The h Bar

barrycarter

17:00

@ACuriousMind I agree. I think we need a fundamental source of random numbers in order to generate other random numbers. The question is: given a source of random numbers on [0,1], can we output random numbers or R with equal probability, and the answer to that is no, unless we do some really serious weirdness thingies.

0celo7

@ACuriousMind What?

If your theory disagrees with obvious reality your theory must be wrong.

barrycarter

@0celo7 You mock @ACuriousMind ... while I'd normally be with you on that, the mock quality is a bit low.

ACuriousMind

@0celo7 This is not about "obvious reality", it's about the existence of an integral.

barrycarter

@0celo7 But his theory does not disagree with reality.

We could settle this by formally defining what an integral from -Infinity to Infinity actually means, but I'm too lazy, so someone else should do this.

It's easy to define if one side is finite.

@0celo7 OK, what do you claim the integral of cos(x) is?

Slereah

Cos is an even function you numbskull

barrycarter

17:05

@Slereah LOL :)

Slereah

It's easy as pie

Secret

(I am not sure if the rules of quantum mechanics (e.g. requiring probability to be conserved will poke holes in the following (possibly hypothetical?) scenario (marked by *) that aims to better illustrate the example)

In ordinary quantum measurements, e.g. check if an electron is spin up or down, you get a bunch of outcomes (a sequences of ups and downs) when you measure an ensemble of them. Quantum mechanics then told you the probability of giving spin up and downs via the $|\psi|^2$, but due to the random nature of decoherence, you cannot predict in advance which eigenstate will the stat…

0celo7

@barrycarter Undefined.

@barrycarter Not mocking

barrycarter

@Secret You can do what @DavidZ says, but I'm not sure you can do what I say. Also, if you put "@" immediately before someone's name (like I'm doing in this sentence), they get a little clicky sound to alert them you're talking to them.

@0celo7 So, you say the integral of sin(x) is 0, but the integral of cos(x) is undefined?

ACuriousMind

Heh, I guess coordinate transformations are out of the question, then :P

barrycarter

17:07

I've got him on the ropes!

0celo7

@barrycarter Correct.

@ACuriousMind For infinite integrals, sure.

barrycarter

@0celo7 But sin(x) and cos(x) are the same function up to a coordinate shift.

ACuriousMind

@0celo7 Why "sure"? What about this thing is a integral if none of the usual integral rules work?

0celo7

@ACuriousMind It's the area.

Secret

@barrycarter I know how annoying constant clicks are, thus I don't do that as often. If I know the conversation will be ongoing, I usually don't @. Sometimes I do "@ Name" so that I can refer to the person without actually sending out the click

barrycarter

17:08

Today on @0celo7's Wonderful World of Math: everything is integrable and most things integrate to 0.

@Secret Feel free to @ me whenever you'd like. If I'm busy, I'll just ignore it.

ACuriousMind

@0celo7 But cos is just sin shifted! How can they have different areas?

barrycarter

@0

@0celo7 You claim to be a mathematician, right?

ACuriousMind

(Also, don't make me dig up examples of integrable functions that have no graph for which you could conceivably speak about the area :P )

barrycarter

@ACuriousMind the indicator function of the reals?

ACuriousMind

You mean the rationals, right?

0celo7

17:10

@ACuriousMind Simple.

It depends on which point you measure the area about.

barrycarter

@ACuriousMind LOL :) I meant the irrationals, but yes.

@0celo7 Could you formally define an integral for us, Mr Ocelot?

ACuriousMind

@0celo7 Just checking: You still claim to be serious?

Secret

@Acuriousmind We know the outcome of a quantum measurement is random. But from the statement about random process mentioned below by Barrycarter, what is the random number generator in the quantum system, or is there one?

> In order to choose numbers from a distribution, you must already have a method of choosing random numbers

Or is quantum randomness (the outcome of a decoherence) is outside the scope of this because the random process outline in barry's quote is classical?

barrycarter

@Secret The output is random, but you can predict what happens if you measure it long-term.

0celo7

@ACuriousMind Well, this isn't Riemann integration.

