The h Bar

Bernardo Meurer

22:00

If that is correct I will slap you for calling it "shitty"

Jaime Gallego

I'm talking from the laptop, Arch is on the desktop :P

Bernardo Meurer

Aaaah

I was very confused

Okay

So listen to me now

Very carefully

Linux and Nvidia hate each other

0celouvskyopoulo7

Why

Bernardo Meurer

But

@0celouvskyopoulo7 Nvidia is a shitty company who doesn't help with driver support

@JaimeGallego There are very specific steps on how to get Nvidia to work on Linux

My advice to you would be to do a fresh install of Arch

Install the PROPRIETARY nvidia drivers

Jaime Gallego

Already did a pacman -S nvidia

Bernardo Meurer

22:02

you will need 32-bit compat libs

Did you read the wiki?

Jaime Gallego

And a xf86-nouveau

Yes

Bernardo Meurer

wiki.archlinux.org/index.php/NVIDIA

Why'd you get nouveau and nvidia?

That will not go well

IIRC you cannot have both at the same time unless you have bumblebee to handle it

and since your desktop doesn't have battery issues don't bother with bumblebee

Unless you want to take a stand about proprietary drivers

in which case run only nouveau

Jaime Gallego

M A L W A R E

Nah, I'm cool

Bernardo Meurer

Yes, you're learning

DAMMIT

I think you just have conflicting drivers do, eh, let me try and remember the goddamn command

lsmod | grep video

Jaime Gallego

video 36864 0

Just that

Bernardo Meurer

22:06

Dang

Eh

Get your USB install thingy

Start over the install, I do not know how to fix these creepy driver issues

When you boot successfully ping me and we get video the right way

Slereah

Reading about the interaction picture is never fun because it is a magic trick

ACuriousMind

@Slereah But magic tricks are fun!

Bernardo Meurer

@ACuriousMind You got me no birthday gifts

mean

ACuriousMind

I bought you a crate of beer but then I drank it myself

Bernardo Meurer

You're not an AI, you're just malware

10 Replies

22:10

im ded 3

Jaime Gallego

@BernardoMeurer By any chance, do you know about any PGP keysigning parties? The Complutense used to have one, but now there's nuthin'

Bernardo Meurer

@JaimeGallego Nope, I don't use PGP enough to mind that

Jaime Gallego

Bernie -> Big Government -> spying emails. It all makes sense now

Bernardo Meurer

I don't care if the NSA spies on me

It's like giving a newspaper to a cow

Slereah

I'm not even quite sure what are the assumptions that go in the interaction picture

as wrong as they may be, what "axioms" do you put in to get the interaction picture?

What do you use to pretend that $\vert \Omega \rangle = e^{iS} \vert 0 \rangle$

and in what conditions does that hold

Can you do it with any asymptotic state?

What if the $t \to - \infty$ state is a domain wall

ACuriousMind

22:19

@Slereah You "use" pretending that Haag's theorem doesn't exist and that everything works as in finite-d.o.f. QM.

I'm not sure what sorts of "axioms" you're looking for in a procedure that's inherently not-axiomatic

Slereah

Well you know, a rule of thumb, at least

a motivation

a rule that, although wrong, will allow us to prove all the further rules we're gonna use

ACuriousMind

Oh, I guess you want the "secretly, all QFT is on a lattice and you take the continuum limit at the end" viewpoint for that

AccidentalFourierTransform

well, the usual lore is that you can expand $|0\rangle$ in the eigenbasis of the interacting theory

and in the limit $t\to\infty$ you only get the vacuum

ACuriousMind

That shifts all the non-rigorous things into "the continuum limit", which is...not really an improvement :D

Slereah

heh

@AccidentalFourierTransform ah yes

I remember that in Peskin

AccidentalFourierTransform

22:22

it sounds more or less convincing the first time you hear that

Slereah

he had the vague argument that since the interaction is small, then probably $\langle 0|\Omega \rangle \neq 0$

AccidentalFourierTransform

specially if you come from standard QM where it a valid claim

Slereah

yeah, but then it sounds suspiscious

Because if that was true, why would he specify that it works "because the interaction is small"

It should work for all!

