The h Bar

0celo7

01:44

@ACuriousMind Going through old records and found about 10 pages of normal vectors on Lorentz manifolds

good times

...20 pages

I was a little obsessed with that problem.

0celo7

02:03

@ACuriousMind he sent out the problem set

Bernard Meurer

Heya

0celo7

02:22

@BernardMeurer hi

Bernard Meurer

@0celo7 What did you need help with the other day? Sorry man, I've been ultra busy with the move

0celo7

Link?

I don't remember

Bernard Meurer

Then it probably wasn't important :)

0celo7

no, it was

link?

I forget important shit all the time.

Bernard Meurer

You literally just said "I need your advice"

I don't have a link

0celo7

02:25

hmm.

Oh, I found a used book for ~$380 less than a new copy.

Wanted to know if I should get it.

I did.

It shipped this morning.

(it was $20)

Bernard Meurer

Meh, seems like an OK decision

Ah, very good call then

0celo7

02:39

@BernardMeurer The issue is the seller doesn't have a good reputation

4 stars but only 70 reviews

some pretty bad

but I figured it was worth a hail mary

Bernard Meurer

Well the possible 380 USD savings it seems like a reasonable bet

although it all sounds a little too good to be true

0celo7

It's an older edition

and paperback

the $400 is the newest edition hardback

but it's not worth that at all

the book is out of print for some reason

it's a 5 volume set and only the second volume is out of print

Bernard Meurer

I wish I could afford TAOCP

It's expensive like that

0celo7

proof?

Bernard Meurer

amzn.com/0321751043

I'm only buying this once Knuth dies though

Otherwise he'll just release 4B and screw up my set

Bernard Meurer

02:56

He's 78, it'll take a while

0celo7

03:11

hmm

not if we accelerate it.

0celo7

03:29

@BernardMeurer what's the math in those books

basic algebra?

Bernard Meurer

@0celo7 Haven't read it

Dunno

0celo7

then again

Hawking & Ellis is "basic algebra and calculus"

@BernardMeurer so when are you moving

Bernard Meurer

@0celo7 September the 5th

2 weeks

0celo7

do you want to learn some math

I'm going to relearn some GR

get some motivation to learn geometric analysis

Bernard Meurer

I want to learn some linear algebra

0celo7

03:40

what do you want to know

Bernard Meurer

In fact that's an understatement

I want to learn all the mathematical base I'll need for Shankar

or at least for starting it

0celo7

vector calculus, linear algebra, complex calculus

Bernard Meurer

So there, that's what I want to learn

0celo7

hmm

I should prove the Minkowski triangle inequality.

how the hell does it work when you don't have Cauchy-Schwarz

Hmm, I once proved the light cone is a lightlike submanifold

wonder where I wrote it down

Shit. Where is that proof

I should not have gone down the GR rabbit hole

I always meant to write down this proof

I'm regretting it now.

Really regretting it.

Bernard Meurer

03:56

Morning @JohnRennie

0celo7

Ok, it's probably an achronal 3-manifold

Probably.

At least it'll be Lipschitz.

Bernard Meurer

@0celo7 You should scan this stuff and keep an organized GDrive

0celo7

@BernardMeurer Trying to reconstruct a proof from memories 6 months ago, I need a few minutes.

Ah, QED.

It's trivial.

John Rennie

Morning.

0celo7

If you look at the tangent space of the light cone, it has to contain a null vector

Hmm, no, not that trivial.

Mar 10 at 20:51, by 0celo7

By Thm. 8.1.3 in Wald, $\partial I^+(p)$ is achronal. So it's either spacelike or null or some mix. If any part were spacelike, the normal would be timelike, which cannot be orthogonal to the null tangents. So it's null. $\Box$

the fuck

I need someone to translate me

AH!

NO.

AH!

Wait, what does spacelike even mean

If the hypersurface is timelike, it contains a timelike curve (have to prove)

which can't happen if it's timelike

Ah!

John Rennie

04:04

@0celo7 Space is deep

0celo7

Mar 10 at 20:59, by 0celo7

Since the metric is Lorentz, pick any timelike vector field on the surface. Then draw the integral curve.

Ah!

Thanks past me :)

AHA

Such a vector field always exists?

Sure if we're working in a spacetime

But what about some pathological Lorentz manifold

Can we always have local timelike vector fields

@JohnRennie you probably don't know...

Although I have a proof of what I wanted

now I'm wondering about the vector field thing

@BernardMeurer Maybe you should read Griffiths first.

Bernard Meurer

@0celo7 DS hates Griffiths

0celo7

so?