Secret

17:13

@barrycarter ok that makes sense. In that case I am guessing the random number generator is the quantum state (because the pdf can be obtained from the $|\psi|^2$?

barrycarter

@0celo7 Oooh, what kind of integration is it?

0celo7

@barrycarter Physical integration.

barrycarter

@0

0celo7

Empirical integration.

barrycarter

@0celo7 Sounds kind of kinky....

@Secret Something like that. Whatever it is, we can draw a PDF for it before running any experiments.

ACuriousMind

17:14

@Secret The "you" that Barry spoke about is you, a human, who wants to draw from some probability distribution. When we model something by a probability distribution, it is supposed to just be random, you don't have any model for "how" it is random (then it would not be random)

barrycarter

@Secret What @ACuriousMind said: in order to pick a random number from a probability distribution, you first need a random number between [0,1] and then you pick where your prob distributions CDF hits that number.

@0celo7 This must be some strange new form of integration with which I am not familiar.

ACuriousMind

@0celo7 That's ridiculous, and you know it. Do you realize I actually tried to explain why integration can't work like that? And that I explicitly asked you if you were serious?

barrycarter

I think @0celo7 is covering up his error by pretending he was being silly all along.

0celo7

@barrycarter Perhaps.

Secret

@Acuriousind makes sense
If I understood correctly, in typical quantum computations e.g. "P(electron is at x=a)" is the random number, the $|\psi|^2$ give the PDF and then $$\langle\psi|x|\psi\rangle$$ is where the CDF has that number?

barrycarter

17:19

@Secret You can use the TAB key after you start typing someone's name to auto-complete (except for Ocelot)

0celo7

I still maintain that the area under $x$ is 0.

I will concede $\sin x$.

DanielSank

@ACuriousMind Yeah that's sort of the conclusion I came to in my half-consciousness falling asleep last night. Makes sense. Thank you.

barrycarter

@0celo7 You believe the infinite integral of the identity function is 0?

DanielSank

My research group's mailing lest is full of discussion of this issue now.

0celo7

@barrycarter What?

$f(x)=x$ has zero area.

That's clear.

3075

17:22

how come peskin and schroeder never mention the fock space formalism

0celo7

It's hard.

ACuriousMind

@Secret What? The random number is $x$, the pdf is $\lvert\psi(x)\rvert^2$ and $\langle\psi\vert x\vert\psi\rangle = \int x\lvert\psi(x)\rvert^2$ is the mean value of $x$.

barrycarter

@0celo7 Now I now you're joking :)

ACuriousMind

@DanielSank Alright :)

3075

@0celo7 are we referring to different things because I found it really simple back when I learned it in my class.

barrycarter

17:24

@0celo7 Any chance you could alert us to when you're pretending to be dumb vs just being yourself? It's a little hard to tell.

0celo7

What?

@barrycarter I'm being serious

The area under $f(x)=x$ should be 0!

barrycarter

@0celo7 Oh dear.

3075

@0celo7 why?

barrycarter

@0celo7 You can't use symmetry that way.

0celo7

@3075 It cancels out

ACuriousMind

17:25

The integral is not "the area" (see the integral of the irrational indicator function).

barrycarter

@0celo7 that's not how you define a -Inf to +Inf integral.

ACuriousMind

The integral is a mathematical tool that often corresponds to the area.

0celo7

@ACuriousMind Ok, and I take issue with that.

barrycarter

@0celo7 @ACuriousMind I agree that the integral from -L to L is 0. But when you take a -Inf,+Inf integral, you do NOT take that L -> inf

ACuriousMind

That's like taking issue with the definition of a vector space. Sure, you can do it, but what's the point?

3075

17:26

@ACuriousMind what is it really I forgot.

barrycarter

@3075 The integral is a formally defined limit of sums.

0celo7

@ACuriousMind My emprical integral was not a joke >:(

It's precisely the area, in all cases.

ACuriousMind

@3075 Well, it's some kind of limit. Which exactly depends on whether you're doing Riemann, Lebesgue or something else.

barrycarter

Or "empirical Ocelot integration", new and improved!

"I'll integrate ANY function for just $99.99!"

