The h Bar - 2015-08-30 (page 2 of 2)

« first day (1760 days earlier) ← previous day next day → last day (3184 days later) »

5:00 PM

Really the whole "scattering event" is just a big cheat because the real interacting QFT is awful

we talk about the particles in the in and out states at infinity

wherein the itneraction is negligible and we can just define particles in terms of the free theory

@Slereah One of the best cheats we've ever invented, though :P

The whole notion of "particle" stems from the Fourier transform of the field which yields a nice superposition of momentum states

Once you're non-linear, though, you can't have superposition of solutions

And that whole notion goes out the window

though really I'm not that fond of calling momentum eigenstates "particles"

It's kind of a weird nomenclature because they are not localized at all

They aren't even proper states, if you want to be exact

People don't even use particle states when they derive scattering events

Well they are if you're in a volume :p

Otherwise they are rigged Hilbert space bullshit

5:04 PM

Exactly :D

but well

IF THE UNIVERSE IS CLOSED

Then they totally are states

No worries

just have to find the universe's topology

I don't have the chance to do QFT (except possibly as a two day short course in October organised by the uni) because I am doing my chem honours thus cannot do physics honours until later (since you cannot do two honours at the same time)

But the above should get me concpetually prepare for it when I read them up since my freinds recommend some books for me to read

O btw, just in case I raised that "how to interpret scattering" question again due to memroy failure, jsut remind me it is state A->blasckbox->state B and I should be able to recall our discussion

Scattering is basically a trick

You pretend that the problem isn't interacting in the past and in the future

Because those you can do mathematically

Then you stitched those two together awkwardly

Using the

SCATTERING MATRIX

We can't solve in the middle but we can approximate by perturbation series

Hence the whole Feynman diagrams

Feynman diagrams are probably the closest you can have to an "intuitive explanation" of it

All possible processes are summed

I see thanks

O btw below is my full background in quantum in my uni:

user image

Pretty good one

5:11 PM

there is not enough hilbert space, however

If you have any QFT questions feel free to ask

Well not every QM methods use Hilbert spaces

Or at least not explicitely

Most introductory courses just use the Schrödinger formalism

Which just use $L^2$ functions instead

(Which do form a Hilbert space)

The library internet works just fine for some reason.

@ACuriousMind It's all those torrenters in Brown Hall!!?

When it comes to scattering, light interacts with light. Not many people know that. Some people think pair production occurs because pair productions occurs. Spontaneously, like worms from mud. That's what you call cargo-cult science.

For shame @ACuriousMind :p

:P

5:13 PM

That is why I said "not explicitely"

ACM likes correcting people.

thus there's a gap in the understanding
Our uni like to stick to wave mechanics for 3rd year courses and introduce dirac/bra ket notations

They use them more only from honours year onwards

It's more commonly taught because you can just use dumb old functions instead of a weirdo Hilbert space

You can totally do QFT with the Schrodinger picture, by the way, but it is very rare

Because you have to do terrible things

You have to use WAVEFUNCTIONALS

And solve functional equations

what the hell are wavefunctionals

like a wavefunction

But a functional instead of a function

5:15 PM

@Slereah It gets funny though when they discuss "boundary conditions" for solving the Schrödinger equation when $L^2$-functions don't have well-defined boundary values

reference?

And what's a wavefunction?

A fucntional is something that takes in vectors and spits out scalar (that's the abstract definnition)

$\Psi(x) \rightarrow \Psi[\varphi(x)]$

@Secret I know what a functional is...

5:15 PM

@0celo7 a transition amplitude is defined by a path integral from a boundary state to a final state

a wave functional is just when you leave out the final state

so it's a path integral that takes the final state as input

explain

I need an explicit thingie

Basically QM to QFT in that formalism is, you replace $x$ by $\varphi(x)$ and $\partial_x$ by $\frac{\delta}{\delta \varphi(x)}$

@0celo7 idk how to write it here

we usually write it in terms of path integral diagrams

too complicated for me

OK, I'm off. @0celo7 : watch out that you don't get lost in math.

