The h Bar - 2016-08-02 (page 1 of 4)

« first day (2098 days earlier) ← previous day next day → last day (2832 days later) »

1:55 AM

@vzn Hmmm, not sure if you mean in the wild or at an aquarium.

Anna and I found a California two-spotted octopus in the tide pools in San Diego once.

It was very curious about us. It tried to pull our fingers into its den.

2:09 AM

California two-spot octopus

The California two-spot octopus (Octopus bimaculoides) is an octopus species found off the coast of California. One can identify the species by the circular blue eyespots on each side of its head. Due to their friendly temperament and relative hardiness, they are considered by most experts to make the best pet octopus. Bimacs usually live to be about two years old. They are closely related to Verrill's two-spot octopus (Octopus bimaculatus). == Range == This species is found in the eastern Pacific, from the mid-California coast to Mexico, the Indo-Pacific from East Africa to American Samoa, north...

user54412

"friendly temperament"

user54412

I'm struggling to imagine how a (small) octopus can display an unfriendly temperament

2:27 AM

@DanielSank wondering if you know of/ have seen other aquariums with them, seems like they are maybe not that common...? hard to keep?

2:46 AM

@vzn I saw octopuses at the Baltimore and Monterey aquariums.

3:02 AM

@JohnRennie I've thought a lot on your last writings. And slept. After I've woke up, afaik I've understood this. Now my best hyphothese is, that you are bored caretakers fighting against (4).

3:12 AM

Guys

What part of AQFT do you think I need to tweak to crowbar in some Colombeau algebras

I'm guessing the $C^*$ algebra part needs to change

Not so much the spacetime part

Well, maybe a little

Some metrics can be generalized functions

@Slereah I need that translation...

It's 5 AM mate

I'll do that later

3:27 AM

arxiv.org/pdf/0707.4257.pdf

This paper totally has Colombeau algebras for nets of observables

Well

Briefly

BUT

Enough to go on

Do you know what a jet bundle is

"Jet manifold formalism provides the adequate mathematical formulation of classical field theory, called axiomatic classical field theory (henceforth ACFT) "

whaaat

No, I do not

Well

I kinda

but not in any detail

Tho for my purpose, I don't think I need to learn about jet bundles

3:54 AM

Maybe I could try to do a Borschers algebra with generalized functions

It's not particularly different

translate pls

Borschrers algebra is like

Free scalar field QFT algebra

Built from Schwarz functions

2/3rds of that paper are appendix

user54412

4:28 AM

I believe that the processing of matter through a black hole acceleration disk is a form of cold fusion not possible anywhere else. I believe Einstein cold fusion does work but only in a black hole. As the matter nears the event horizon the extreme magnetic fields pull any radiation from matter and that is what causes Hawking radiation which is the last bit of radiation a proton produces during rapid decay. — Jen 3 hours ago

@ChrisWhite Yes that user posts interesting things.

@knzhou We don't discuss the details of suspensions. I will say that if you're not getting votes invalidated by the automatic script, you probably don't have to worry about your votes counting as serial voting.

'sup DZ?

Astrophysics still workin' like it should?

user54412

despite some people's dedication, astrophysics is still working

@ChrisWhite Ahhhh, very good.

4:31 AM

I doubt that

I got the funniest referee report ever today.

What's a referee report?

After solving an almost decade-old problem in our field, the referee rejected our paper because the result was "too obvious".

@BernardMeurer Feedback on a submitted paper.

Ah, lol, maybe he was salty

user54412

I would be tempted to make some snarky reply, but that's probably not wise

4:33 AM

@DanielSank lol... I would think that would be easy to refute

@DanielSank What do you do once it's rejected?

@BernardMeurer Depends on the journal. In this case I'm going to:

1. Rewrite the introduction to make the paper sexier (because referee #1 said we didn't show broad appeal)
2. Explain to the editor that referee #2 (the one who said it was obvious) is a moron.

@DanielSank Can you verbatim call him a moron?

