The h Bar

think_meaning_buildß

00:26

::crickets::

naturallyInconsistent

00:49

@User1865345 happy mew year~~

Allie

01:34

meow

meow?????

peterh

02:03

Afaik MOND thinks that gravity on large distances switches to 1/r from 1/r^2.

I think that is not logic. Why it does not suspect a fifth interaction which decreases by 1/r?

Allie

02:36

@naturallyInconsistent any good texts you know of for DFT? one that assumes you know the fundamentals of quantum chemistry/HF

one text that i was told about by a grad student i work with is "Density-Functional Theory of Atoms and Molecules" by Parr and Yang but idk if its a good text

Silly Goose

03:04

Is there a simple condition for when the time-evolution of a quantum state is periodic?

ACuriousMind

@SillyGoose Hint: Expand in energy eigenstates.

Silly Goose

Why do we want the canonical and microcanonical ensemble to produce the same thermodynamic results? To me, canonical and microcanonical ensembles describe quite distinct scenarios.

@ACuriousMind okay i shall think on it thank you

Silly Goose

03:34

So I believe $\exists t \neq 0 \text{ s.t. } \Psi(t) = \Psi$ if and only if there exists a set of integers $m_n$ (not all zero) such that $2\pi m_n/\lambda_n = t$ for all $n$.

thus, time-evolution is generically not periodic since the initial state does not even recur once for a Hamiltonian whose eigenvalues do not satisfy the above constraint.

huh. for some reason i had in my mind that closed evolution is periodic...then i was reading stat. mech. and thought surely the ergodic hypothesis does not apply for a closed quantum system...but then here we are

Allie

05:55

Hi John

John Rennie

Hi :-)

Allie

Hope u are well this evening

John Rennie

Morning in the UK! It's 05:56 a.m. here.

I'm good. The first coffee of the day has worked its usual magic :-)

Allie

yay

yall: would it be correct to say that Hartree-Fock treats the electrons as if they are in a potential that is an "average/effective" repulsion from the other orbitals, while KS-DFT treats them as if they are in a potential that comes from a functional of the density?

where of course the kohn-sham potential is trying its best to replicate the true potential in some sense

John Rennie

That's certainly true for HF. I don't know enough about DFT to comment. I don't think DFT had been invented when I was doing HF-SCF calculations.

Allie

05:58

yay

at least that one is right

youre an early riser

its 1 am here

John Rennie

As I got old and grey I found I was waking up earlier. I believe this is fairly standard for humans. I quite like it - the world is a very peaceful place at 6 a.m. :-)

Allie

my biological clock seems to always tend towards sleeping later

John Rennie

When I was a student getting up before 9 a.m. was a major struggle. Unfortunate really since the university had arranged lectures at 9 a.m. six days a week (yes, including Saturdays!).

I'm sure I have read that the human circadian rhythm is longer than 24 hours when you're young and shortens to less than 24 hours as you age.

That would certainly tally with my experience.

Allie

i guess im slightly young

lol

John Rennie

I don't know what age the crossover is i.e. what age your circadian rhythm is exactly 24 hours. In the 40s I'd guess.

Allie

06:05

yeah im like half that

lol

John Rennie

Are you studying for a degree at the moment?

Allie

yurp

John Rennie

In physics?

Allie

i am in the last semester of my BS, i am currently waiting on decisions for phd programs right now

nope, chemistry

computational chemistry tho so a LOT of physics lol

and i am very nervous about the decisions but im trying to just not think about it

John Rennie

Aha, so you've been studying HF-SCF for doing molecular orbital calculations?

Allie

06:10

yurp. i am currently doing research in DFT

i am most interested in new methods and software, since i am a programmer as well

John Rennie

The university I went to didn't separate chemistry and physics - you just took whatever science courses you wanted, and I specialised in quantum chemistry in my final year.

Allie

in the next week or so when i finish reading over HF theory ill write my first quantum chem program from scratch, for HF-SCF

@JohnRennie i go to a small liberal arts college--one of five chemistry majors

maybe four actually not sure

John Rennie

Writing HF-SCF code from scratch seems like a lot of work. There must be a ton of code already out there. I did a project modifying code to study the electron densities of molecules with a view to distributed multipole analysis.

Allie

well the point is to force myself to understand every little detail about the process :3

John Rennie

But the HF-SCF code already existed so I just had to add the extra code needed.