5:17 PM

and my reading list is waaaay to long

One thing I wanna do someday is like

Make a list of all the QFT formalisms

how do I reduce my reading list

And try to solve free scalar fields in all of them

Weinberg's formalism is all you need

cluster decomposition

@0celo7 you start doing physics research

your reading lists will dramatically decrease quickly

5:18 PM

@FenderLesPaul that might start soon-ish

next week is when I have to go around and beg for a position

I know another freshman is doing research, but tbh it's boring as all hell

he's looking at how motors burn out...

pretty hard topic to do with QFT

@JohnDuffield The only thing I can thought of from "light interact with light" is https://en.wikipedia.org/wiki/Breit%E2%80%93Wheeler_process
Cannot wait when they start the experiment and report the results

@secret depends on what you mean by "light interact with light"

you can write down Feynman diagrams with incoming photons that scatter at higher order through intermediate electrons say

BF4 premium is on sale

oh no

if I buy this I won't surface for a month

e.g. Delbruck scattering

5:27 PM

halp

@FenderLesPaul what do you say to: help Ryan do analysis exercises Skype calls

@0celo7 alright :p

you're lucky I have no social life

dude origin.com/en-us/store/deals

I'm about to spend so much money

ooh dragon age

I have the demo

I need to play it to see if i like it

but holy shit these sales

Mass effect bundle for $30 o.o

user image

I get that if I get the premium version of dragon age

that just looks too cool

I think you mean "eeeh Dragon Age"

5:31 PM

@Slereah is it not good?

I'll bbl

have fun guys

I've always wondered who buys these silly premium editions.

I have not played it so I cannot confirm or deny

From what I have heard, it's okay

@ACuriousMind I'm going to get BF4 premium

But it is from Yahtzee, and he is a notoriously grumpy man

5:32 PM

5 DLCs, more guns, more random shit

I liked the first part, never played the others

you have to have premium for BF4 it seems

all the servers have DLC on

so 500GB is nothing near what one needs for gaming

@ACuriousMind BF3 premium was a great deal

4DLCs, priority server reservations

you could cut in front of people in server queues :D

6:32 PM

oh wow the whole school network crashed

they are having some problems today

6:58 PM

My hard drive Big blue blob: GTA 5

user54412

7:14 PM

@0celo7 My university's job fair had exactly two types of jobs: finance for math majors, and coding for CS majors.

user54412

If you weren't in one of those categories, they weren't interested in talking to you.

user54412

Hopefully your school does a better job of inviting companies.

@ChrisWhite I haven't spoken to anyone yet who didn't get a summer job with a company.

At least within my department.

7:55 PM

@ChrisWhite Yikes! Does a physicist trying to learn math count? :P

8:22 PM

o.0

what is so shocking

idk

what is shocking is that BF4 is 100GB

I mean that's just insane

damn I gotta tutor today

le sigh

I hate making ends meet

@0celo7 The larger the file is the more poorly the game was engineered.

8:27 PM

@0B3 Maybe

@FenderLesPaul Elementary school?

But GTA5 is 60GB and it's a technical masterpiece

Also my BF4 contains 5 DLCs and a dev version

no I tutor intro physics here

@0celo7 I think it's reasonable then.

typically freshman and sophomores

sometimes juniors if they fail at life

8:33 PM

user image

forgive handwriting, doing this with the trackpad

That's 365GB

@0B3 that's not really true anymore...GTA5 has a lot of uncompressed audio...there are hundreds of thousands of lines of dialogue in the game

It's all very high fidelity

also the textures are all full 4K

you can make Skyrim huge by getting 4K textures for everything...but I think I've hit the limit where it just doesn't like loading everything anymore :p

@0celo7 What program is that? ^^

@Danu windirstat.info

getting it for my Mac partition too

it helps rooting out wasteful files

8:51 PM

Neat, thanks

np

9:32 PM

0

Q: Does anyone thing that it would be useful to notify the question asker when someone has started drafting a response to their question?