@DavidZ I actually think that it's because we explained the physics really well that this referee thought it was "obvious".

user54412

@BernardMeurer If only...

4:36 AM

@ChrisWhite :p

Frankly, if after reading a paper the main idea is not obvious, the paper probably isn't so great :P

@ChrisWhite I'm not sure what would happen if we actually wrote that.

Wait what

My strategy is to always be positive.

If after reading it it's obvious it isn't good?

Therefore, I will say something like "We're glad referee #2 thought our result is obvious as this strongly indicates we have explained the new effect clearly".

4:37 AM

AH

@BernardMeurer See edit.

Okay

I was like "Wtf Dan y'all were trying to be oblivious all along"

I will also say that the fact that the problem existed for almost ten years, and the fact that the father of the field publicly declared it "the skeleton in the closet of our field" strongly indicates that it is not "obvious".

I'm thinking of actually quoting that guy in the main text of the paper.

I think that would cover the "not sexy enough" comment from referee #1.

Annex a picture of the wedding cake

that'll do em

@BernardMeurer zomg that's hilarious.

For those wondering what that means: my mother made my wedding cake and adorned it with various physics references, including plots from that paper.

4:40 AM

You totally should man, they'll put it in the cover of the journal

@BernardMeurer :P

Hey, @BernardMeurer quit chatting here and go fix cappy issues :P

Imma watch TV meanwhile.

It's 1:43 AM, do you really want me to code on our precious project? :p

zomg or go read about type classes!

Oh uhhh, nah. Go to sleep.

I'm going to be writing "foo+bar/1*1" all over the place

PSA: Anyone here who runs code on the JVM and in particular anyone who distributes libraries, read about type classes.

This is one of the neatest programming patterns I've ever seen.

4:44 AM

I gotta read that, good call!

It's SO cool.

Stack Overflow cool?

gotcha

You can essentially write libraries which allow extension to types you don't know about yet.

Without other people writing adapters!!!!!

::jaw hits floor::

::brain asplode::

You're way too excited about this lol

2

no

Dude, just read the article.

I'll help you with any bits of Scala you don't understand.

4:47 AM

def median(xs: Vector[Double]): Double = xs(xs.size / 2)

it's a function median that takes an argument xs of type Vector[Double] and... wait what

It takes xs, calculates Double and returns Vector[Double]

!

No?

omg

no, nevermind

Yeah, don't get it

5:07 AM

@DanielSank QM interpretation/ the measurement problem was called a "skeleton in the closet" by Jauch as quoted by this book quantumenigma.com

> “The interpretation [of quantum mechanics] has remained a source of conflict from its inception… For many thoughtful physicists, it has remained a kind of "skeleton in the closet."” —Jauch

5:26 AM

@peterh I can only speak for myself, though I suspect I'm reasonably typical, when I say I'm very proud of this site and I want to keep I a vibrant, useful and interesting place. My main (though possibly not the only) motive is that I want everyone to enjoy physics as much as I do.

5 hours later…

10:19 AM

> I would like to take this opportunity to briefly apologize for my past insurrections and disrespect for your lovely site. it is with a great grievance that i reflect on my past mistakes and i will learn from this experience and never repeat them again. Once more I sincerely apologize for this.

> ---- Albert Einstein, 2016-08-02

1 hour later…

11:32 AM

In a sense is the spectral theorem in operator theory, the result which proves that the eigenvectors of a self-adjoint operator are an orthonormal basis?

@JohnDoe yes

@ACuriousMind That's great...They don't mention that in two introductory QM books that I am using, they take it as an axiom, probably trying to protect people that are new to the subject from moving into areas which require too much mathematical motivation.

Could I ask a question about Gauss' law here? I posted a question but for some reason it's been closed. Not sure if this is the right place to go when that happens

@Hobbyist It's a gathering of the good and the great, sure someone will be able to help you...

@Hobbyist I suppose you could, but it's better to ask here how to fix your question to get it reopened.