Allie

06:14

i dont expect anyone to ever use it even myself

i just thought it would be a nice little project to make everything clear and stick with me

ill be honest it doesnt seem like that much work to me. but i tend to fully devote myself to my programming projects and im used to working on very large projects

of course ill only be sure of that once i actually do that, there will definitely be snags ;)

and in terms of all the linear algebra ill just use existing implementations in python

would probably be much more tedious in something like fortran or C

John Rennie

It's always interesting to work through these things for yourself, though it can take longer than you think!

Allie

definitely

qwerty

@JohnRennie it always takes longer :')

Allie

but i am an ambitious person by nature. i have to at least try

John Rennie

The code I used was written in Fortran. C had been invented by then but it wasn't widely used in scientific research.

Allie

06:18

yeah i really dont know fortran. i only had a brief encounter with it while using DFT to study catalysts during my summer research program

it seems like a funny little language lol

i like C though. but not if youre trying to make regular every day stuff

John Rennie

HF-SCF is not as accurate as one might wish. The problem is that in reactions we are usually dealing with small changes in large energies so even if the error in the HF-SCF calculation is quite small it could still result in large errors in interaction energies of molecules.

Allie

ofc

John Rennie

As I recall we had to use an additional configuration interaction step to get more accurate results.

Allie

yurp. that would be my next step! (if i have time)

implement CI

of course that gets ridiculously computationally expensive very fast, but it would all be a learning exercise

John Rennie

Yes, CI isn't hard but you get very lengthy calculations.

Allie

06:23

CI is the next chapter so i only have been briefly introduced to it--the difficulty is constructing the 2 electron integrals of the molecular orbitals from the 2 electron integrals of the basis set, right?

John Rennie

I can no longer remember the details - I went on to work in colloid science as an industrial scientist and I haven't done a HF-SCF calculation for four decades.

Allie

LOL

qwerty

@Allie may I ask did you also do a comp sci degree or were you self taught?

Allie

self taught for about 10 years :3

Hi user!!!!!!

User1865345

Hey.

Allie

06:30

you guys saying its difficult has motivated me to start tonight....

i think i have everything i need to get started

User1865345

@Allie great. Always appreciate self taught people.

Allie

i dont learn well in a class environment. i have always excelled if my learning is self-directed

User1865345

@Allie you doing Hartree Fock (hope I spelt that correctly, lol)?

qwerty

I think in groups where I've worked there's a saying that goes actual project timeline often is like 1.5-2x your estimate

Allie

yup! HF-SCF to start

User1865345

06:32

@Allie Most of the things I learnt are self studying books that I love

Allie

@qwerty Eh I don't really have an estimate. I just wanna start :) these projects are always rewarding and enjoyable and i just get lost in them

User1865345

my hands shivering in cold to type...

qwerty

I also did hartree fock for like 3months in an undergrad project but I've forgotten it all now

like completely

User1865345

@Allie 👏🏻👏🏻

Allie

the project has been created.

qwerty

06:33

I think I was working in Fortran too

nice

User1865345

@JohnRennie 😱👏🏻

I wish I could wake up early especially in this frigid 🥶...

@qwerty you folks use Fortran as primary language?

naturallyInconsistent

@Allie arrived at office and immediately got roped into day-long meetings

very crycry

Allie

mew mew mew

qwerty

@User1865345 no. it was just the group I was working with then

User1865345

Ahh. Okay.

qwerty

06:36

@User1865345 where are you based? :)

naturallyInconsistent

@Allie miao miao used a few texts, Vignale seems to be nice though miao miao did not read it.

User1865345

@qwerty I prefer not to say but at this point, I think anyone could guess.

naturallyInconsistent

@Allie On the contrary. Doing it in python will be a lot more involved than in Fortran; Fortran is literally geared for these manipulations, and plenty of free code exists that helps you with it. C, however, is a mess.

User1865345

Back in undergrad days, I got option to learn a language. Fortran was one option. But I didn't opt for that.

It has hardly any applicability in statistics. So I opted for python.

John Rennie

@User1865345 The downside is you have to be old - on balance I don't recommend it :-)

2

Allie

06:40

@naturallyInconsistent 2 things: free code that performs what functions exactly?

