The h Bar

Semiclassical

15:02

I guess the point to be made is just this. Consider a function of the form $p(x)e^{-x^2/2}$ where $p(x)$ is polynomial in $x$. Is this square-integrable?

Phase

Yeah because the exponential vanishes

like when you do a transform

Semiclassical

more precisely, the exponential vanishes more quickly than any polynomial increases

Phase

Oh

Semiclassical

so it goes to zero instead of blowing up like the polynomial alone would

Phase

How does one prove that?

That the exponential vanishes more quickly than any polynomial increases

nevermind

brain aneurysm, i got poly and function switched in my head

Semiclassical

15:05

now suppose you multiply $p(x)e^{-x^2/2}$ by $x$. Is there any possibility of this not being square integrable as well?

Phase

uh im not entirely sure on the condition that makes the exponential vanish more than the other function increases so I cant say

0ßelö7

@Semiclassical what does "sufficiently fast" mean in this context?

Phase

I guess id wager that for most cases it ends up being square integrable

Semiclassical

@0ßelö7 $\phi(x)$ normalizable but not $x\phi(x)$.

0ßelö7

@Semiclassical No

Semiclassical

15:07

indeed not.

0ßelö7

$e^{-x^2}$ is a Schwartz function

Semiclassical

the point is that if $p(x)$ is a polynomial in $x$, then so is $xp(x)$.

0ßelö7

it and all of its derivatives decay faster than any polynomial

Semiclassical

so if $e^{-x^2}$ decays faster at infinity than any polynomial, then it decays faster than $xp(x)$ as well.

0ßelö7

@Semiclassical Oh. Something like $1/x$ but smoothed near $0$.

Semiclassical

15:08

right.

I was thinking something like $\phi(x)=(1+x^2)^{-1/2}$ but couldn't remember if there was a nice potential for that.

Phase

hang on

another dumb question

since the exponential is infinitely differentiable

couldn't you expand it AS a polynomial?

Semiclassical

Not for large $x$, you can't.

0ßelö7

wot

Semiclassical

For small $x$, sure. But small $x$ isn't the point here.

0ßelö7

first: not all smooth functions are analytic

second: never taylor expand

Semiclassical

15:11

@0ßelö7 just to check, we're agreeing right now right? :P

0ßelö7

@Semiclassical exp is analytic, but it's best to just forget about analytic functions altogether

you get into all sorts of trouble if you assume functions are analytic

Balarka Sen

SemiC is saying e^(x^2) is not analytic at infinity

Semiclassical

true.

right.

e^(x^2) at infinity is almost a canonical example of how nice-looking functions can behave weirdly

0ßelö7

the series converges uniformly on compact intervals

Semiclassical

(one usually talks instead about e^(-1/x^2) at small x, but same difference)

Balarka Sen

15:13

@Semiclassical bump functions 101

0ßelö7

@BalarkaSen that's not a bump function

the usual bump function is $\exp(-1/(1-x^2))$

Balarka Sen

everyone knows that

@Semiclassical er, I meant e^(-x^2), typo

(I hope you meant that too)

Semiclassical

right

concrete example time: Suppose $V(x)=(2x^2-1)/(1+x^2)^2$. This has a valley in the center, peaks away from that, and then goes to zero as $V(x)\sim 2/x^2$ for large $x$.

Turns out if $\phi(x)=1/\sqrt{1+x^2}$ then $-\phi''(x)+V(x)\phi(x)=0$, so $\phi(x)$ is a zero energy eigenstate of this potential.

however, $x\phi(x)\to \pm 1$ as $x\to\pm \infty$. So $x\phi(x)$ is not normalizable despite $\phi(x)$ being normalizable.

Phase

now im confused in a differeny way

Semiclassical

So this is an example where $\

Phase

15:20

I thought this conversation was about why xf(x) IS normally normalisable

Semiclassical

It's turned into me giving an example of where it'd go wrong, I guess.

Phase

o

Semiclassical

One thing to note, I guess, is that in this case there's no binding energy: the potential energy at infinity is zero, and the total energy is zero

Phase

so let me check if I have this right

Semiclassical

by contrast, I think any normalizable state $\phi(x)$ with $E<0$ would also have $x\phi(x)$ normalizable

Phase

15:23

Actually

First

You write a lot about how a function tends to another for large x, and how that describes its behaviour when it goes to zero

To say that something vanishes "fast", does it have to approach a different, standard function that is known to also vanish "fast"?