Bernard Meurer

There must be a good reason

0celo7

Oh, I think you can get such a vector field in a strongly convex neighborhood by parallel transport.

Bernard Meurer

04:09

Geodesics have parallel transportation right?

You taught me that

0celo7

@JohnRennie Ah, I've rediscovered GR

@BernardMeurer Parallel transport is a way of moving vectors along geodesics

and it's an isometry so you can use it to move a timelike vector around

and it stays timelike

Bernard Meurer

timelike?

When did time get into this

0celo7

we're doing GR

Bernard Meurer

To me we were just talking about a curved space

0celo7

tfw you're reading a GR paper and don't need to look at what ref 204 is because you know that on page 193 of HE is the figure they're describing

Bernard Meurer

04:12

does that imply GR?

0celo7

what

I'm obviously going on a late-night GR rant though

now I want to buy O'Neil

but no

I shall restrain myself

FOR NOW

I should read Penrose

DanielSank

04:27

@BernardMeurer Ha! You thought you could outwit me by not @'ing my name!

Nice try.

Bernard Meurer

@DanielSank I don't like pinging you :p

I never know what chaos that'll bring

DanielSank

@JohnRennie Morning.

Bernard Meurer

and hey, I'm not lying am I?

John Rennie

Morning

DanielSank

Well, it's not a good book IMHO.

Bernard Meurer

04:28

Ghost - Cirice (Official Music Video)

►

DanielSank

I'd even say it's bad.

Bernard Meurer

Also, p. interesting

DanielSank

Actually yeah, screw that book.

Bernard Meurer

See you hate it

0celo7

@DanielSank What's wrong with it

I haven't even looked at it

but if @BernardMeurer is not ready for Shankar...

DanielSank

04:30

@0celo7 The first page has the Schrodinger equation for a 1D potential written out in the position basis.

The first chapter or two make you compute commutators and integrals without having the slightest idea of where any of it comes from.

0celo7

@DanielSank I'm not a quantum physicist, can you please explain why that's bad

The first comment

DanielSank

Then chapter 3 introduces some matrix notation etc.

@0celo7 Well 1) it makes you think quantum mechanics is all about wave equations, 2) it fails to capture much of what makes quantum mechanics different from classical, because we have waves in classical. In other words, he makes you slog through a differential equation to see the discrete energy levels, whereas really it's discrete occupation of energy levels that makes quantum special.

0celo7

I'm trying to recall how Shankar does it

DanielSank

3) He forces tons of special functions and differential equations down your throat first and then shows you the finite dimensional systems where things are easier to deal with from a math point of view.

0celo7

I'm pretty sure you solve an ODE there too

DanielSank

04:33

@0celo7 Shankar spends 100 pages reminding you about normal modes. Then he goes on to quantum systems.

It's wonderful.

0celo7

Hmm

@DanielSank So Bernardo has to just learn enough math for Shankar

DanielSank

@0celo7 I would say the special functions are considerably more difficult than matrices.

Special function expansions don't make any sense unless you understand linear algebra anyways!

0celo7

I guess

@BernardMeurer so...learn some linear algebra

Bernard Meurer

@0celo7 I know, I really gotta work on that

0celo7

@JohnRennie Uhhh, what's $R^{\mu\nu\rho\sigma}R_{\mu\nu\rho\sigma}$ called

the something scalar

Kretschman!

John Rennie

04:46

Kretschmann scalar

In the theory of Lorentzian manifolds, particularly in the context of applications to general relativity, the Kretschmann scalar is a quadratic scalar invariant. It was introduced by Erich Kretschmann. == Definition == The Kretschmann invariant is K = R a b c d R a b c d {\displaystyle K=R_{abcd}\,R^{abcd}} where ...

Ah, you beat me to it.

Bernard Meurer

Dude this Ghost band is p. good

0celo7

stop listening to music and read linear algebra

Bernard Meurer

I'm just chilling before I head to bed

almost 2 AM

0celo7

I'm chilling by reading mathematical GR

chill with more linear algebra

DanielSank

@BernardMeurer It's turbo important and I will help you as much as possible.

Bernard Meurer

04:54

@DanielSank Turbo important? That's a new one hahaha

<3

DanielSank

No seriously.

It's at least as important as calculus and probably more.

Bernard Meurer

O.O

DanielSank

I am not exaggerating.

@BernardMeurer want to play a game?

Bernard Meurer

I know, that's why I'm surprised

Ah, sure

DanielSank

Ok

Suppose I have a box of mass m on a spring, and the spring constant is k.

Do you know the oscillation frequency of the box?

Bernard Meurer

04:56

Nope

DanielSank

oh

crap, ok

Bernard Meurer

Ha, take that

DanielSank

Well you know F=ma, right?