User1865345

😅

Allie

and, will python really be more involved if doing it in fortran would require me to.... learn fortran

naturallyInconsistent

@Allie whatever it is you want: FFTW has Fortran and C bindings, BLAS and LAPACK and so forth are best in Fortran, etc.

User1865345

You do programs in what language @Allie? Python?

Allie

are "best"?

naturallyInconsistent

06:41

A lot of matrix stuff are done in Fortran, and most languages simply import from Fortran.

Allie

exactly

naturallyInconsistent

@Allie modern fortran behaves similarly to python...

Allie

so id rather use python and have someone else handle that complicated matrix math :D

@naturallyInconsistent ok i am not, get it outta your head

naturallyInconsistent

lol

Allie

@User1865345 I've done web dev (HTML, CSS, JS), my main experience tho is low level programming. assembly and C for operating system development, GPU programming, etc

User1865345

06:42

@naturallyInconsistent this is true. R language, that we use, is based on Fortran partly.

naturallyInconsistent

miahahahaha

User1865345

Bulk is C though. Wehen I made packages, I did that in R-C interface.

Allie

but i genuinely dont really understand what youre saying. how it would be much more involved in python

User1865345

@Allie 👏🏻👏🏻

Allie

i plan on doing everything from scratch besides matrix manipulations so i just dont see how it would be that different

User1865345

06:45

@Allie you will be making your own libraries?

Allie

libraries for what?

here, im just gonna get started and then we will see if i run into any trouble ;)

User1865345

I mean you said you would do everything from scratch. I thought you would be writing packages and libraries for your computations.

naturallyInconsistent

@Allie I think your way of phrasing it does not emphasise the essential differences (nor similarities). In HF, you insert the electrons "in pairs" (subshells) upwards from the lowest energy, with each pair seeing a mean-field from the other electrons. (Of course, in practice, this is done one-step with all the electrons coming in together.) This way, each electron sees a unique mean field. There is also a Slater determinant step, or else it is not Fock.

User1865345

@Allie sure.

Allie

right the Slater determinant is where the fock operator comes from

naturallyInconsistent

06:48

Whereas in DFT, the picture is actually the same. However, it is noted that the exchange interaction from Slater determinant, the Fock part, is way too strong, and instead would be weakened by the correlation. Hence, you bundle the exchange and correlation into a functional that you approximate, and thus omit the Slater determinant. This saves you a tonne of computation and yet gives you a better result. But that means that you are approximating the exchange interaction, even in

Exact Exchange. Yes, you can verify this, even on Wiki, IIRC.

The point is that, since the exchange interaction is too strong, i.e. overestimating, you might as well omit the step and just go straight to approximating both exchange and correlation together.

Allie

too strong in HF you mean

naturallyInconsistent

Yes

So in DFT you just omit it

Allie

lovely

meow

naturallyInconsistent

@JohnRennie It certainly had. DFT started becoming popular in the 1970s.

@Allie M E O W ~

John Rennie

@naturallyInconsistent In 1982 all the code that was floating around at my university was HF-SCF. But bear in mind that the world of scientific computing was much more fragmented before the Internet was invented.

naturallyInconsistent

06:59

@JohnRennie that is very much not denied. Of course it would be, and it shouldn't be everywhere until the applications became routine. Early efforts werent great for anything

@SillyGoose You already noted that they describe different scenarios. You definitely know that when you have few particles only, i.e. far from thermodynamic limit, the behaviour disagree quite a lot. If they don't produce the same thermodynamic results, then why are you even interested in them at all? Can't you see that they would have no place in the scientific curriculum if they don't satisfy some important and nice properties?

@SillyGoose Your intuition is completely out of whack. If it is periodic, then it is extremely far from ergodic (except in the subspace where it can be operating in.)

naturallyInconsistent

07:16

@Allie this is correct. If it were $X^{-1}FX$ then the result is necessarily unitary. Because it is orthonormal, the adjoint is sufficient to obtain the transformation, yet you dont want the resultant to be unitary at the end, so that is how it is.

Allie

yay thx <3

naturallyInconsistent

@think_meaning_buildß If you use rocks, then you are still establishing explicit isomorphism, just a different kind of isomorphism. The silly story is about inventing numbers as a way to break this need for isomorphisms.

think_meaning_buildß

Hmmm, I thought it had to do with inventing silly words...