3075

@ACuriousMind can it be interpreted as something other than area though?

I didn't find anything on irrational indicator function.

0celo7

17:28

@barrycarter No.

barrycarter

@3075 If you interpret the function as velocity, you could interpret the integral as distance, for example.

ACuriousMind

@3075 Well, it's the antiderivative.

0celo7

Any empirical function.

barrycarter

What's an empirical function?

3075

@ACuriousMind I know that, but anything else that is not obvious?

Secret

17:29

ah yes, my mistake

Going further, what are the analogous terms in probability theory for the following commonly encountered expressions in quantum mechanics

1. Some generic (possibly hermitian) operator $\hat{A}$
2. Some generic state $\lvert \psi\rangle$
3. superposition of states $\lvert\psi_1\rangle+\lvert\psi_2\rangle$
4. Composite system $\lvert\psi_1\rangle \otimes \lvert\psi_2\rangle$
5. Entanglement $\sum_{i,j}\lvert\psi_i\rangle \otimes \lvert\psi_j\rangle$
6. $tr(\rho_A)$
?

3075

wait then what about this: en.wikipedia.org/wiki/Differintegral

0celo7

@barrycarter "Reasonable" ones.

Working on a precise definition.

barrycarter

@0celo7 Good luck with that :)

ACuriousMind

@3075 The function $f(x) = \begin{cases} 1 & x\notin\mathbb{Q} \\ 0 & x\in\mathbb{Q}\end{cases}$ has no "area", it is discontinuous everywhere, yet it is perfectly integrable.

0celo7

^not very reasonable.

3075

17:30

Cool though it isn't obvious to me why, any link for a description?

barrycarter

@0celo7 I think we're taking issue with your believing that $x$ integrates to 0.

You can confine the rational numbers into arbitrarily small intervals.

ACuriousMind

@Secret They do not have corresponding things. Quantum mechanics is not purely classical probability theory. See e.g. this answer

barrycarter

I'm pretty sure the differintegral is a meta-definition once you know differentials and integrals. It's cool, but not basic.

3075

guys I feel depressed and completely devastated because after not doing physics for 5 months I forgot a lot of what I knew and I don't know what to do.

barrycarter

@3075 @0celo7 may have a solution for you.

3075

17:34

I want to re-begin qft but I realized I forgot a lot of mech and em and qm.

and I forgot a lot of qft too so I have to begin from the beginning.

John Rennie

@0celo7: The integral of x from -L to +L is zero, but the distances from +L to +\infty and from -L to -\infty cannot be compared so we can't say they are the same. So you can't argue the symmetry persists all the way out to infinity in both directions.

barrycarter

All hail @JohnRennie

ACuriousMind

@JohnRennie We've been trying to tell him that for quite a while now :P

barrycarter

We really just need the definition, and it will become clear.

3075

@barrycarter idk if you will believe this but 0celo7 was the reason I began self learning physics 1 year ago. xD

John Rennie

17:35

I've been resisting the temptation for comment for a while, but I thought I'd found a good way to explain the point.

3075

and the funny thing is I'm basically back to where I was a year ago.

barrycarter

@3075 You know him? You have my sympathy.

@3075 You wouldn't happen to be toothless and from Alabama, would you?

3075

@barrycarter no.

I'm canadian.

barrycarter

@JohnRennie It's an interesting interpretation of the -Inf,+Inf integral (as the limit of the -L,+L integral), just not a correct one.

@3075 Once again, you have my sympathy.

3075

why?

barrycarter

17:37

Well you know, Canada.

John Rennie

@3075 If/when you go back to QFT I'll bet you find it comes back to you a lot quicker than you originally learned it. You'll have remembered more than you think.

@barrycarter that was my point, wasn't it?

3075

@JohnRennie hopefully.

barrycarter

@JohnRennie Yes. I'm not sure why I typed that... I actually mentioned that earlier in chat I think.

John Rennie

@3075 Barry has an almost British sense of humour - he is best not taken seriously :-)

barrycarter

@0celo7 is either teasing us or in hard core dummy mode :)

John Rennie

17:38

I mean that as a compliment of course.

0celo7

@barrycarter What?

barrycarter

@0

0celo7

@JohnRennie Did you read my closed geodesic proof

barrycarter

@0celo7 Dude, the infinite integral is not the symmetric integral.

0celo7

It's very slick

barrycarter

17:39

@JohnRennie Of course.

0celo7

@barrycarter I think it should be.

Slereah

Ugh integrals

QFT is really an integral fest

barrycarter

@0celo7 And you believe it should be filled with unicorns and puppy dogs?

@0celo7 There's a poster with a picture of Einstein that says... you must be THIS smart to create your own system of mathematics ;)

Slereah

Einstein didn't do that

0celo7

Einstein was dumb.

John Rennie

17:40

@0celo7 but you can't add real numbers to infinity, so how can you extend the symmetric integral to the \pm infinite limits?

0celo7

He shat all over the place and failed the driving test a bunch of times

@JohnRennie What?

You take a limit.

Secret

The indicator function cannot be drawn
A possible proof might be as follows:

There exists no map $M : (I don't know what is the discontinuous version of \mathcal{C}(\mathbb{R})\rigtharrow \mathbb{R}^2$ such that M(I_A) is injective

Pretty sure I made an error in my logic somewhere, to be checked...

barrycarter

@Secret Wait, you're saying you can't map the complex numbers to R^2 injectively?

0celo7

@barrycarter You can map them to R injectively too

Slereah

What about $(x,y) \rightarrow x+iy$

0celo7

17:44

Bijectively even

Secret

ya, make sense

barrycarter

@0celo7 Yes, R^n isomorphic to R. I was just complaining about @Secret's specific case.

Slereah

I'm pretty sure you could even map them bijectively to $\Bbb R$ too

Since they have the same cardinality

barrycarter

@Secret Were you talking about the Riemann mapping theorem... that you can't do it preserving angles and distance?

ACuriousMind

I think he didn't mean the complex numbers :P

Although I have no clue what he meant.

Secret

17:46

Well, I am not sure if there is a mathematical way to express the fact that "attempting to plot the indicator function of the rationals on a piece of paper will basically end up with something that look like two horizontal lines no matter how far you zoom in the axes, yet these lines are not continous because they are bascally a set of dots"

barrycarter

Everywhere discontinuous?

Secret

yes

Slereah

You can't draw it because paper is not continuous, for a start

barrycarter

That was my answer :P

@Slereah What?

ACuriousMind

This chat makes even less sense than usual today...

Slereah

17:47

Let's talk about cakes instead

barrycarter

Yeah, we now have 3 people who don't know what they're talking about.

Up from the mean of 2.

Secret

Basically, the above statement is one of my attempt in me hitting one of my habits on its head by trying to prove that some given mathematical object simply cannot be illustrated so I can stop wasting time trying to visualising it (because visual learnign is such a strong tendency of me)

This is in response to too many complains about me gietting too wild on visualisation to the point it does not make sense anymore

Slereah

maybe just stop without trying to prove it

It's just not particularly useful to try to illustrate EVERYTHING

barrycarter

@Secret This actually relates to my Theory of Physicists, which would offend here, so I won't say anything more.

Secret

you knew I have stopped that for a very long time, right?

Slereah

17:50

Keep on doing it then

barrycarter

Mathematics is beyond visualization.

Slereah

the last thing we need is more Penrose notation

barrycarter

It would be like looking upon the face of God.

Truth is math, and math is truth. That is all you need to know.

0celo7

@ACuriousMind are you mad at me

John Rennie

But doing SR calculations is a lot easier if you draw a spacetime diagram ...

barrycarter

17:52

@JohnRennie Nooooooooo!

Secret

@JohnRennie A LOT EASIER, in fact

barrycarter

Oh come on.

0celo7

@JohnRennie If you have to draw a picture you're obviously not good at math.

barrycarter

It's math, it's all math. Diagrams are a crutch. You don't need them. You can walk. YOU CAN WALK!

John Rennie

Guilty as charged

0celo7

17:52

If you can't do the calculations axiomatically, well, you don't understand the theory.

barrycarter

I'm tempted to star that, but I'm currently feuding with the Ocelot

0celo7

Pictures only dilute the purity of the math.

barrycarter

Stop making me love you.

Secret

@0celo7 I am not very good at maths indeed, although for unkwnon reason people tend to have an impression I am good at it

I can handle maths purely abstractly, by treating them as some rules between some objects

barrycarter

@0celo7 Doesn't that contradict your own statement about "empirically integrable"?

@Secret That's what math is. A set of rules. THE RULES. Way better than those religious rules.

Secret

17:54

Mathematics is a study of patterns

barrycarter

@Secret That's not true, though I've heard people say it.

Secret

what is missing?

0celo7

Sigh, @ACuriousMind are you ignoring me

I was not trolling

barrycarter

@Secret It's entirely wrong. It's proving whether patterns actually exist or are just observed but incorrect.

Secret

i see

barrycarter

17:55

"3, 5, 7 are prime, thus all odd numbers are primes" <- pattern

There are much worse cases.

Secret

Acuriousmind probably gone BSOD due to 3 guys having undertermined and incomprehensible mental states

And I must say this is the first ever joke I made about people in this chat

barrycarter

@Secret How did you know he was a bot?

@Secret We all use humor to hide our low-level hatred of each other.

0celo7

@barrycarter why do you hate me

barrycarter

@0celo7 Do you remember the starred comment? A better question would be "why would anyone NOT hate you"?

Secret

People can go BSOD, we are ensemble of some extreme advanced quanutm computers after all

I rarely use humor, because it is mostly irrelevant to the seeking and discussion of knowledge. As a result when peopel are mad at me, it is often true that they are REALLY made at me

andI have no reason to hate anyone here, btw

barrycarter

17:58

@Secret Might try a spell chequre there too ;)

@Secret Yet.

0celo7

@barrycarter :(

barrycarter

@0celo7 You have two choices. One is to improve yourself.

Secret

My questions told me one thing: I don't know what I am talking about despite I knew exactly what I am trying to talk about

It's as if my Self is decoupled from my conscious thinking

barrycarter

@Secret The ability to translate your inner mind into speech isn't easy.

skill patrol

@barrycarter do you think hate is good?

Secret

18:05

One professor said I have too many ideas, but no coherent way in expressing it

Even my thesis suffer from a highly nonlinear introduction

barrycarter

@skillpatrol I was joking, but, at a fundamental level, yes.

Secret

and too much redundancies

barrycarter

@Secret That could be a compliment, though. Ideas are good. Improve on expressing them and you might really have something.

skill patrol

on what fundamental level @barrycarter?

barrycarter

@skillpatrol The fundamental level of judging people who want to harm those who don't harm others. IE, the intolerant, the hateful, the physics moderators.

skill patrol

18:09

power has a way of doing that to people

barrycarter

Power corrupts, absolute power ... is kind of cool.

skill patrol

absolutely

barrycarter

Yes, we need a new election for physics.SE, and the elected moderators need to take more responsibility than the high-reputation members.

ACuriousMind

...what?

barrycarter

I'm unhappy with the overly aggressive moderation on the physics.SE site.

ACuriousMind

18:11

@0celo7 No, I am just not instantly responding :P

John Rennie

@barrycarter you are in the minority

barrycarter

@JohnRennie Of what group?

John Rennie

@barrycarter those who think the moderation is excessively aggressive

And on that note I have to go. I'm off drinking with some programmer friends.

ACuriousMind

@barrycarter "Overly aggressive"? I would not use the word "aggressive" at all to describe moderation around here.

skill patrol

cya later

barrycarter

18:12

@JohnRennie No, I meant, what's your population here? The majority of overly aggressive moderators believe in overly aggressive moderation?

John Rennie

If I suddenly reappear and start posting gibberish in about six hours time, that's the reason

barrycarter

@JohnRennie How is that different?

John Rennie

More gibberish than usual, that is

ACuriousMind

@JohnRennie I'll be waiting for that ;)

barrycarter

@ACuriousMind They seem to close a lot of questions quickly and unnecessarily, and with a zeal bordering on compulsive.

ACuriousMind

18:13

"they"...most questions aren't closed by moderators at all.

barrycarter

Yes, I was talking about high-reputation users who have moderator-like abilities.

ACuriousMind

@barrycarter In other words, people like me?

barrycarter

@ACuriousMind Nice weather we're having, huh?

@ACuriousMind I haven't checked your closures specifically, but yes, people analogous to you.

ACuriousMind

@barrycarter there's a thunderstorm outside :P

barrycarter

@ACuriousMind You better run, you better take cover.

"We have closed your question and not migrated it, because there is the theoretical chance that it could be useful to someone somewhere."

ACuriousMind

18:16

@barrycarter We cannot migrate questions to somewhere else than math.SE

barrycarter

@ACuriousMind Which is often the right choice, actually.

OK, I'm off for a bit, will return to debate this later, maybe.

ACuriousMind

All other migration targets need an actual moderator

skill patrol

cya

ACuriousMind

@barrycarter Yes, and there are quite a few questions migrated there. You'll have to be more specific about what the problem is.

AccidentalFourierTransform

18:28

whats going on with these electrostatic-related homework questions today?

sigh... 20 votes a day is far from enough...

0celo7

18:46

@ACuriousMind :(

@ACuriousMind I know the argument that locally Riemann flat implies locally isometric to Euclidean space, but if I know the topology of space is $\mathbb R^n$, how do I strengthen that to "globally"?

If I know it's Riemann flat at all points.

ACuriousMind

Why would I know that or care about it?

0celo7

Because you took a course on Riemannian geometry?

0celo7

19:03

@ACuriousMind Do the topological charges in QFT like $\int \operatorname{tr} F^2$ give information about the base space or the total space of the bundle?

ACuriousMind

@0celo7 Neither, sadly, except in the $\mathrm{U}(1)$ case, where that thing is actually an element in the second cohomology class of the base space. (Well, them being non-zero also means that the bundle is not trivial, but other than that, there is no directly evident topological information in there)

0celo7

@ACuriousMind Wait, so why call them topological charges?

I know that integral has significance wrt. the topology of 4-manifolds

ACuriousMind

Oh...well, they are a topological invariant of the bundle.

It's just that knowing them doesn't directly tell you anything specific about its topology

Maybe I misunderstood your question.

barrycarter

@ACuriousMind (resuming) My problem is that many questions that COULD be migrated to math are simply closed instead.

0celo7

@ACuriousMind What is $\mathrm{tr}F^2$? Is it a Chern class?

I've forgotten

ACuriousMind

19:10

@0celo7 Yes, it's the second Chern class.

0celo7

@ACuriousMind Ok, so suppose I have two gauge theories

and for one I get $\int_M c_2=a$ and for the other $\int_N c_2=b$, $a\neq b$.

From this, what can I say about $M$ and $N$?

Just that they're not homeomorphic?

@ACuriousMind The bundles have the same $G$, of course.

And $M$ and $N$ have the same dimensionality.

ACuriousMind

Wait, are $M$ and $N$ your bundles or your base manifolds?

0celo7

@ACuriousMind We integrate characteristic classes over manifolds.

ACuriousMind

The bundles are manifolds :P

0celo7

Sigh...pedant

The base space manifolds then

ACuriousMind

19:14

Then it tells you nothing about $M$ and $N$.

Since the same base manifold can have different bundles with different $\int F^2$ on it, they could even be the same.

0celo7

@ACuriousMind ...I thought Chern classes did not depend on the exact connection you picked.

ACuriousMind

I didn't say that they did.

But you can have the same base once with a trivial and once with a non-trivial bundle on it, the Chern classes of those bundles will be different.

0celo7

Then how could different bundles over the same manifold have different integrated Chern classes!?

@ACuriousMind Ah...

ACuriousMind

@0celo7 Because they are different bundles? I don't understand what the issue is. The Chern class is an invariant of the bundle, not of the base manifold.

0celo7

@ACuriousMind Yes, Jeez.

ACuriousMind

19:21

@barrycarter Are you talking about homework-like questions?

0celo7

@ACuriousMind Is there a general way to define Wick rotation on a curved manifold?

barrycarter

@ACuriousMind Yes, since I know Math.SE accepts those. Plus, the closures are often without providing a good link to other sources.

ACuriousMind

@0celo7 Not that I'd know

barrycarter

@ACuriousMind I also object to banning worked-out solutions, since those would be most helpful for physics students in understanding step by step how something works.

ACuriousMind

Well, object all you want, I don't think we're going to change that. Have you looked at math.SE? I find that site terrible precisely because it is flooded with low-level homework that is gladly answered in minutes by many of their users.

barrycarter

19:26

@ACuriousMind Yes, but that's actually a good reason to move questions there. It'll get an answer and the people in math.SE don't mind.

ACuriousMind

@barrycarter The overall rule for migrations is "Don't migrate crap". We don't migrate questions unless we are sure they are actually on-topic at the target site and are of good quality. And, many of the questions we close are just crap, just copy-pasted exercises with no thought behind them.

And they are physics questions, after all - those aren't really on-topic at math.SE, they even sometimes migrate physics HW to us that we then have to close.

barrycarter

@ACuriousMind All physics is math, but most of the questions asked are highly mathematical anyway (ie, high school physics). The problem is: when you define the word 'crap', do you mean as viewed from the migrating site or the migrated site? If math.SE is OK with them, I'd say they're not crap.

@ACuriousMind And, at the very least, can we give the poster SOME help in finding resources, instead of pointing them to a post which tells them why they are a bad person first?

@ACuriousMind I think we also need a canonical answer to "I've got this crackpot theory..."

ACuriousMind

@barrycarter The close reason is fixed, and we can't just change it on a whim. There is an on-going meta discussion about the homework policy, but any reformulation of that close reason will take time.

@barrycarter The canonical answer to that is "non-mainstream physics is off-topic on this site".

barrycarter

@ACuriousMind My point is that you can migrate instead of closing. And we could have a canonical answer that gives people some advice (eg, ask people you know, does your theory predict something, is it consistent, etc).

physics.stackexchange.com/questions/248857/… We sort of need a canonical "physics is an approximation to the real world, we don't actually model individual protons"

ACuriousMind

"Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics and astronomy." - We are not here to debunk crackpots, or give laymen advice. We are also not here to do high school student's homework, or tell them where they can get help. It's just not what this site is meant to do, why do you insist it should?

barrycarter

19:38

If it's a site for physics students, it should certainly help students with homework, no? And, remember, crackpots see themselves as researchers. Overall, I don't see the harm in having a more positive policy towards questions that don't fit.

ACuriousMind

@barrycarter The harm is that it drives away the experts. Quite a few of the high-rep users express frustration at the amount of homework and pop-sci and plain gibberish they have to wade through to find worthwhile questions. The point is that we simply don't want to answer homework questions, and we don't want to help crackpots with their "research". It's about the signal-to-noise ratio and closing questions is the way of filtering out noise.

barrycarter

@ACuriousMind Oh, for the crackpots, I wasn't suggesting keeping the questions open. Just close with a link to a single "how can I tell if my theory is any good" question. No need to debunk on a case by case basis, just something that provides a first level of help.

@ACuriousMind As for homework questions, either remove the tags for homework, put up a bigger warning, or edit the site description not to imply that students and welcome to ask questions.

ACuriousMind

@barrycarter Why? What is to be gained by that? Why should anyone spend the time and energy of writing something like that? And how would it fit into the Q&A system?

@barrycarter Students are welcome to ask questions. It's pretty sad if all questions you have as a student are homework questions.

0celo7

@ACuriousMind ...

Seriously?

barrycarter

@ACuriousMind Because it helps the overall universe of knowledge? Because it makes the world a better place... hell, I'd write it myself if you'd like.

0celo7

19:45

Not everything has crazy conceptual questions that pass the bar at PSE.

I certainly don't.

ACuriousMind

And I think a warning that pops up when you add the homework tag is under investigation, but I think that as to be enabled by an SE developer.

barrycarter

@ACuriousMind No, but you're arguing that students should have NO homework questions. Most of their questions will be homework related.

ACuriousMind

@barrycarter No, I'm not arguing that. They can have all the homework questions in the world. We just don't want those here.

barrycarter

Working out a sample problem (not necessarily with the exact numbers they have), explaining each step is uber-helpful.

ACuriousMind

Yes, it is. Still, we don't want to do that here.

barrycarter

19:47

@ACuriousMind Well, that's totally not clear from the site description and the existence of the homework tag. Also, isn't there some sort of overall philosophy behind stackexchange itself?

ACuriousMind

@0celo7 You're also not a student of physics, as you don't get tired to mention.

0celo7

@ACuriousMind Only physics students are allowed to post here?

barrycarter

It's not even clear from the comment replies. Some say "show us your work and we'll help" while others say "you can't ask that here, period"

Students of mathematics are students of all things.

ACuriousMind

@barrycarter That is because we have people who disagree with the current homework policy leaving comments like "show your work and we will help" which are how they wish it would work.

barrycarter

@ACuriousMind Oh, it should probably be made clear that those comments are contrary to site policy.

ACuriousMind

19:50

That's the fundamental issue I keep bringing up in the chat sessions - many of our users seem to not really care at all about policy.

barrycarter

@ACuriousMind OK, but you high-powered mods can delete those comments or at least add "no, we really fucking hate you here. we will NOT help you EVEN IF you show your work"

Clarify that the other comments are invalid.

ACuriousMind

@barrycarter No, I cannot delete any comment.

barrycarter

@ACuriousMind OK, so correct it. "@xyz is wrong. We won't help you. Period."

ACuriousMind

And every time I leave such clarifying comments, I get drawn into exhausting and repetitive comment discussions.

0celo7

How did Euclid actually define "straight line"?

barrycarter

19:52

@ACuriousMind OK, modify it a bit to say "I have moderator powers and I'm telling you it's against policy for us to help you; all the low level users who say otherwise are violating site policy, and we will round them up and kill them shortly"

I seriously thought this site helped people who showed their work. I didn't even realize that was counter-policy.

Obliv

@0celo7 was it one of his axioms of geometry?

Euclid described a line as "breadthless length" which "lies equally with respect to the points on itself"

0celo7

@Obliv I can google too

barrycarter

Although I have to say: this site's current policy seems to go out of its way to not help people and make them feel bad about themselves, after entrapping them into believing it will help them.

Obliv

lol

0celo7

That's not a definition

barrycarter

19:55

@0celo7 Two points.

0celo7

@barrycarter What

ACuriousMind

@barrycarter This is a template comment that many users have started to leave. So far, I'm not sure it makes any difference.

barrycarter

@ACuriousMind You're saying that template is contrary to policy?

ACuriousMind

@barrycarter Okay, I agree with the first statement, we don't help people who post homework questions. But what do you mean by "entrapping them into believing it will help them"?

@barrycarter No

Obliv

you can resolve a definition out of that probably. @0celo7 a line is one without width and every point within the line lies in a straight line to one another. what kind of definition do you want..?

barrycarter

19:56

@ACuriousMind You have a homework tag and offer help to physics students in the site description.

ACuriousMind

@barrycarter And the homework tag says: "Please READ THE GUIDANCE IN META before asking homework-like questions. "

barrycarter

@ACuriousMind Homework-like questions should ask about a specific physics concept and show some effort to work through the problem. -- doesn't that mean we WILL help them if they show effort?

ACuriousMind

@barrycarter No. It means we will help them if they ask about a specific physics concept and show some effort. Just showing some effort is not enough.

That's an "and", not an "or".

barrycarter

@ACuriousMind The (presumably site-approved) template almost literally says we will help you ..... holy crap

0celo7

@Obliv what

barrycarter

19:58

@ACuriousMind Who wrote this crap? A deranged hateful lawyer? I always read that to mean "show some effort"... every physics question is about a physics concept.

0celo7

a straight line is a line in which every point lies in a straight line?

are you serious?

Obliv

lol

barrycarter

@0celo7 Can't you define a line as two points?

Obliv

connected by a straight line

0celo7

@barrycarter No, it's an axiom that there's a line between two points

but what the hell is a line

Transcript for