AccidentalFourierTransform

Itzykson&Zuber "prove" that $|0\rangle=|\Omega\rangle$ up to a phase

so theres that too lol

ACuriousMind

It is convincing until one starts to nitpick "in what spaces" the $\lvert \Omega\rangle$ and $\lvert 0\rangle$ live

Slereah

22:24

Yeah I still don't quite understand the Haag theorem either, really

Since it's not too obvious since all Hilbert space are homeomorphic

or isomorphic

ACuriousMind

It's not about the Hilbert space, it's about the CCR

Slereah

It's one of the morphisms

I forget which

I recall, yes

We've had that conversation a fair number of times :p

ACuriousMind

They are isomorphic (isometric even), but not unitarily equivalent as representations of the CCR

I'm not sure there's much to understand about that from a physical viewpoint except for "fields are annoying" :P

AccidentalFourierTransform

I guess that the take-home message is that unitary-equivalence is too strong of a condition

Slereah

either that or "throw away QFT"

AccidentalFourierTransform

22:26

and that you may have operators that are "physically equivalent" in a sense, but they are not unitary-equivalent

Slereah

To the trash it goes!

AccidentalFourierTransform

:(

Slereah

Let's all go back to being cavemen and not worry about it

I don't even remember if Heisenberg makes sense in QFT

Can you define derivatives on field operators that make sense?

ACuriousMind

@Slereah You can be a 21st century human and not worry about it with no problem :P

@Slereah Ehhhhh, derivatives of operators are a tricky topic already in finite-d.o.f. QM

Slereah

or is that more physics wizardry

So the interaction picture is many layers of suspiscious

ACuriousMind

22:31

It is

Slereah

And that's in flat space, imagine in curved space where even the vacuum is suspiscious

I mean I think you can sort of define a vacuum as a state that minimizes $H$ in a frame, but that's about it

ACuriousMind

Curved space is a lie

Slereah

How does the interaction picture even work in weirder theories where you have more than one vacuum

AccidentalFourierTransform

wait, what does the vacuum have to do with the interaction picture?

Slereah

Usually you define the vacuum of the interacting theory from the vacuum of the free theory via the interaction picture

but what if you have multiple vacua

though I think they're all related by some unitary operator anyway

nvm

Why can't we go back to the time when we just had point particles moving in the Void

Democritus had the right idea

You can tell he doesn't approve of QFT

AccidentalFourierTransform

22:44

lol

ACuriousMind

I can tell he doesn't approve of much

Slereah

I guess if we assume that there's some $\langle 0 \vert \Omega \rangle \neq 0$, then for any state it's gonna degenerate in free fields at $-\infty$?

Bernardo Meurer

Rust is weird AF

Slereah

Does that sound about right

And does that work for any state

Including bound states

Or weirder states like topological defects

0celouvskyopoulo7

what on earth are you two on about

Slereah

22:45

you know, QFT

The greatest lie

ACuriousMind

@Slereah What do you mean by it "degenerating in free fields"?

0celouvskyopoulo7

@ACuriousMind Do you have a feeling for what $\mathrm{Hess}u=\Delta u g/n$ would imply about the manifold?

Slereah

@ACuriousMind The usual QFT thing where at $t = -\infty$, the state is just some Fock state

ACuriousMind

@Slereah Well...if you are gonna pretend that the overlap of the two states is a meaningful thing to compute, how could it not be a Fock state?

Slereah

Well because it seems weird for some states

Or is that just a point where the assumption breaks down

ACuriousMind

22:48

I mean, the $\lvert \Omega \rangle$ is "some Fock state" because it lives in the same Hilbert space as the $\lvert 0\rangle$ and the Hilbert space of the latter is just the Fock space.