11:42 AM

That would be better, yes. It's physics.stackexchange.com/questions/271207/…; not quite sure how to reword it

@Hobbyist Problems such as the one you posted are off-topic here as homework-like no matter whether they are actual homework or not. We want questions on this site to ask a specific conceptual question instead of how to solve some exercise.

Right, so the main idea I was getting at is that I don't want someone to post the answer, but I do want to know how Gauss' law applies here, since it's not so symmetrical as when you normally apply it

Or are these questions just blanket banned?

One quick thing to get out of the way is that you shouldn't put commentary on the question in the question itself. So that last edit you made after I closed the question, you shouldn't have done that. (You could have posted it as a comment.)

@Hobbyist You can ask something about how to apply Gauss's Law to an asymmetrical situation. That would be fine.

But you shouldn't ask for the solution to the problem you're working on, which is what you did.

Fair enough, wasn't aware of that

Thanks

Oh, and you shouldn't make an edit just to remove the last bit of the question. Wait until you have accumulated all the changes you want to make to the question and do them all in one edit.

Don't worry about it now, but just for future reference, it's best to minimize the total number of edits you make. If you find yourself editing one post more than 3 or 4 times, it's probably too many.

11:47 AM

By the way, is [on hold] like my cue to delete the question or to reword it and re-title? It's just kind of sitting there and unable to be answered as it is

@JohnDoe it also tells that this is not always true ;-P

Ideally you should edit it. If you edit it to address the problems that got it closed in the first place (and other problems that would get it closed, if that applies), it gets reopened. That's the goal.

Thanks

Oh, BTW "closed" is a synonym for "on hold". They used to call it "closed" until they changed the wording to make it clear that it's supposed to be a temporary condition.

That's why you'll see us saying "close" and "reopen" even though the question says "on hold".

Anyway, the point is, it's usually meant as a way to improve questions. Close (or hold), edit, reopen, that's the cycle.

@yuggib Okay thanks. I'm revising some operator theory to try to understand how to prove this result.

11:50 AM

However it's up to you and if you really want to delete the question, you can do that. I wouldn't recommend it, since I think you could turn this into a good and on-topic question which would get reopened, but it's your choice.

You can also leave it and do nothing. Again, I wouldn't recommend it, but you can do it.

1 hour later…

1:18 PM

Anyone home?

@0celo7 ossu~

1:36 PM

what

Hi :)

2:01 PM

@yuggib !

2:41 PM

19

A: Safe distance from a black hole?

A black hole does not have any magic properties, it does not "radiate" time dilation or any other nonsense like that. Noticable time dilation happens when one observer moves at relativistic speeds with reference to another, or when one observer is under much higher gravitational acceleration than...

This makes me sad :-(

@Slereah Dude, can I please have the translation

It's two lines

The beings of hyperspace can be defined precisely as those of ordinary space, and if we can't represent them, we can conceive them and study them

What

that's the translation

I know

What does it mean

2:51 PM

He's basically saying that it's doable to do math in more dimensions than 3

ah

Can one define a Hilbert space on a non-archimedean field

What's a non-archimedian field

A field that admits infinites and infinitesimals

what is an infinitesimal

2:54 PM

I think that shouldn't be a problem?

Non-archimedean fields still have the field structure

@Slereah Of course you can

Dude, what's an infinitesimal

An infinitesimal $x$ is such that $\forall n \in \Bbb N, n \times x < 1$

what the hell

@Slereah You get trouble outside $\mathbb{R}$ and $\mathbb{C}$ because you don't get a norm from the inner product anymore, so you have to say what you mean by "completeness" in that case.

2:55 PM

That's literally insane

Hm

@ACuriousMind outta nowhere

But wait, why would there not be a norm

@ACuriousMind Read a load of analysis last night about ODE, was put to sleep.

@0celo7 Nobody expects the curious inquisition!

2:56 PM

I mean I guess the norm can be infinite, but it's still part of the field, right?