Allie

ugh i forgot i broke my glasses

life kinda sucks

naturallyInconsistent

@Relativisticcucumber The wacky fowl is correct. This is not any $T$, it is $T^0$. Both $S$ and $T^0$ have zero magnetic quantum number and thus will not be affected by magnetic fields.

User1865345

07:23

@Allie 😱

Allie

im sad now

bed time

think_meaning_buildß

cya

User1865345

@Allie 🥲

naturallyInconsistent

@think_meaning_buildß sure, as long as you dont think of it as inventing silly walks

User1865345

Bye @Allie.

naturallyInconsistent

07:24

@Allie oh nose!

good night

Allie

gn $\hbar$

qwerty

cya

think_meaning_buildß

@naturallyInconsistent walk like an Egyptian?

naturallyInconsistent

@think_meaning_buildß is that a silly walk at all? Do they walk in a different way at all? There was a historical story of a walk that is definitely different, but in ancient C culture that is now lost.

But it survives only as a phrase.

A cautionary tale

John Rennie

Missed cultural reference error?

For those of us even older than the Bangles a silly walk reference means the Monty Python sketch.

think_meaning_buildß

07:33

whatever happened to the flying circus?

naturallyInconsistent

Death visited some of them, sadly.

think_meaning_buildß

RIP

naturallyInconsistent

news.ycombinator.com/item?id=23175216

qwerty

07:55

just out of curiosity, do any of you know much about aerodynamics/ how planes fly? obligatory xkcd xkcd.com/803

User1865345

I thought I saw somewhere, maybe a video, that this is oversimplified.

How do airplanes actually fly? - Raymond Adkins

►

Can't remember anything though what I saw. 😬

naturallyInconsistent

08:17

@qwerty The standard treatments are SOOOOOO BADDDDD. Some of them even assert that there is a disagreement between N2L view and Bernoulli view of the problem, which is just straight up nonsense. The picture is complicated, but there is no contradiction with general laws of physics that necessarily has to apply. Complementary pictures work together to explain what we get.

qwerty

@naturallyInconsistent where did you pick up what you know?

naturallyInconsistent

@qwerty serendipity~~

User1865345

😺

qwerty

lol xD

Tobias Fünke

@Allie Parr and Yang is nice. Dreizler and Engel is also nice. And Dreizler and Gross, too (it also includes other many-body techniques)--I know you did not ask me, but I could not help it :d

that being said, none of these I fully read through, but only consulted chapter by chapter if needed

morning everyone :)

qwerty

08:34

hey :)

naturallyInconsistent

@TobiasFünke mia~~

Tobias Fünke

08:49

have you had a good start into the new year so far?

nI you are working already again already? :d

naturallyInconsistent

@TobiasFünke crycry

non-stop meetings!!!! brain melts

qwerty

pat pat

naturallyInconsistent

mew mew

Last two days were great, though

Tobias Fünke

09:09

:(

qwerty

09:50

is the vixra founder a crackpot or did the site just get taken over by them?

I found a dubious looking answer of his on phys.SE which I don't trust (and contradicted the top answer) but nonetheless had a number of upvotes

John Rennie

Phil Gibbs is a well respected physicist.

In fact he's very good at explaining complicated features of GR in a simple way and I like his writing a lot.

He founded Vixra with good intentions, and I'm sure he's disappointed with what it has become, though really it should have been obvious to him what would happen.

Anyhow, you can trust any answer he posts here.

qwerty

hmmm

John Rennie

What's the answer you're unsure about?

naturallyInconsistent

@JohnRennie I'm not sure that he is disappointed, but it tends to be the case that the people who do not agree with his philosophy, would have the takeaway that the gatekeeping rules of arXiv is necessary

John Rennie

I haven't asked him, but Vixra has become a cesspool of crackpots.

There are some good posts on there, but if you reach in at random you're far more likely to pull out a turd than a gem.

naturallyInconsistent

10:03

His point is precisely that one never knows which crackpot might be onto something and so everything needs to be documented and made free for everyone to peruse.

qwerty

@JohnRennie there's a couple of them. I haven't thought deeply about it but just on first read they sounded fishy.

12

A: Energy conservation in General Relativity

Energy conservation does work perfectly in general relativity. The overall Lagrangian is invariant under time translations and Noether's Theorem can be used to derive a non-trivial and exact conserved current for energy. The only thing that makes general relativity a little different from electro...