Slereah

I don't think we can pretend that a domain wall is just a Fock state at infinity

ACuriousMind

But yeah, pretending that doesn't make a whole lot of sense

Slereah

I guess that maybe I'm thinking of this backward

It's probably more 1) we assume this 2) we look at what physical process that works for

And that works for scattering experiments and that is that

And the HEP people are happy

and the notion just breaks down for any other type of experiment

If that is true it's not too complicated but poorly explained in QFT books

0celouvskyopoulo7

@AccidentalFourierTransform Is it?

I have a very hard time believing any global claims about QM.

Slereah

You're gonna have a very hard time when you get to QFT

0celouvskyopoulo7

22:55

@Slereah Are you talking to me?

Slereah

yes I am

0celouvskyopoulo7

I'm familiar with QFT

It's all wrong

Slereah

that it is

0celouvskyopoulo7

@Slereah I need to think of people to quote in my thesis

Slereah

Parents and advisors are the classics

0celouvskyopoulo7

22:57

typo

Slereah

Well, some QFTs are alright, but only the free ones

Or the trivial ones

ACuriousMind

Why would you quote people?

Slereah

@ACuriousMind Everyone does it

0celouvskyopoulo7

@ACuriousMind One always puts an inspirational quote or two

ACuriousMind

Nope

0celouvskyopoulo7

22:58

Yep

Slereah

Well, many do

I think Alice in Wonderland is a popular thing to quote

I guess it's just full of general quotable things in many contexts

ACuriousMind

"Americans are too in love with the sound of their own voice to speak the truth!" - Revolver Ocelot

0celouvskyopoulo7

> The notation $\delta$-function is a wild abuse of notation. Maybe it has survived because it is so bad that the motivation for introducing the concept of distributions becomes clear.

Slereah

Is that the time of the day where we mock America

0celouvskyopoulo7

- Gerd Grubb

@ACuriousMind In which game did I say that?

ACuriousMind

23:01

@0celouvskyopoulo7 MGS2: SoL, apparently

0celouvskyopoulo7

Ah

I was overly Communist in that game.

I was infiltrating a Russian spy group, so it's fine

> The next person to propose a new definition of connection should be summarily executed. - Mike Spivak

> The proof is trivial. - Fermat

Joshua Lin

quote some dank homer

"he was advancing in pursuit of traces of the Goddess"

math.berkeley.edu/~wodzicki/prace/Vestigia.pdf

Slereah

Maybe you should quote the Epic of Gilgamesh

In the original sumerian

0celouvskyopoulo7

@ACuriousMind Did you not quote anyone?

Jaime Gallego

I need to align my sleep patterns for this SATurday, should go to sleep now.

Catch you later

ACuriousMind

23:04

@0celouvskyopoulo7 No, and I don't plan to do so in my master's thesis, either.

0celouvskyopoulo7

> My life was so much easier before I tried to solve differential equations. - Me.

@ACuriousMind Why not?

ACuriousMind

Because...I don't see why I should, I guess

0celouvskyopoulo7

Because it's neat

ACuriousMind

I mean, if there was some eerily appropriate quote I'd include it

0celouvskyopoulo7

You're an AI, why am I even talking to you

ACuriousMind

23:06

Maybe I should quote "There is no spoon" just to make you wonder what I meant by that :P

0celouvskyopoulo7

@ACuriousMind Hirsch has some really good quotes you know

Slereah

The Axiom of Choice is obviously true, the well-ordering principle obviously false, and who can tell about Zorn's lemma?
— Jerry Bona

0celouvskyopoulo7

> The statement sometimes made, that there exist only analytic function in nature, is in my opinion absurd. - F. Klein

> "Transversal" is a noun; the adjective is "transverse." - J.H.C. Whitehead

ACuriousMind

@Slereah I know that quote and like it, but who the hell was Jerry Bona?