Bernard Meurer

Yeah

WAIT

It's coming to me

F=kx

right?

DanielSank

F = - k x

Bernard Meurer

04:57

Why -?

DanielSank

If the spring is pulled to the right (i.e. x>0) then the force the spring is exerting is to the left (i.e. F<0).

Bernard Meurer

When did we say we were pulling it to the right though?

DanielSank

We didn't, but if the box is to the right, then the force on it is to the left, and vice versa.

So (force on box) = - k (displacement of box)

i.e. F = - k x

Bernard Meurer

Okay, reasonable

DanielSank

(By the way, this is all to motivate linear algebra)

Bernard Meurer

05:01

But when you asked about the osc. frequency how will I know?

DanielSank

@BernardMeurer Well, you know F = m a, and you know F = - k x.

Now use your math skills.

Bernard Meurer

I mean, you didn't even say it was oscillating in the first place :p

DanielSank

-_-

Bernard Meurer

I think like a programmer, you can't blame me

DanielSank

Dude...

Bernard Meurer

05:02

Ah, okay, anyway

I don't see where you're going with this, neither F=ma nor F=-kx seem to have anything to do with frequency to me

DanielSank

That's because you need to go on...

Do you have mathjax on?

If not, turn it on.

Bernard Meurer

$-kx = ma$ ?

I always have it on ^^

DanielSank

@BernardMeurer Very good.

Let's rewrite this as $\ddot{x}(t) = -(k/m) x(t)$.

Bernard Meurer

I was thinking that couldn't be it, it was too simple hahaha

DanielSank

That's a differential equation.

Do you know how to solve it (and do you know what I mean by $\ddot{x}$)?

Bernard Meurer

05:05

No²

DanielSank

Ok, $\ddot{x} \equiv d^2x / dt^2$. Is that notation more familiar?

Bernard Meurer

Hmmm. I think so yeah

it's the second derivative of x?

DanielSank

...with respect to $t$, yes.

So our equation says (second derivative of displacement with respect to t) = -(k/m) displacement.

ok?

Bernard Meurer

Not everything is lost, okay

DanielSank

Ok, so any idea how to solve $\ddot{x}(t) = -(k/m)x(t)$?

(This is the single most important differential equation in physics, by the way)

Bernard Meurer

05:08

Get rid of the derivatives?

DanielSank

@BernardMeurer Haha, well that would work, yes. Do you know how to do that?

Bernard Meurer

Integrate!

DanielSank

@BernardMeurer I doubt that's going to help, unfortunately.

Integrate what?

Bernard Meurer

Tsc, don't ask me hard questions

Ahn lemme think

can we even get rid of them?

DanielSank

@BernardMeurer Yes, but you need linear algebra to do it :) However, I don't want to get into that quite yet.

There is a way to solve this without getting rid of derivatives.

Often, the only way to solve differential equations is to guess.

Bernard Meurer

05:10

Ah, put it on WolframAlpha, QED

DanielSank

@BernardMeurer that is acceptable.

What's the result?

Bernard Meurer

Dunno, trying

Oh wait, I still have a calculator that is not a computer!

Tsc, Wolfram doesn't speak TeX

$$-\frac{kx(t)}{m}$$

DanielSank

x''(t) + u x(t) = 0

^ It understands that

Here u is just k/m.

Bernard Meurer

$$x(t) = c_2 \sin(t \sqrt{k/m})+c_1 \cos(t \sqrt{k/m})$$

Ok, wat

DanielSank

Very good.

How much time do you have and how tired are you?

Bernard Meurer

05:19

10 minutes maybe 20 if we push it. I'm level 7 tired

DanielSank

@BernardMeurer Go to sleep. Let's continue when you're more awake. We're about to get to a place where you have to use your brain.

Bernard Meurer

@DanielSank Nice! That hasn't happened in a while :p

Apart from x86, god forbid

DanielSank

@BernardMeurer Good night.

Bernard Meurer

@DanielSank Goodnight man!

Secret

05:44

Attempt to understand this advice or comment (I cannot really tell which) have caused my mind to literally blown last night.

At the end of the dream last night, I felt my brain was fried from the inside. Right now I am in a bit disoriented state, lol

DanielSank

Someone asked me in my AMA if there is one physicist I really admire.

I mentioned Robert McDermott.

I should also mention Tony Leggett.