I dunno if this is a good idea. I just it might keep people more engaged with their questions so that an answer doesn't come in that doesn't get marked.

user54412

9:43 PM

Something I don't think I'll ever find the motivation for: writing the introduction in a paper.

@ChrisWhite are you motivated to help me with my ODE homework

user54412

Most readers skip it, it contains nothing particularly novel, and every introduction in a given subfield is basically the same as every other. "Ever since <Nobel laureate> discovered <ubiquitous thing>, <field> has been studied extensively..."

2

user54412

@0celo7 Motivate me. Make it sound interesting.

If you give us an arXiv link to your paper, I promise I will read it.

@ChrisWhite I need to find the second order equation whose solution is a circle of fixed radius $r$ centered at $(h,k)$.

I am taking this to mean a two-parameter family in $h,k$.

So I wrote $(x-h)^2+(y-k)^2=r^2$, and took the derivative twice. This gives me $$k=\frac{1+y'^2+yy''}{y''}$$

Then I substituted it above into the first derivative equation and got $$h=x+yy'-y'\left(\frac{1+y'^2+yy''}{y''}\right)$$

I then have to substitute both of these into the circle equation...but the simplification is horrendous.

...what problem do you have except not wanting to do the calculation? :P

9:52 PM

There has to be an easy way that I'm not seeing!

Much like analytic solutions, the easy way is also rarely certain to exist.

user54412

I'm a little confused by what you're trying to do.

@ChrisWhite What part is confusing?

user54412

so you're looking for an equation for y as a (possibly implicit) function of x

No, I have the implicit equation.

I'm trying to find the ODE for which it is a solution.

user54412

9:58 PM

strike that

Hmm, maybe I have to sub variables or something?

user54412

an equation relating x, y, y', y'' such that only x,y pairs on the given circle work?

@ChrisWhite I...think so? You're making this more complicated.

user54412

i'm trying to figure out how you end up with a second order equation whose solution is that particular circle

$h,k$ are the integration constants.

$r$ is fixed and can feature in the ODE.

10:02 PM

@ChrisWhite You don't want a particular circle. He wants a second-order equation such that the "initial conditions" determine the center of the circle.

user54412

where do initial conditions come into this?

user54412

@ACuriousMind ahh

Welp

I sent an email to Visser

I hope he will know better than Stackexchange :V

Did you link him the PSE question?

Nah

10:04 PM

The solution is apparently $((y')^2 + 1)^3 = (ay'')^2$, by the way, so I don't think there's a trick to this exercise other than just chugging through the calculation

I just asked him directly

@ACuriousMind I know what the solution is.

Did you find the book online or something?

math.SE

...

I was looking at the first answer on that page

Did not even see the second one

lol

10:07 PM

yay $k$ is right

and $h$ is right

(I see I did not simply all the way)

@ACuriousMind thank you

so I'm not a failure after all, just lazy

17 mins ago, by ACuriousMind

...what problem do you have except not wanting to do the calculation? :P

None, then

well I derped and missed a very useful simplification

so I would have gotten there eventually

@ACuriousMind where is your Skype avatar from

some people think Yu-Gi-Oh

@0celo7 Dark Eldar from Warhammer 40k

that's so nerdy

Well, I am a nerd.

10:12 PM

are your socks matching ATM

Not wearing socks ATM

(that was a weird turn in the conversation :D)

Germany is cold

you should wear socks, you might get kidney/bladder problems

@0celo7 Today was a 35°C day, it's not cold

high today was 31

damn Germany boiling

::looks on floor::

>nerdy

>dark eldar

That is so third edition man

10:18 PM

Oh, someone clipped their nails here. That's just nasty.

Totally sold out by that point :V

::hangs head in shame::

ACM is a quantum sellout!!?

I can nerd pretty hard on Warhammer

>playing tabletop games

You people...

10:20 PM

@ACuriousMind : Res. recom. qs are a touchy issue. On one hand, they tend to be primary opinion-based (duplicate?) list question, so they should ideally be closed. However various meta posts indicate support for their existence. On the other hand, they are easy to ask and easy to answer, so one school of thought is that they should only be CWs.