14

A: Is the total energy of the universe zero?

The claim that the total energy in the universe is zero can be rigorously justified. To answer your specific questions: General Relativity is required. It does not apply for Newtonian gravity. It has to be assumed that classical general relativity, with or without cosmological constant, is corr...

plus linking to your own paper and nothing else

is a bit of a red flag imo

handan_toddler

@naturallyInconsistent nice story, thanks 🙏

naturallyInconsistent

@handan_toddler lol you

John Rennie

@qwerty I read that article, and the point Gibbs is making is that it depends on how you define the energy of the universe, and that's quite correct.

He points that that if you include an extra term then the total energy is conserved.

naturallyInconsistent

@qwerty both answers are fine

qwerty

10:15

in that case, it's interesting that this doesn't seem like textbook stuff

actually motl's take is what I remember from books

Madder

You often find in books examples of unit cells, and sometimes they give squiggly lined unit cells to show none uniqueness. EG: Aschcroft. But the problem i see is, the primitive unit vectors are not able to produce such shapes. So how can they be considered primitive unit cells?

Madder

10:42

Or is it such, that the unit vectors, span one possible primitive unit cell among many primitive unit cells? and not every primitive unit cell has vectors spanning it (as triple product)

naturallyInconsistent

11:12

@Madder Why is it necessary for the unit vectors to make up the shapes of the unit cells? Every point on the boundary of the unit cell can be translated by unit vectors to find another corresponding point on the boundaries, and that is all that matters.

@qwerty ??? They are pretty much equally good? Just emphasising different aspects

qwerty

> Despite these general formulations of energy conservation in general relativity there are some cosmologists who still take the view that energy conservation is only approximate or that it only works in special cases or that it reduces to a trivial identity. In each case these claims can be refuted

that's what Gibbs said, and he is right that that is what I was taught

although not "it is approximate"

but special cases or trivial identity yes.

naturallyInconsistent

Because in cosmological contexts the importance of the dark energy is far more emphasised

qwerty

and this is what Motl was saying

naturallyInconsistent

I also think that it is more surprising, and thus more informative and entropic, to state that, hey, in cosmological context, energy conservation is not something you can assume

whereas in other areas of physics, we tend to want to emphasise that energy is conserved, because it is such a great organiser of the current understanding of physics

qwerty

sorry, I don't quite follow. cosmology is only an application of GR; whatever goes for GR should go just fine, DE or not...

naturallyInconsistent

11:20

Well, that almost completely depends upon whether you consider cosmological constant as part of GR or not

qwerty

I do!

naturallyInconsistent

And that is so big an assumption that you should not be taking that at face value

I mean, most physicists who are not in cosmology would probably make the opposite assumption

Madder

why it is necessary? because i thought that was the definition. I am guessing it is not.
i do not understand the boundary thing you mentioned.

qwerty

@naturallyInconsistent ehh maybe it's bias but how many general relativists are there whose work does not interface with cosmology

naturallyInconsistent

@Madder Even when curvy, each point on the boundary will have its corresponding other points. This is just the same as in waves, where you define "constant phase difference" even if the whole thing is a curve.

@qwerty I dont know why you would restrict yourself to "general relativists"; quite a lot more physicists outside of the particular field would have some training in GR, and they understand the issues being discussed. The same maths in GR also help to organise the concepts in condensed matter, for example. It is also not clear why a GR specialist trying to work on quantum gravity, would necessarily have to interface with cosmology. Or MOND ppl, say.

qwerty

11:25

@naturallyInconsistent in any case, Gibb's answer included mention of the cosmological constant, so it can't just be that...

naturallyInconsistent

@qwerty still, miao miao hath no idea what your issue is with that answer. It is fine as far as miao miao can see

qwerty

just that surprisingly, as he himself says it refutes claims made by some cosmologists. and that includes people/books I learnt from as well as the answer given by Motl

qwerty

12:00

@naturallyInconsistent actually I'm confused by this statement still. it's accepted to add the vacuum contribution to the RHS of the EFE, i.e. the stress-energy tensor these days. it works either way.