Anthony James Leggett

Sir Anthony James Leggett KBE FRS (born 26 March 1938), has been a professor of physics at the University of Illinois at Urbana-Champaign since 1983. Leggett is widely recognized as a world leader in the theory of low-temperature physics, and his pioneering work on superfluidity was recognized by the 2003 Nobel Prize in Physics. He has shaped the theoretical understanding of normal and superfluid helium liquids and strongly coupled superfluids. He set directions for research in the quantum physics of macroscopic dissipative systems and use of condensed systems to test the foundations of quantum...

DanielSank

06:29

Zomg @DavidZ we need to fix teh homewurk policiez.

user116211

07:16

0

Q: Confusion about what exactly is meant by "Field"

Does the "field" in physics same as the "field" in mathematics? Which one those is more abstract in Nature?

user116211

Too broad?

John Duffield

@0celo7 : you'll like my latest chapter.

@0celo7 : tell him they aren't constant, and I said so.

David Z

08:20

@DanielSank Yeah, I know, that's the plan for today's chat session

David Z

08:38

Incidentally if anyone hasn't rated sample questions, there's still time!

Secret

08:49

I am guilty of "Asks "what am I doing wrong" or equivalent" and "Asks "is this correct" or equivalent" type questions ocassionally

In particular, the What am I doing wrong type question are sometimes caused by a conceptual bottleneck, making it harder to work out whether it is too rely on the question asking medium (be it forums, PSE, chat room etc.) as a error checkign advice (too specific), or more on trying to understand the correct flow of logic to avoid these bottlenecks (more general)

This question return error 404 physics.stackexchange.com/revisions/248024/1

I am also at loss of handling questions that ask for "other approaches". Personally I found having multiple ways to solve a problem is important as not onnly it halps as a sanity check but also train our problem solving skills to look at the problem at multiple perspectives (which is an important skill in solving real life problems). In practice, few people are bothered to answer these type of questions because they are in some sense asking others to show the workings

https://physics.stackexchange.com/revisions/250549/2

David Z

@Secret ah, it's been deleted. I'll mark that.

@Secret Indeed, having multiple approaches is good. Something we can discuss is whether questions that present one approach and ask for another way to do the problem should be acceptable.

Secret

https://physics.stackexchange.com/revisions/247999/1

These questions are a confusing breed. It can be a flat out show workings to explain in maths how that happens, or (more commonly if the asker is layman) a soft type question where the asker are onyl interested in the qualitative treatment. I'll le others to decide on how to handle that

@DavidZ One thing we can add to that is how should the answerers give the answers to these type of questions. Personally, a guiding approach is better such as how when people in maths ask proof type questions, a sketch of the proof is given so that the readers will do the thinkinig. After all the important thing in problem solving is the thinking and the logical flow needed to solve the problem, not how the details are carried out

David Z

Yes, that will be part of our new homework policy discussion.

Secret

09:07

https://physics.stackexchange.com/revisions/253025/1

In real life, I like to ask these types of what if questions a lot. Various professors have different approaches to them, most tend to remind me on what I learn and what aspect I need to think more carefully about to arrive at a solution

However, these questions tend to be short, thus they can be dealt with using more real time medium such as chat rooms. Posting them as PSE questions, in my opinion, will be a bit too cluttered

There are however exists a subcateogry of these "what if" questions that actually lead to very deep questions.…

Dead link: physics.stackexchange.com/revisions/253477/2

https://physics.stackexchange.com/revisions/247730/3

These are very common small questions that is asked by students in a daily life basis. The answers to them are also quite short. I have no comment on how to handle these questions

Dead link: physics.stackexchange.com/revisions/253583/1

David Z

I've already identified all these links as deleted - nothing to do there

I only need to know about them if they're not already marked as such in the form (in the dropdown where you pick which question you're responding about)

Secret

https://physics.stackexchange.com/revisions/247547/1
https://physics.stackexchange.com/revisions/254734/1

The only comment I can give is that these two questions have something in common...

David Z

All this would be a lot more useful if you actually filled out the form, you know.

Secret

https://physics.stackexchange.com/revisions/248365/3
This might be an example of "Asks "what am I doing wrong" or equivalent" type question where the mistake is caused by a conceptual issue. In this case probably a consideration of 3D is missing somewhere

Now that I have ran through all the questions, I should try to pick one to fill in the form

David Z

The intent is that you fill in the form for all of them. Or some of them, if you don't have the patience for all. There's no need to pick just one. (Well, I guess you do have to start with one, then you can do another, etc.)