I'm still a bit tripped up by the language

Semiclassical

this is me being sloppy

if you want $\phi(x)$ to be square integrable, though, you need the integral $\int_{-\infty}^\infty \phi(x)^2\,dx$ to be finite.

Phase

Is converging to zero equivalent to "vanishing fast"?

ok cool

Semiclassical

and that won't work if $\phi(x)^2 \sim 1/x$, since then the antiderivative of $\phi(x)^2$ would behave like $\ln x$ (which diverges at infinity)

Phase

i meant the antiderivative converging to zero as $x \to \infty$

that was shoddy communication from me

Kyle Kanos

@ACuriousMind (or other mod) got hit again with another seemingly serial dv attack :( :(

Semiclassical

15:27

I don't follow.

Phase

i edited it

Semiclassical

if $\phi(x) \sim |x|^{-1/2}$ for large $x$, then it goes to zero as $x\to\infty$. But in that case $\phi(x)^2\sim |x|^{-1}\implies$ the antiderivative of $\phi(x)^2$ behaves like $\ln |x|$ at infinity

Phase

yeah

Semiclassical

and $\ln |x|$ blows up at infinity, so you don't have a convergent integral

so $\phi(x)\to 0$ isn't sufficient.

Phase

cool

I was asking about the integral rather than the function

Semiclassical

15:29

ah

Phase

which brings me to a side question

Semiclassical

the integral needs to be finite

which puts certain conditions on how $\phi(x)$ can behave asymptotically for large $x$.

Phase

I was trying to communicate that the integral tends to zero but I mean that the Value of the function you get my integrating tends to zero

How do you actually communicate that

$\int f dx (x)$?

Semiclassical

the antiderivative goes to zero at infinity? that's what I've been doing above.

Phase

Im just asking a notation question really

Semiclassical

15:31

(more generally, if $\phi(x)\sim |x|^{-p}$ for large $x$ then $\phi(x)^2\sim |x|^{-2p}\implies$ antiderivative behaves like $|x|^{1-2p}$, which goes to zero if $p>1/2$.)

Phase

$lim_{x \to \infty}f^{(-1)}(x)$

I guess that

where the superscript is like the old dash notation for derivatives?

Semiclassical

eh, maybe. can't say I really like that

Phase

What would you recommend? I just want a way to write it people will understand

because im bad at explaining with words

Semiclassical

$\int^x f(x')dx'\to 0$ as $x\to\infty$

Phase

ahhhhhhhhhhh of course

Semiclassical

15:34

(i'm omitting the lower endpoint because it's irrelevant here)

Phase

didnt think of using the \int with just one limit

could I ask some other basic questions?

Semiclassical

I think I'm going to have to take a break.

Phase

fair, thanks anyway

0ßelö7

Lol

Emilio Pisanty

0

A: Energy loss in Capacitors

question is physics unit 1 by electrostatic

that's a new one

Semiclassical

15:47

@EmilioPisanty da fug

Emilio Pisanty

@Semiclassical yeah, I know

why would someone do that?

Semiclassical

so: NAA or VLQ? :P

(I flagged as NAA)

Emilio Pisanty

@Semiclassical NAA

Semiclassical

yeah

Emilio Pisanty

Hey @JohnRennie a.k.a. GRguru, any chance for a nice essay on curvature?

1

Q: When a massive object warps the space around it, does the amount of space expand?

I may have watched those physics shows for too long, but the warping of space is usually described as a surface which bends. Whenever that happens, the surface they are showing increases in surface area. (like those weighty spheres on a stretched rubber surface) I was wondering if this is practi...

general-relativity

Semiclassical

15:50

for it to be VLQ I feel like there'd actually have to be some kind of answer however specious

and that one is just a non sequitor

Emilio Pisanty

@Semiclassical VLQ is a complicated flag, but if it falls under NAA then just go with that and rid yourself of the complications

Semiclassical

right

any time spent debating on that question is wasted

Emilio Pisanty

yeah, I know

that time should be spent admiring the images

I mean

what's not to love?