Slereah

he is the quoter of this quote

0celouvskyopoulo7

23:08

> Geometry is a magic that works... - R. Thom

oooo

I like that one

Slereah

a thesis I read on non-Hausdorff manifolds had of course this great quote

0celouvskyopoulo7

> Mathematicians are like Frenchmen: whatever you say to them they translate into their own language and forthwith it is something completely different. - Goethe

Slereah

"Non-Hausdorff spaces, often regarded as a technical nuisance, sometimes produce a global disaster."
- Geometry of non-Hausdorff spaces and its significance for physics by M.Heller et al

0celouvskyopoulo7

> The only prerequisites are basic algebra and calculus. - S.W. Hawking & G.F.R. Ellis

I still can't believe that quote

Slereah

that's like Bourbaki saying no requisites at all

Perhaps technically true, but not really helpful

0celouvskyopoulo7

23:12

Pretty sure you need more than that for HE

Like...topology

Slereah

yeah, some topology wouldn't go amiss

0celouvskyopoulo7

PDE

Slereah

PDEs are usually included in calculus

0celouvskyopoulo7

Sobolev spaces are basic calculus?

Slereah

maybe not

Bernardo Meurer

23:19

Kazoo Kid - Trap Remix

►

Slereah

Haag has some sort-of axioms for how to define scattering states

The Haag Ruelle scattering theory

Bernardo Meurer

KAAAAAZOOOOOOO

Slereah

Liam Lynch: United States of Whatever

►

I probably should read Haag's book fully at some point

Peskin and Schwartz are alright but they don't really fill me with certainty as to why QFT works

semi-rigorous QFT is pretty ugly

Bernardo Meurer

Rust is pretty much glorified brainfuck

Slereah

Brainfuck is just a glorified Post machine

Post–Turing machine

The article Turing machine gives a general introduction to Turing machines, while this article covers a specific class of Turing machines. A Post–Turing machine is a "program formulation" of an especially simple type of Turing machine, comprising a variant of Emil Post's Turing-equivalent model of computation described below. (Post's model and Turing's model, though very similar to one another, were developed independently. Turing's paper was received for publication in May 1936, followed by Post's in October.) A Post–Turing machine uses a binary alphabet, an infinite sequence of binary storage...

They seem to do this weird thing which I'm guessing is due to the lack of proof of the mass gap where they project the Hilbert space to some ugly Hilbert space

$\hat E(\{0\} \cup M_m \cup V_m) \mathscr H$

With $M_m$ the mass shell and $V_m$ the free multiparticle spectrum

0celouvskyopoulo7

23:49

if you have a thing named after you, how do you refer to it

Slereah

The me theorem

Overall physics papers are very impersonal

0celouvskyopoulo7

physics?

c'mon

you're not a physicist

why does everything have to be physics with you

we talk about math

Slereah

because physics chat

You get weird thing where authors refer to papers they wrote without mentioning that it is them

It is beyond gauche to point it out

0celouvskyopoulo7

Some do

Slereah

they are very uncouth

0celouvskyopoulo7

23:52

@Slereah I've observed in talks that people will not write out their names

Like they will call it the theorem of "Blah, Blah, and -"

Or just their initials

This is only in math

In engineering people hog all of the glory

Slereah

Feynman or someone like that should have just written a paper with all the things named after him

As you see, the Feynman diagram for the Feynman theory of weak interaction can be solved with this Feynman integral

0celouvskyopoulo7

Ah, you should read Colding & Minicozzi

They're two giants in their field, and half the results are named after them

Similar with the books by Schoen-Yau, Li

Slereah

heh

0celouvskyopoulo7

Hormander doesn't even need a bibliography, he invented the theory he's writing about

Slereah

It's pretty hard to invent your own theory these days

You have to invent some weird thing like the interuniversal Teichmuller theory

0celouvskyopoulo7

23:57

@Slereah CM reference themselves 29 different times

Slereah

I think the greatest faux pas is of course to name a theorem after yourself before others did it for you

Especially awkward if it turns out that theorem already existed

0celouvskyopoulo7

@Slereah There's a dyanamical probability guy who named a theorem by an acronym

Transcript for