Secret

09:22

O I see, I thought you can only pick one (because that's how most questionnarie works

David Z

I'll clarify that

Secret

(Currently rating this question https://physics.stackexchange.com/revisions/247719/1)
Do you have some kind of comment box in the questionnaire, because for this question I felt like the user have not done any prior research simply because it is not applicable (His only external source is the physics text he found and from there he knew it is a computational problem and is wondering whether there are non computational ways to solve it (which might be because it is a new way to look at the problem, there may be no research material to reference on at all))

David Z

Yep, just keep it simple and pick the closest answer in the questionnaire. If the user did as much prior research as can reasonably be expected, even if that amount is small, answer accordingly.

Secret

09:38

(Side note 247829/1 is quite rude, although emotions carry no useful information)
*Onwards to the next question*

Noah P

09:51

@ACuriousMind Is it not a genuine university?! My bad... And I think it is definitely fair to say that although it may not be a huge gap, the top UK universities are certainly better than their European counterparts?

Secret

(currently on https://physics.stackexchange.com/revisions/247730/3)
This question also reminds me that next time if i have an unsure of some thing, i should double check whether i have overlooked something obvious (and hhow exactly do i felt unsure about something) before asking (wavelength vs wavelength & length)

David Z

@NoahP I would be quite surprised if there were actually a university named after sausage :-P

Noah P

@DavidZ You never know with these Germans

David Z

Indeed

Noah P

Though, who am I to speak - we do name our buildings after vegetables...thegherkinlondon.com

"How could the unique properties of black holes be exploited?"

How about that for a title?

(5000 word essay)

Secret

10:18

(3 questions left https://physics.stackexchange.com/revisions/256585/4)
From my experience, laymens don't know what "entropy generation" is, thus this rules it out as a layman level question

David Z

@Secret Probably, but it depends on the question.

Secret

I have 4 guesses on the nature of the asker:
1. A physics geek
2. A college student who have done some degree of stat mech or information theory
3. A researcher who ponder about the issue when he/she saw a result of some experiment (home based or not)
4. A parent of some background asking this question because they pick it up from their kids

I am going to assume he/she is 1-3 and rate accordingly. Others might have different vie on this question

all done, all question filled into forms

David Z

Cool, thanks

Secret

Btw, i hope the stats are weight against number of participants in order to avoivd bar chart results being determined by only one or a few data point

Alex

Hi all

David Z

10:33

@Secret of course; I'm not sure how you think it would be otherwise

Secret

there might be cases where there are only e.g. 2 people rated a question and thus the bar chart or pie chart that resulted will be not very good representative of the question's (insert word)

David Z

Gotcha

Secret

how many people have filled in the forms so far and what are the rough distribution for which questions being chosen to be rated?

or maybe we should detail the stat later to avoid introducing errors...

David Z

There have been 85 responses, approximately evenly split between the questions. I haven't done a detailed breakdown yet though.

Alex

Hi @secret I have a question regarding something we discussed briefly yesterday. I stated the de broglie relations as $\lambda = \frac{h}{p}$ and $\nu = \frac{c}{\lambda}$ and $E = h \nu$. Should the equation for the frequency be $\nu = \frac{v}{\lambda}$, where $\nu$ is frequency and $v$ is velocity? or was I right to state it as $\nu = \frac{c}{\lambda}$?

Secret

10:41

I don't recall discussing this question with you yesterday (although I do read through it in the background). However $\nu = \frac{v}{\lambda}$ is the general formula for wave speed and frequency and for light $v=c$

For light $v=c$ is a constant and thus only frequency and wavelength changes
For sound waves and other waves with a medium, $v$ is determined by the material's properties

Alex

@Secret Okay but in this case of De Broglie relations we are referring to probability waves, so $v$ could be anything right, it does not have to be $c$?

Secret

For probability waves, the $v$ is thus reflect the momentum of the system in question. Eg. for a non relativistic free particle, $v$ is proportional to momentum

Alex

@Secret Okay thanks.

Secret

wikipedia's article on matter waves have more in depth info (my mind seemed to be a bit rusty on de broglie relations because I have been thinking to mcuh abotu qubits recently)

Alex

@Secret Okay, I was just wondering why neither you are acuriousmind corrected me in stating $v = c$ as part of the de broglie relationship I gave yesterday (we discussed the commutativity of momentum and energy). But I guess that could be a valid $v$ as well for some system...

Secret

10:53

In general $v$ is tied to the relativistic energy of the system. For light the energy is E=pc, therefore using De broglie to rearrange, you get the expected $\nu=\frac{c}{\lambda}$

hence $v=c$ for light (and other massless particles

Alex

Yeah that makes sense.

ACuriousMind

11:55

@NoahP Better at what? I'm neither convinced that the quality of education can be ranked on a one-dimensional scale nor that the differences between the renowned universities of Europe are large enough to make for meaningful comparisons.

@0celo7 There will be a feast in his honor.

Transcript for