John Rennie

@EmilioPisanty you're so too late :-)

Emilio Pisanty

@JohnRennie =P

Semiclassical

15:53

as an answer to the question, it's worthless. as an instance of dadaism, it's brilliant

Emilio Pisanty

on the other hand, you very casually use the word "radius" like it can be unambiguously defined

Semiclassical

(actually, it could probably be made even more WTF.)

John Rennie

@EmilioPisanty what baffles me is that someone obviously went to some trouble to provide that answer. Maybe it is an artistic statement :-)

Balarka Sen

@Semiclassical Nice one!

Emilio Pisanty

@JohnRennie yeah, that's what baffles me as well

@JohnRennie I was thinking more in terms of the nonlinear relationship between radius and circumference

Balarka Sen

15:55

I'mma post an answer to the question John Rennie just answered with a snapshot of the cover page of Hawkings-Ellis and Crash Bandicoot photoshopped over it now

John Rennie

@EmilioPisanty the Schwarzschild $r$ coordinate is of course unambiguously defined. The question is whether I can squeeze that into the answer without causing confusion.

Emilio Pisanty

@JohnRennie I tend to think that the current formulation is too ambiguous to be useful

John Rennie

@BalarkaSen Crash Bandicoot! I'd upvote any answer with Crash Bandicoot in it :-)

Balarka Sen

Woah.

John Rennie

The hours I spent playing Crash Bandicoot games!

Emilio Pisanty

15:58

@BalarkaSen well, you know, both you and @JohnRennie should watch out

Semiclassical

what

Emilio Pisanty

> Hawking Built The Big Bang On A Foundation Of Quicksand

Balarka Sen

Beautiful

Emilio Pisanty

what it says on the tin, pretty much

Semiclassical

not that the other talks in that session look very good, but that was seriously at an APS meeting?!?

Emilio Pisanty

16:01

@Semiclassical mind the session title

> Session Y9: New Directions in Astrophysics

Semiclassical

sure, but

Emilio Pisanty

I'm not sure when APS started hiring brits for the position in charge of euphemistically naming sessions

but whoever they got was a brilliant hire

Semiclassical

calling that a new direction in astrophysics is like calling "tripping over your own shoes tied together" a new direction in human transportation

Emilio Pisanty

well, that session is what some people might colloquially call the crackpot session

Semiclassical

Ah

Emilio Pisanty

16:03

it's a staple at APS meetings

i.e. it's too much trouble for APS to turn them away so they often just give them a session to themselves and that's that

Semiclassical

"Abandon all hope, ye who enter here"

Emilio Pisanty

@Semiclassical you'd be surprised at how many people actually turn up

Semiclassical

i find it depressing that you're probably right

Emilio Pisanty

though not all of the session attendance is motivated by a scholarly desire to learn from the speakers

in fact rather the opposite

Semiclassical

well, sure

Emilio Pisanty

16:05

it's a bit cruel to turn up to that kinda session just looking for a laugh

but hey, I've done it, so there

Semiclassical

lol

"If making fun of crackpots is wrong, I don't want to be right" ?

0ßelö7

@EmilioPisanty what do they talk about

Semiclassical

well, did you see the abstract linked above?

i'm not entirely sure that's representative, but it seems a good starting point

0ßelö7

I did not

Emilio Pisanty

16:21

@Semiclassical @0ßelö7 the abstract above is representative (link)

0ßelö7

@JohnRennie I'm not going to say your answer is wrong, but spacetime isn't a metric space so the analogy with balls and circles doesn't work.

It's true on the spatial slices, but the curvature of the whole thing cannot be interpreted like that.

Emilio Pisanty

the one I saw that I remember the most was this one meetings.aps.org/Meeting/APR15/Session/H14.8

John Rennie

@0ßelö7 yes agreed. It's better than pretending it's a rubber sheet though :-)

Emilio Pisanty

no, wait

meetings.aps.org/Meeting/MAR15/Session/Z23.14 this one

meetings.aps.org/Meeting/MAR15/Session/Z23.5 on that same session also memorable

Semiclassical

hnnnnng

John Rennie

16:24

If someone is going to be a crackpot they could at least make the effort to be a coherent crackpot :-)

Semiclassical

maybe it's coherent in their own head, but decoheres rapidly in the presence of an observing audience? :P

I guess this is the Crackpot Measurement problem: Is a crackpot internally consistent within their head?