naturallyInconsistent

@qwerty but that difference is basically the only thing breaking the conservation of energy at the cosmological scale. While none of us are claiming that it is a controversial add, it is relevant to the discussion at hand, that is why it is important to bring it up

qwerty

anytime you have a spacetime without a timelike killing vector you would have this argument that energy isnt conserved (or at best you would say the stress energy tensor is covariantly conserved)

naturallyInconsistent

Yes, which is also agreed by him. He is not denying this fact. He is just saying that the energy is being radiated away, leading to the observation of a lack of a conserved energy quantity

qwerty

I don't see the phrase 'radiated away' in his answer... :/

I don't know what that means, either

i don't understand why you're saying he isnt denying these things when he plainly says "it refutes claims made..."

naturallyInconsistent

@qwerty Hello? The paragraph starting: "The debate continued for many years especially in the context of gravitational waves which some people claimed did not exist. They thought that the linearised solutions for gravitational waves were equivalent to flat space via co-ordinate transformations and that they carried no energy. [...] The issue was finally resolved when exact non-linear gravitational wave solutions were found and it was shown that they do carry energy.

That's literally describing radiation of energy via gravitational waves

qwerty

12:14

but we're not discussing gravitational waves?

naturallyInconsistent

But it literally impacts the conversation.

You cannot get conservation of energy without also considering gravitational waves

qwerty

you model gravitational waves when you perturb some background spacetime?

naturallyInconsistent

It is one of the things making Killing vectors not exist

ACuriousMind

@qwerty The problem here is that it's more a debate about terminology/usefulness than disagreement about the facts: The "energy" that Gibbs constructs depends on a specific vector field $k^\mu$. That this expression dependent on a choice of vector field is conserved has been known for ages (the "Komar superpotential" he keeps referencing is from 1959).

But most people do not consider this a suitable thing to call "energy", as there are potentially infinitely many different choices for such time-like vector fields.

qwerty

@naturallyInconsistent you do not need to consider gravitational waves in something like an FLRW universe. and you don't need dark energy or a cosmological constant. you find the same argument eg for a radiation dominated universe

@ACuriousMind omg

naturallyInconsistent

12:21

@qwerty your statements here are missing some bits that, sadly, crucially change the truthiness of your statements.

qwerty

@ACuriousMind why can't he just have been clear he was using an non standard definition

@naturallyInconsistent feel free to point them out

ACuriousMind

@qwerty because he's fighting against the orthodoxy suppressing him

qwerty

@ACuriousMind i told everyone i had ~vibes~

ACuriousMind

The shortcomings of Komar's quantities have also been known for ages, e.g. iopscience.iop.org/article/10.1088/0264-9381/2/3/018 points out that while this gives you something that looks like contributions to an energy, it doesn't always give you the right versions of it. For Kerr black holes you would expect it to yield the energy from the angular momentum of the black hole, but it gives you double the value you'd "like"

Madder

these definitions are very unclear. So a unit cell need to reporduce the lattice by translation of the vector $T=ax+by+cz$ for some unit vectors and constants. But does it mean i fix one vector and keep translating along side? well i will not cover whole space!
If i translate along ALL the vectors, i will get double regions! and thus break my "no overlap" rule
EG: square lattice. 2x+y translation will put the square shape of my unit cell in that area, then i will have to do with 2y+x combination, but this is the same, and i will draw it double! breaking my no overlap rule...

qwerty

12:26

@ACuriousMind thanks for coming in with citations. i felt like i was going mad.

naturallyInconsistent

go queue up behind Mr Madder here

Madder

@naturallyInconsistent what :D ahhahahha oh i get it now

naturallyInconsistent

lol

qwerty

@naturallyInconsistent Please do point out what you think is wrong with what I said, as i remain fairly confident in it. I don't mind criticism as long as it's constructive.

ACuriousMind

My personal opinion is that some people just are too obsessed with "energy" because we raise it to this quasi-mystical status in many (especially introductory) contexts instead of just treating it as a conserved quantity like any other that's useful but may or may not exist.

naturallyInconsistent

12:32

@qwerty It aint wrong. You were just missing caveats and specifications that are necessary to judge if you were correct or not. You are most likely correct, and so I didn't find it necessary to poke at that

qwerty

@ACuriousMind i agree strongly.

naturallyInconsistent

@ACuriousMind well, given how important it is in modern physics, its possible non-existence is quite disturbing

qwerty

that said, i still want to unpack it, and based on how this evening has gone plus my main site question being unanswered, i fear it may be annoying me for a long while yet.