Although the proof of Picardy Lindyhop was cool, I guess

It's not much of a problem

@Slereah A norm by definition goes to $\mathbb{R}_{\geq 0}$.

If you're not over the reals or the complex numbers, you have no guarantee that $\langle x , x\rangle$ is real.

Well you can extend it to $\Bbb R^*$ without too much troubles, I guess?

> the complex dielectric function is defined by $\epsilon'-\mathrm i\epsilon''$

2:57 PM

@Slereah ...what's $\mathbb{R}^\ast$?

WHY A NEGATIVE

@ACuriousMind In my case it would be generalized numbers $\mathcal E$

@ACuriousMind You mean $\Bbb R_{\ge0}$, of course.

Well, $\mathcal K = \mathcal E_M / \mathcal N$

Does anyone know what he's talking about

2:58 PM

Though I guess restricted to positives, as well

@Ocelo7 Who ß

?

@acuriousmind Our chief weapon is surprise! .. surprise and fear..our two weapons are surprise and fear.. and our ruthless efficiency. Our 3 weapons are surprise, fear, and ruthless efficiency... and an almost fanatical devotion to the pope

@Slereah The point is that the norm provides the notion of convergence by inducing a metric $d(x,y) = \lvert\lvert x-y\rvert\rvert$. You can't replace the target space of a metric without having to redevelop the theory of metric spaces.

The rambling one @PhysicsGuy

@ACuriousMind : Shouldn't be too difficult, though, no?

2:59 PM

@Slereah I recommend Jost's analysis book

He has a chapter on metric spaces

Go check it out

@Slereah I expect it to be very difficult, and I am not sure if it makes sense at all to have a metric that's not real-valued.

The convergence will just be when the metric becomes $\asymp 0$

What is that symbol o.o

I am not familiar with that kind of notation, so I dont care

For generalized numbers it means $= 0$ up to an infinitesimal

3:00 PM

@ACuriousMind Do you know anything about finite-time blowup of ODEs? Just the very basics ofc

I'm looking at $f'=\phi(f)$, where $\phi(y)\ge y^\alpha,\alpha>0$ with $f(a)>c$

@Slereah You often make crucial use of the triangle inequality for the metric. If < behaves differently in your field than in the reals, you have to reexamine everything, and I suspect it won't go through without major modifications.

Hm

@0celo7 I have no idea what that is.

I should try to work it out, I guess

@ACuriousMind Ok then

3:01 PM

sciencedirect.com/science/article/pii/S0019357708800237

@yuggib ~~~~~~

Oh wait

Apparently it's been done before

Let's investigate

projecteuclid.org/euclid.bbms/1197908895

also

do you have access to this stuff?

it's free access

communists

3:04 PM

@Ocelo7 Dont be radical

@ACuriousMind Let $M$ be a manifold and let $K\subset M$ be closed. Let $f:M\to\Bbb R^n$ be smooth on $K$ and cont. on $M$. Then there is a $C^\infty$ function $g:M\to \Bbb R^n$ that agrees with $f$ on $K$ and $||f(p)-g(p)||<\epsilon$ $\forall p\in M$.

Using this, I FINALLY showed that one can smooth kinks in curves

@Slereah I'm thinking one can prove transitivity of causal relations using this theorem, too

Man who cares about GR

GR is for losers

QFT is where it's at

@Slereah Read the second page of your first paper. The Hilbertian spaces studied there are used because the usual notion of Hilbert space doesn't exist in the non-Archimedean case.

Well all I want is to do QFT with it

Doesn't matter too much if the notions don't exactly coincide

As long as it serves its purpose

Which hopefully it do

Well, but if they don't coincide, you have to build things like operator theory from the ground up!