Ideally the mods are supposed to closed the res. recom. qs that don't meet quality requirements but in practice it is difficult to draw the line without entering a fight with OP. VTC from the community could be helpful.

done with ODEs!

woot

time to do LA, then go eat, then do physics, then do english

@ACuriousMind what should I eat tonight, you have to choose

if you don't, I will go hungry

user54412

I might add that it doesn't quite seem right that resource recs should earn reputation. The question proper doesn't really contribute anything or show research effort. And the answers get votes based on how much the community likes the book, not based on how well the answer is written.

@Qmechanic So, uh...we don't really have a consistent policy?

@ChrisWhite I could go there now and get a discount...

@ChrisWhite the new school of thought from SE ("new" since a couple years ago, I think) is that if it doesn't seem right that a poster earn reputation for a question, the question is bad and should be closed. Community wiki shouldn't be used as an excuse to have bad (in that sense) questions on the site.

10:28 PM

@0celo7 Uh...I don't know what to choose from!

@ACuriousMind Well, we have Subway and some taco place in the dorm building.

Then there's the dining hall right next to the dorm.

They have everything there, pizza, salads, sandwiches, Chinese, fried potatoes.

Might be some pulled chicken left over from lunch.

user54412

@DavidZ Well, I always thought we shouldn't have resource recommendations in the first place. Allowed but CW seems the lesser evil compared to allowed and earning rep.

Then there's Chick-Fil-A below the dining hall (but I went there last night on the way home from the library).

Oh Chick-Fil-A is closed (Sunday)!

I agree with Chris, res. rec.s should be either CW or disallowed, if CW is not an option, then forbid them.

@0celo7 Chinese it is

@ACuriousMind I don't like their Chinese, try again

10:32 PM

@ChrisWhite well, in my view (based on input from the network and the "overlords") allowed-but-CW is an abuse of the system and shouldn't exist. Actually IMO community wiki mode shouldn't exist at all.

I could go down to Panda express on the strip

But that's a good walk and I want something not in the opposite direction.

@0celo7 Well, you can't transfer your choice and then not accept it :P

@ACuriousMind Truth be told, the only time I had Chinese at the dining hall, I had to go home and sit on the toilet for an hour.

I know one data point is not correlation, but I'm very weary.

user54412

I think resource recs as CW would be like "Tips for <language>" on Code Golf, which I thought were CW. But @dmckee this seems not to be the case anymore? I'm curious what happened there.

@DavidZ Why do mods have the possibility to make questions CW, then?

user54412

10:37 PM

@0celo7 That's possibly a sign it was legit Chinese food.

@ACuriousMind I don't know. It used to be slightly relevant and maybe the people at SE thought there would be too much complaining if they removed it.

Maybe they use it for something on other sites? But I don't know what.

::stomach rumbles::

I guess it might make some sense as a mod-only feature, for things like policy recommendations on meta. But even there, it would only be an informational marker to indicate that the question is considered to be owned by the community, in some sense, not associated with the original poster. The valid use cases are quite rare IMO.

@ChrisWhite indeed

D:

My dad says his time stationed in Vietnam and Taiwan has allowed him to eat anything without repercussion.

I have a question to the room. I'm just getting to the theorems of homology/cohomology isomorphisms (along the lines of, "simplical homology $\equiv$ smooth singular homology $\equiv$ smooth singular cohomology $\equiv$ de Rham cohomology" if I get that right). What physics-related problems does this all solve?

10:43 PM

Gauge theory.

I mean obviously it solves some problems of homotopy. What does figuring out homotopy of smooth manifolds do for me?

oh?

It also has applications in string theory.

You need to know the cohomology of Kähler manifolds to do compactifications. (I know this because I've skimmed some stuff, I haven't worked through it. Consequently, that's a very rough [possibly wrong] sketch.)

Although in String Theory you don't use de Rham, you use the Dolbeault cohomology.

