Griffin Scared of Puppets

►

Philosophy of Science 14 - Novel Prediction

►

worth a watch btw for the people who think string theory replicating existing experimental data isn't good enough :p

ACuriousMind

14:29

@Slereah last time I checked, it still can't really replicate existing data because the best we can get is some supersymmetric version of the SM

I mean that's basically why we're explicitly looking for traces of superpartners in our colliders

Slereah

@ACuriousMind It can still replicate GR

nothing to sneeze at!

ACuriousMind

indeed

Slereah

but the argument isn't typically specifically about phenomenological string theory not perfectly replicating the standard model

It's usually more that it's not a good enough criterion

ACuriousMind

unfortunately it always tacks that SU onto the GRA :P

Slereah

Which is a weird idea because it basically posits that string theory is less good than QFT because it was developped after it

ACuriousMind

14:36

that's right but then again ST and QFT are not disjoint, you arguably cannot develop ST without QFT

at least in the standard approaches

Slereah

Still a bit of a weird argument imo

It's not like string theory was developped with the express intent of aping those experimental results

ACuriousMind

oh, I'm not exactly defending the objection here

Slereah

It's not this :

Dec 20, 2018 at 14:57, by Slereah

" Let me provide you with an unnatural theory that I will call Leo Vuyk 2020 theory. He hasn't invented this precise theory yet but he's free to plagiarize me. The theory says that the world is a giant strawberry whose vital characteristic, the Vyukness, is a real number, a parameter known as 𝐿𝑉. It defines the ratio of the diameters of the strawberry and its pit. Strawberries usually don't have pits but Leo Vyuk's strawberry has one.

Now, the parameter 𝐿𝑉 contains the answer to all questions in the world. It's approximately equal to 42.05121973…. Fourty-two was explained in the Hitchh…

Slereah

14:49

I did good to post that here because I think it disappeared from Motl's blog

Slereah

15:13

web.archive.org/web/20170718095210/http://motls.blogspot.com/…

Strong opinions on shape dynamics

> Their formalism is ugly, too. All the first-class and especially second-class constraints are annoying, unnecessary, and the need to use them usually indicates that something is seriously wrong with the theory.

Ryder Rude

16:02

note that the proper terminology shud be : "observables in GR are frame dependent", and not "observables in GR r co ordinate dependent"

6 hours ago, by Ryder Rude

or we can use another methodology which makes energy a scalar. for e.g. to define the "energy relative to an observer", u can define it as the time component of the energy momentum four-vector in that fixed basis. this energy wud b a co ordinate invariant scalar @Sanjana

here i have given the methodology to define co ordinate invariant observables while keeping frame dependence

naturallyInconsistent

16:42

@Slereah errr, just to not be totally mistaken, I think you are being sarcastic here...

Slereah

Am I

naturallyInconsistent

@Slereah I think this is making a huge mistake: When we make a new theory and try to accommodate some stuff, the usual situation is that the new theory has parameters to tune to precisely replicate the accommodated stuff. This is different from the situation whereby the theory that has accommodated stuff is then used to explain data that isn't used in the accommodation step. In a sense, what I am arguing here is that a theory gains some value if, like AI models, it can reliably "predict"

what isn't in its training data set.

Slereah

It isn't that common for physical theories to actually be "fitted" to data

String theory was most certainly not fitted for that, it was discovered out of entirely unrelated research

naturallyInconsistent

It is in this sense that, say, quantum theory swiftly gains ground: It is the only logically internally consistent theory at the time that, given mass of electron, reduced Planck's constant, Boltzmann's constant, mass of nuclei, fine structure constant, and so forth, and then goes to explain every physical phenomena that was confusing scientists prior to quantum theory.

@Slereah I was asking you, not asserting that you are. Are you?

Slereah

I was being serious

Predictivism is way overrated imo

naturallyInconsistent

16:51

Well, then I would have to point out that the video also misses another rather important point: We change theories when there are sufficient value-add from the old theory to the new theory. I'm not saying that we won't at all entertain proto-theories. I am saying that string theory has to give us something more than the Standard Model under QFT. At this point, we don't even know which of the string proto-theories are the ones that would successfully reproduce known physics and have a good chance

Slereah

Who's to say you have to change theories

You can just keep several competing theories

naturallyInconsistent

In particular, the added complexity cost of string proto-theories have to be less than the value-add that it brings.

Slereah

The concept of complexity of a theory is also pretty vague :p

naturallyInconsistent

@Slereah Sure, this is at least somewhat tolerable. However, you then cannot blame people for not paying that much attention to string proto-theory as long as it does not do any better than that.

Slereah

Oh I don't care for attention

But plenty of people seem to have strong opinions about string theory!

Quite the opposite of no attention

Physical theories where you actually manually fit the theory to data tend to basically be the ad hoc formula like the liquid drop model or the perfect fluid equation before thermodynamics

naturallyInconsistent

16:58

Anyway, I do not mean to give the impression that I didn't learn and/or enjoy the video, but it isn't convincing

Slereah

It's not very common to do it for like fundamental theories

naturallyInconsistent

Yes, those of us who have thought about that (and better: actually just straight up taught about that) know this.

Claudio Menchinelli

17:14

Guys how does he get a triple scalar product even though there’s no sum in the $l$ index

unless the book implicitly switches to Einstein convention

Slereah

it is implicit Einstein yes

Claudio Menchinelli

Ok thanks

Sanjana

17:47

@ACuriousMind Yeah, sure. I was being very quick while sayin' that. The actual article has about two pages pointing out the various subtleties.

Supersymmetry

18:14

@naturallyInconsistent What do you mean by the statement that "string theory has to give us something more than Standard Model"? For string theory to be consistent, you need to impose supersymmetry which predicts the existence of partner particles. This isn't something Standard Model gives. Mirror symmetry relates symplectic and complex structure, in particular type II string theory on Calabi-Yau folds has far-reaching physical implications

Nihar Karve

18:33

> has far-reaching physical implications

(citation needed)

ACuriousMind

@Sanjana which article?

Slereah

21:18

"Immediately after Einstein’s first suggestion, the matter was taken up by Dr. E. Freundlich, who attempted to collect information from eclipse plates already taken; but he did not secure sufficient material."

"At ensuing eclipses plans were made by various observers for testing the effect, but they failed through cloud or other causes"

Truly cursed

Slereah

21:40

Reading Eddington's report you can get a good feel of the toughness of experimental physics

Relativisticcucumber

22:37

HONK

Transcript for