No spectral theorem for you

3:07 PM

We'll see how things go

No theorems about extensions of operators, essential self-adjointness, nothing

Hopefully I'll get distracted by something else in the meanwhile

I'm pretty sure that most of the important theorems will be the same

Same type of proof

And a few will not but

Physical states will be those of norm $1$ and I'm pretty sure that they will roughly be the same for those

Also hopefully there's a few papers out there on the topic already

I found a few, we'll see what's in them

@Slereah The notion of "Hilbertian" is grossly weaker than that of being a Hilbert space, see, for instance, statements like "Every subspace of a Hilbertian space is Hilbertian". I don't know where your confidence comes from that this is a rather trivial modification

I HAVE FAITH

Gee how do you even do anything if you don't try to solve problems

Just look into it and see what happens man

If it don't work out it don't work out

books.google.fr/…

Apparently there's a whole book about it

Let's obtain it

Does defining a Hilbert space over a non-Archimedean field sound like a decent idea to define linear operators as generalized function valued, by the way?

Or am I losing my way on the path to Jesus

i think ur already off that path

something tells me jesus didn't care about QFT and preferred you to work on CTC/time-travel theories and weird general relativity effects @slereah

3:22 PM

@Slereah What do you mean by a linear operator being "generalized-function valued"?

CTCs are a harder sell to find a thesis

A linear operator is valued in the vector space it acts on.

@ACuriousMind Well same thing as QFT operators being operator valued distributions, but for generalized functions

Such that $\hat{\mathcal{H}}_\varepsilon | 0 \rangle = \mathcal{E}_\varepsilon | 0 \rangle$ is a well defined Hilbert vector

@DanielSank So I bought a bag of these really small red peppers.

Turns out they're insanely spicy---and I've got like 50 of them. They will probably go bad in a few weeks. Any ideas on what to do?

I thought about making jam---my mother used to do that, but I'm not sure if that's maybe too much effort.

@Danu dry them?

3:25 PM

Dude, the new MathJax option is great.

@Slereah you need to tell the \hat what it's supposed to be on.

Fuck what did I fuck up in that latex

This SE chat mathjax extension is great.

If only it worked when just viewing the room without joining

There we go

$\mathcal E_\varepsilon$ is a well defined quantity in generalized numbers $\mathcal K$ but divergent on $\Bbb R$

0

Q: Number of close votes

I just reached 3000 rep (yay!) yesterday and I immediately went to the close vote review queue. After working through twenty it said I was out of close votes. Same thing today. Now, I remember reading somewhere (not sure where) that you gain close votes with rep, or review tasks done, or somethin...

discussion vote-to-close

3:28 PM

@Slereah What are those "generalized numbers"? Hyperreals?

Then the renormalization is just a proper choice of bare parameters in $\mathcal K$

@ACuriousMind Basically like generalized functions but, y'know

Not functions

Renormalization via hyperreals? Skepticism mode on :P

@Slereah In my world a "generalized function" is another name for "distribution", so that doesn't help me any.

^

Generalized functions are more general than distributions

user image

^the thing

$\mathcal N$ is the set of negligible numbers ($\to 0$ with $\varepsilon$), basically infinitesimals

$\mathcal E_M$ is moderate numbers, which is "reals" and infinite numbers

@Danu There's a few papers on renormalization with generalized numbers, if you wish

It's nothing special, really

Just another renormalization scheme

Hm, I think there's an error in that definition

Negligible numbers should be $\forall q \in \Bbb N, q > 0$

p-adic numbers aren't archimedean?

3:52 PM

@ACuriousMind Does everyone still hate you?

@BernardMeurer I think the wave of hate has passed :P

@Danu Hmm?

What MathJax

@JohnRennie Polishing a cerium ingot by hand

Hm

The sparks are real holy shit

Non-archimedean fields aren't metric

BUT

3:57 PM

I might need sunglasses

They are ULTRAMETRIC

Much better than a regular metric

Yes.

"Better" is arguable. But it certainly has more structure than just a metric space, of course.

Wait, I'm not sure how it works

$|x + y| \leq \max (|x|, |y|)$

But if I pick, say, $1 + 1$

$|1 + 1|\leq \max (1,1) = 1$

Obviously not the case

« first day (2098 days earlier) ← previous day next day → last day (2832 days later) »

Transcript for

Aug '162

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