Hmm. Do you know if it relates at all to the relation between a lie algebra and its lie groups?

I mean, is useful*

I know ACM has talked about cohomology in context of Lie stuff.

yay!

10:46 PM

But my knowledge of Lie groups is quite limited I'm afraid.

@ACuriousMind I'll give the Chinese a shot.

Well the chain "smooth singular cohomology of a manifold is an invariant <-> simplical cohomology of any triangulation of a manifold is invariant <-> euler characteristic is an invariant" is enough for me to learn it but it's hard for me to see homotopy as useful in physics.

@NeuroFuzzy de Rham cohomology is closely tied to integrability for closed forms, so it tells you for example when you might find global gauge potentials for field strengths

@ACuriousMind OH.

@NeuroFuzzy There is a notion of cohomology for Lie algebras, and certain things in the canonical resolution used to compute this homology may be seen as the physical ghosts and anti-ghosts, see this post

Generally, the mathematical application of "stuff with different invariants is not homotopic" of (co)homology is not relevant to physics, but the ties of the cohomology to forms and other differential geometric properties makes it important in some settings

For once something that's not from QoGS

11:01 PM

@0celo7 Well, they do talk about this, too, they just never tell you the complex whose homology you are computing is the Lie algebra cohomology of the Chevalley complex because QoGS is in a setting where the constraints are generally not considered as a Lie algebra (I'm not sure under which assumptions they are).

user54412

Ok, international and latex-using folk: how do I properly enter Ma in a .bib file?

Write it out as Maria? :P

Lol

user54412

I try to enter names exactly as they appear on the byline, but this might be too much.

user54412

Not only is the a superscripted, it's underlined.

11:20 PM

Probably depends on whether you're using biber or bibtex

Alternatively, see this:

3

Q: How do I insert an underlined ordinal indicator, if the font does not have that glyph?

Basically what I want to do is underline ordinal indicators ª and º, because some fonts do not underline them.

fonts text-decorations

@ChrisWhite Nearly a year ago there was a big round of re-debates on CodeGolf because it had become clear that the site was never going to graduate as it was.

@ACuriousMind I went to get Chinese but no one was there!

I might even have set it off by announcing that I wouldn't be enforcing some of the "what challenges do we accept" rules for a while as an experiment.

They must not have "international" food on Sundays.

In any case, the site got a third wind from that and because all the original pro tem mods were burned out the team grabbed us a new one from the revived membership.

Since then I have not been active as a mod and have been expecting them to kick me out, but I keep forgetting to take any action that way myself.

Any way, long story short, one of the new decisions was to treat "tips" as full citizens.

So there are new challenges that I don't like very much, but given that it seems to have upped the activity a much that is all for the good.

user54412

11:32 PM

@DavidZ That actually works. But now I see my journal's documentclass eliminates all but the first letter anyway.

Lol, should I put "on all time top users list for general relativity tag on physics stack exchange" as an "achievement"?

@0celo7 I don't see you there.

very bottom

I have not been very active lately

That always displays the current user's stats.

seriously

11:37 PM

Yep.

fuuuuu

That's why I'm on there. In my view, at least.

who is the lowest

oh it's ACM then

It looks like @ACuriousMind, with 121.

I only have 91

damn

11:39 PM

Just answer a really popular GR question and get 31 upvotes. Problem solved.

"computer skills"

my high school did not even have a CS class

I know some java, but I'd hardly classify that as a a skill...

user image

huh?

"LaTeX" is not a computer skill, sadly

@0celo7 It's never crazy to ask a scientist for pedagogical resources.

It's never crasy to ask someone for help learning something.

If they choose to ignore you, no harm done.

@DanielSank Ok, I'll look into the group and asking someone there.

Should I outline what kind of background I have?

I want a serious textbook on the subject.

« first day (1760 days earlier) ← previous day next day → last day (3184 days later) »

Transcript for

Aug '1530

The h Bar

General chat for Physics SE (physics.stackexchange.com). For M...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile