hey I actually made a perfect score on a GR homework for once

@Semiclassical I have a working theory :P (it does not involve you being a dick)

lol

he grades them kind of brutal but says it doesn't really matter because it's a grad course ( \o/ )

is my not being a dick a necessary or sufficient condition of that theory? :P

I wish internal set theory wasn't so obscure because it's kind of cool :3

one of the big guys in nonstandard analysis emailed me and asked for my internal set theory thesis a couple weeks ago and has been emailing me comments

so far my favorite comment was when he emailed me saying I had a typo in my Idealization axiom and then emailed me the next day saying he was mistaken

my thesis contains the most significant worked portion of real analysis within internal set theory that I'm aware of by quite a bit

@GPhys It's obscure because non-standard analysis itself is kinda obscure, no? :P

@ACuriousMind This is true, but even among nonstandard analysis internal set theory is an obscure approach

So...there's one person using it? ;)

>.<

to within an order of magnitude :>

there are a number of sources out there for nonstandard analysis with hyperreals that work out the usual real analysis stuff but the stuff for internal set theory is all at the level where "development of this entire topic is left to the reader"

...lol. my "i'm perhaps being a dick" message from earlier just came up as spam/offensive flagged

as in, it popped up for me to give a vote on it since I'm a 10k user

(I didn't)

my math advisor was convinced there are some meaningful non-analysis uses for Internal Set Theory/NSA, and there are some interesting things

Invalid flag (and marked as such). Being concerned about whether one is being a bit of a dick oneself is not rude or offensive. (I nevertheless removed the messages from the starboard because the starboard doesn't really need dicks on it :P)

indeed.

for example you can write a finite field containing all the usual integers (but is still finite)

@Semiclassical That's odd, I think you shouldn't be allowed to vote on your own flagged message (that you see it is also sort-of a bug, but I think it's always been that way)

nobody ever asks interesting NSA questions on math.se

i didn't attempt to vote either way, so maybe it wouldn't have allowed me

but i did see it :)

I suppose limits are interesting but at the end of the day you will realize you've just redone some words to make things a little bit more directly provable without changing much

the more interesting things are continuity, or proofs of classic analysis results like the mean value theorem and the like

@GPhys I wondered for far too long why math.SE would entertain questions about a secret service until it clicked

whose proofs in NSA are clearly entirely separate from their standard analysis counterparts

@Semiclassical Ah, that's possible

chat UI is a mess

i imagine part of what's going on re: NSA is that it's not obvious why one would do the non-standard argument when the standard one works just as well.

whether or not that comparison is accurate, i can well believe that that's the prejudice

@Semiclassical got a 100 on the liquid test

You're 100% liquid? How do you sit in a chair and type?

the standard one of course works just as well lest it is flawed, but if you mean in the subjective ease-to-come-up-with sense then I'm not sure that's justified

to make an analogy to QM: Why bother to include the particle trajectories implied by the pilot wave interpretation if they don't make the theory any more predictive?

after all, the NSA arguments were thought of first by Euler et al

(even if they didn't know the rest of NSA to back it up, the arguments work virtually unmodified when you add the NSA language)

💦🌊💧

@GPhys Sure, that might well be true. But perception matters when it comes to MSE questions.

💯💯

I'm awful at emojis. That took way too long to find

@ACuriousMind same way you do I guess

with reference to modern times there is at least one result that was first published with a proof in NSA while a standard proof wasn't known. There are a number more where the NSA proof was found first and then a standard proof was worked out for journal submission

You're a brain in a jar

My brain is just dissolved a bit

a standard proof was found for the result using NSA fast enough that it was published in the same journal issue as the original NSA proof though

(to follow up on the analogy, one argument I've seen in favor of pilot-wave theory is that it makes it easier to understand scattering theory and stuff like time-of-flight measurements)

@S

Hmm. Was that for @Semiclassical or @Slereah

Perhaps it was for @Secret

is he here?

I'm not really sure I'd argue NSA should really be taught instead of standard real analysis (in fact I'd argue maybe it shouldn't?) but it's a simple enough tool for a professional mathematician to learn, I think

|S> = <Semi|S>|Semi> + <Slearah|S>|Slearah> + <Secret|S>|Secret>

?????

@Slereah I need to understand a weak Bianchi identity. Is there a review paper on weak metrics?

convincing outside of analysis might help with this

Are weak metrics $C^0$ metrics?

W^{2,q}

Then I do not know

@GPhys yeah

Or q,2

I never remember the order

that's about how i feel about pilot wave stuff, though the threshold of knowledge required for NSA is a good deal higher

@Semiclassical The important question is now what these coefficients are

yep

I'm not even sure if the curvature tensors make sense with that regularity

I need to write up a page on weighted Sobolev spaces. I can't remember all of this crap

@ACuriousMind Though I dunno how I'd interpret (|Semi>-|Slearah>)/sqrt(2)

phases, man, I dunno

Semiclassical without Sam's insane wormhole stuff

So if you have insane wormhole ideas, that's no more

instead, i just know pilot wave stuff which no one gives a damn about :P

@0celo7 There's still a chance to measure that state being |Slereah>, though

dangit, StreamPlot

why can't i make you plot more lines

@ACuriousMind I don't know any QM

you liar

of course you don't.

you just know functional analysis :P

Born's rule is physics

I'm not a physicist

> I knew the physics before I learned the math.

Always remember |Secret> = <weirdness> (abusing abstract algebra)

AND THEN I FORGOT IT

@0celo7 :'(

StreamPlot seems to take the StreamPoints argument as a polite suggestion

All too plausible with that sieve you call memory :P

@Semiclassical honestly I think it's overblown a little. When people see the most common (non-axiomatic non-internal set theory approach) to NSA they see all this scary model theory and are pushed away, or for the IST approach they see they have to learn axioms. I don't really think any of this is much harder than the construction of real numbers, and maybe even easier in the case of IST

You once forgot you had played 1000 hours of Battlefield

Funny story, on my QM 1 midterm people wrote down imaginary numbers for probabilities

aiii

Some people got answers greater than 1

although the construction of real numbers is not so trivial itself

I'm sure they were i% confident these were correct, too

stop violating unitarity people

I think 2 was a common mistake because of a normalization

I'm not entirely unsympathetic there; time pressures of an exam and all

that said, I'm firmly of the belief that if you get a nonsense answer on an exam you had better be able to recognize that it's nonsense

@ACuriousMind yeah there's something wrong with my mind

@Semiclassical I've noticed my numerical BS detector has gotten better over the years

in intro physics I tend to rely on units for my "is this answer nonsense" criterion

@ACuriousMind Why do you think I keep playing Skyrim? I just wait 6 months and forget everything.

though there are other considerations. for instance, on the last exam one of the problems was to get <p^2/2m> for the harmonic oscillator states

which really means "know how to use ladder operators"

a lot of people made little mistakes, like getting $\sqrt{n+1}\sqrt{n}$ when it should be $n+1$

but at least that answer has $n$-dependence.

by contrast, if your answer doesn't depend on $n$ then you're claiming that all the energy eigenstates have the same expected kinetic energy

and that's plainly nonsense.

@ACuriousMind @Semiclassical what's the best way to collect a lot of research where the technical details are important?

Do you make summaries in one big file?

Attach notes to the printed papers?

@Semiclassical when I was taking the master's/PhD calculus/analysis course set at my undergrad on the exams each problem was worth 20 points and you started with 5 points but lost them if you wrote something that was wrong

The idea I guess being in that there's value in knowing you don't have the answer (and knowing something is wrong)

right. there's an evaluation step

i'm more okay with someone doing physics derivations and coming to a nonsense answer if they can see that it doesn't make sense. (even better would be if they can use that to diagnose what's going wrong and fix it, but time pressures don't always allow that)

@Semiclassical we've had this discussion

(i have the qualifier of 'more okay' because it's possible the reason they got the wrong answer was some misstep in logic rather than just a algebra/arithmetic flub. the latter is not nearly as problematic as the former)

probably. i do tend to be a broken record on such things.

I wrote a note on my QM2 midterm that an expectation value HAD to be real but for the life of me I couldn't get a sensible result

Turns out I multiplied matrices wrong in the first line

oof

to me, that's an error which shouldn't result in a major penalty.

It didn't

good

Because after that error everything was correct and I knew there was an error

I was fooling enough to think I could multiply matrices however

@Semiclassical it's an approach that makes more sense in a proof course

there's a difference between leaving out details of a proof and making an argument that is wrong

Eh, I think applies in most long-form problems in physics. The point isn’t just to reply with an answer, but to show the work in deriving it from physical principles

@0celo7 I haven't found a good way, honestly

Somewhat disappointingly, a good memory seems to be the best tool.

Which is to say, what’s asked for is not so much the answer but to outline an argument for that answer

sure

on a previous point: I think the intuition NSA provides in standard point set topology is nice, akin to how nice continuity is

NSA?

oh, non standard analysis

Non-standard analysis is nice and all but there are many traps

Like there's 3 different definitions of continuity in it

And there are discontinuous things that you wouldn't expect

there are also things people say about NSA a lot that I'm not even sure the origin of

although I know some to be quirks of the hyperreal approach or the like

Like what?

well, almost all of these things don't really apply in internal set theory where none of your definitions change at all

I don't really like to use IST for hyperreals

It's a bit much

You can just do it with sequences

@0celo7 I'm struggling with this too

Know what's great about the hyperreals, though?

There's an embedding of distributions in smooth hyperreal functions

Me too!

How... wonderful

Why is that important?

@DanielSank So it turns out there is a path integral definition of quantum computing

which is neat

@Slereah k

Currently, there are around 400 papers (related to my thesis) that I have organized, read all of the abstracts and conclusions (know their idea and what they've done), and read some of them completely. But no (well organized) note so far :/

Not a good idea

@Maxwell Dude, don't try to distinguish those categories a-priori.

People at math.se are unlikely to know really physics-related math questions

@Maxwell no

They're unlikely to know about renormalization or causal structures or whatever

Dude, you can't just name an entire category "mathematical physics" and hope to decide whether or not it's on topic.

The category isn't well-defined enough.

Maybe you're just bad at it

^

Or maybe those particular questions were off-topic.

You can't always take your experience and generalize it to all of mathematical physics and all of this site!

Link us to your questions.

I'll comment/edit them as appropriate.

@Maxwell Well, that sounds less than useful.

It's important information though.

Ahhhhhhhhhhhhhhhhh

Ok I can explain that.

ME TOO

The so-called "homework policy" is NOT about homework.

The "homework policy" is intended to prevent questions that ask for others to do a calculation for you.

You have to show your effort and ask a specific question about the place that you're stuck.

Ok, well, if you don't link to the question then I really can't say any more about it.

I can talk with you to help you feel better, but if you want real help about your question, send me a link.

@Maxwell Lot's of people have ideas about it. We've talked about this many, many, many times.

Why don't you say it instead of hinting?

Why are both h bar and maths recently so ran over by weirdos

Insert deranged emotional state

Or perhaps I have become a lot more mentally unstable

whatever that means

@Maxwell you have neither linked your questions nor offered an opinion on the homework policy.

::Back to do some transcendental number theory::

@Maxwell They are not mathematical physics

If your question involves equations that doesn't make it mathematical physics

Will someone link the damned questions?

@Maxwell I see a few ways that question can be improved.

Before the Polyakov action, they used to choose in Nambu-Goto from physical considerations what you get from the EM tensor of Polyakov and get the same EOM from the crazy NG eom, absolutely incredible

@Maxwell what the hell is $Ii$?

@Maxwell It is very common for new users to come to the site and immediately think they have a better idea of how the site should work.

I would advise you to observe and listen for a while before expressing opinions on how to change a community.

@Maxwell No. If you think so, then you don't know what mathematical physics is. None of your questions were about mathematical physics - if you had read the tag description, you'd know that the occurence of math in a question is almost entirely unrelated to it being mathematical physics.

^

That's an unanswerable question.

It depends on what you're asking.

@Maxwell Do consider the possibility that he read the question and left that comment because it was appropriate to leave. Several users here use userscripts to insert certain comments they have to make rather often.

The question you linked above could be closed as homework on Physics.SE because you show no effort at all to solve it yourself.

You just ask "how do I find $F$" without showing any attempt to do it yourself.

@Maxwell Not if they are of the form "Please solve this exercise for me".

@Maxwell Listen to me. I read your question, edited it, and think it's not good.

Instead of assuming everyone's out to get you, consider that we're trying to help you and keep the site quality high.

You should read this famous essay.

@Maxwell Why are you so sure they didn't read the question before voting to close it?

@Maxwell I see, so you have a crystal ball and you know that other people didn't read your question. Cool!

I want a crystal ball too. Can I borrow it?

@Maxwell yes on the math.se you will more likely get an answer, and even multiple answers, than on the physics one, as you see you got two on that question and comments trying to help.

@Maxwell I didn't take you to mean me.

@Maxwell Please stop accusing John Rennie of being a bot or not reading the questions he votes to close unless you have evidence for that. He's a valuable contributor to this site.

@ACuriousMind that's a bit of a non-sequitur, but whatever.

@DanielSank Right, bots could be valuable contributors, too :)

@Maxwell go to physicsforums as well

@Maxwell Ok

I've now spent about half an hour trying to help you get over that fear and you've done nothing but complain.

I am losing interest in trying to help you.

This makes me sad, because helping people with physic sis one of the highest joys of my life.

@Maxwell I'm losing time. Oh my goodness, the fact that you would even say that indicates a massive lack of respect.

Good day, @JohnRennie @ACuriousMind. See you later.

@DanielSank Have a nice day!

smbc-comics.com/comic/smell-this

Do we get a choice?

@Maxwell You can. But it would do you good to read up on the rather large amount of extant discussion on Physics Meta about it - it is almost certain it has been brought up before.

You're permitted to share it. Just understand that everyone else is permitted to not be persuaded :)

@ACuriousMind wtf did you did delete his account?

@0celo7 Yes. It was a sockpuppet of a currently suspended user.

@ACuriousMind whose?

@0celo7 None of your business

Fine, MOM

@ACuriousMind Yikes, he got ERASED pretty good

should that not cause the suspension time to increase or something

sockpuppeting, i mean

i dont know the policy

I have many socks.

but you're not evading a suspension

That's the official story, yeah.

@BalarkaSen That's for the mods to deal with

Oh man, so tough to not comment...

Ignoring... Most confusing thing about string theory so far has been conventions regarding constants and Clifford algebras in n dimensions,

As a general rule, anything one does to evade a suspension can cause us to increase the length of the suspension.

Gotcha

@bolbteppa How much of that is sign convention misery?

I would say equal, both just so infuriating

The latter has involved a heck of a lot of work to even see as relevant let alone to set up, but once it's done you have it, the conventions however...

@ACuriousMind @BernardoMeurer finishline.com/store/product/… :o

How tacky

I am getting angry watching that video

is it about corny physics

@BernardoMeurer finally got the fortran homework

the test output had multiple typos

78 format ("Step Size:", 1X, F10.6, 1X, "Numerical Result:", 1X, F10.5, 1X, "An. Res.:", 1X, F9.4, 1X, "Err[%]:", F11.5, /)

look at that amazing code

He says he doesn't know what frame dragging is

yet here he is

talking about CTCs

what is frame dragging