ACuriousMind

@naturallyInconsistent I'm not at all disturbed by it - that it doesn't exist (or is ambiguous to define) in certain cosmological models has no impact at all on its practical uses which are well-approximated by models with unambiguous time-translation-like symmetries and hence nice conserved energies.

naturallyInconsistent

@ACuriousMind but surely you can understand that people who are disturbed by that non-existence are not just "too obsessed"...

qwerty

12:36

i was in a bookshop today and was leafing through their copy of oxford dictionary of physics. it actually claimed without caveat that energy is always conserved and any statement otherwise was crackpottery.

naturallyInconsistent

lol

ACuriousMind

@naturallyInconsistent My claim is precisely that being disturbed by its non-existence is a sign that they attribute more meaning to a conserved quantity (whose existence is always dependent on the model!) than they should. By "obsession" I mean that instead of re-evaluating their beliefs about energy they instead insist it must always be there without having any rational reason to believe that.

naturallyInconsistent

@qwerty well, you see, outside of cosmology, the rest of the scientific community has to, you know, defend the entire scientific community from the scammers trying to violate the 2nd law of thermodynamics or something like that

@ACuriousMind hmm; as long as their existence is on grounds as firm or as shakey as momentum also is, then I'm fine with your position on the matter

qwerty

@naturallyInconsistent "you can't handle the truth!" ? a caveat saying GR is different would have sufficed ;)

naturallyInconsistent

lol

ACuriousMind

12:40

@naturallyInconsistent The funny thing is that (kinematic, conserved) momentum obviously has exactly the same "problems" in GR as energy but we see orders of magnitudes more discussion about "energy" in GR than momentum. That's an effect of this unreasonable obsession with energy compared to other conserved quantities.

naturallyInconsistent

@qwerty Well, you might get to comment on that if you first earn 100k rep on science advocacy.

@ACuriousMind which is why miao miao stated it the way miao miao did~

ACuriousMind

@naturallyInconsistent I regret to inform you that earning 100k SE rep does not make you an author of the Oxford Dictionary of Physics.

qwerty

@ACuriousMind they always forget the second term in the energy momentum relation too

@ACuriousMind giggles

naturallyInconsistent

Anyway, yall, im really enjoying the whole series of videos that Angela Collier put on youtube. She's great to listen to

@ACuriousMind drats!!!

qwerty

I'll send a firmly worded letter to the editor

lol

ACuriousMind

12:45

@qwerty You mean the endless parade of questions that are answered by pointing out that $E=mc^2$ is only at rest? Yeah, that's a related failure of pop-sci that plasters $E=mc^2$ everywhere whenever Einstein is mentioned so people think it must be some important or central formula that always holds...

qwerty

@ACuriousMind yup

uh btw @JohnRennie just tagging you in case you were interested in how the rest of the discussion went. any further 2cents welcome :)

naturallyInconsistent

Miao miao makes it a point to write that it is $E=+\sqrt{(m_0c^2)^2+(\vec p c)^2}$ hence $E\approx m_0c^2$ when $|\vec p|\to0$ just so that students get to see the full expression before uni.

Slereah

@ACuriousMind It's even easier to see really

If you drop an apple, it falls

Very clear example of an object gaining momentum

due to the break of translation invariance in Schwarzschild

qwerty

@ACuriousMind ACM, is it worth leaving this as a comment to his answer (plus what you said it about being a non standard definition basically)? I feel it would be useful

Slereah

It's even radial momentum bc that's the degree that's broken

ACuriousMind

12:56

@Slereah Yeah, that's a good point, because it means that whatever Komar's conserved quantity for the radial vector field would be, it's not what we would call momentum, so why should we assume the corresponding quantity for a time-like vector field is what we would call energy.

qwerty

oooh.

Slereah

though I'm cheating a bit because that's not true of energy even though naively it also should change

I should check what the momentum gain in Schwarzschild is

ACuriousMind

on the other hand of course the corresponding Newtonian model also doesn't have momentum conserved as long as you don't include the Earth moving towards the apple, too, so people don't find this disturbing :P

Slereah

Imagine you try to drop an apple and the entire Earth just goes up instead

qwerty

@Slereah the real meaning behind en.wikipedia.org/wiki/The_Son_of_Man

Transcript for