0ßelö7

@JohnRennie I dunno, I think of geometry in terms of rubber sheets.

Emilio Pisanty

@Semiclassical I can testify, from experience, that the presence of an audience does little to alter the speaker's impression of the coherence of the material

John Rennie

@0ßelö7 that's because you're Riemannian and the rest of us are just pseudos

ACuriousMind

@EmilioPisanty I've seen a lot but that one is truly exceptional

Emilio Pisanty

16:28

I got a whole new appreciation for that kinda thing that week

nowadays, when I see e.g. a Concave Earth video, and part of me says "surely the guy is just doing it for a laugh, he can't be serious", the other part of me goes "remember all them guys in APS who were about as crazy but were definitely serious about it?"

0ßelö7

@JohnRennie in the pseudo case, curvature has nothing to do with anything we can imagine

We can't even imagine what 2D Minkowski space is. We have no idea what that metric means

Semiclassical

@EmilioPisanty The conspiracy theory I actually saw someone on campus advocating for on the street was that satellites don't exist

0ßelö7

So the rubber sheet analogy is fine because there is no correct interpretation in human terms

Semiclassical

I don't want to believe that person was serious, but I find it hard to doubt

0ßelö7

@ACuriousMind the last line is gold

@Semiclassical I'm agnostic to such things

I don't know if there are satellites or not

Semiclassical

16:34

...

okay then

0ßelö7

i have no evidence of it

Just people telling me

IMO that's the only reasonable conclusion

John Rennie

@0ßelö7 Haven't you seen satellites passing at night?

0ßelö7

@JohnRennie no

John Rennie

Really?

0ßelö7

I live in a city, I can't even see stars at night

John Rennie

16:35

That's ... a bit sad really.

Balarka Sen

BALEETED

0ßelö7

@JohnRennie I used to not live in the city and I saw mysterious things in the sky

Semiclassical

"That's a bit sad." "You're sad." "...yes, that's what I just said."

0ßelö7

But no evidence that it was a satellite

John Rennie

When I was a teenager we used to have few beers, smoke a little, then lie on our backs in a field gazing up at the sky. All very cosmic (man).

0ßelö7

16:37

@JohnRennie also I'm a bit offended by the implication that I'm only Riemannian

John Rennie

I'm positively definite that you are :-)

Balarka Sen

Damn, that's a good pun.

John Rennie

(and you thought there was no such thing as a relativistic Dad joke)

0ßelö7

I'm mad but also smiling

Semiclassical

@JohnRennie nah, just gotta tell it fast enough

Slereah

16:42

Hey folks

I'm trying to make a graphical representation of a quantum computer in javascript

it's not too easy

Single particle state is fine

samuel-lereah.com/articles/CS/quantum-computing

Emilio Pisanty

@Slereah you sound surprised

Slereah

Doing entanglement is gonna be complicated

Emilio Pisanty

@ACuriousMind I'm a bit sad that it's gone

it was a work of art, that one

Slereah

Entanglement is probably okay to represent with 2 particles, but that's gonna be cluttered

More than 2 and it's gonna be a mess

0ßelö7

17:06

@Slereah I need to come up with a reading plan

Like one book for 1.5 hours every day

Rotate books

Slereah

You can't read a book upside down

0ßelö7

@Slereah yeah I can

It's trivial

0ßelö7

19:07

@ACuriousMind how the fuck do these algebraists just know that A_5 is a symmetry group of a dodecahedron

@BalarkaSen nice

Going for the god memes

Balarka Sen

lol you noticed senpai

Bernardo Meurer

@JohnRennie Are you around?

Balarka Sen

@0ßelö7 I think some nontrivial amount of algebra goes into that

Like, first you have to show it's simple

That's a class equation manipulation

Emilio Pisanty

@DavidZ I hope you don't mind this edit.

Balarka Sen

Once you know that, you have to invoke there's only one simple group of order 60

That's A5

John Rennie

19:19

@BernardoMeurer Hi :-)

What's up?

Bernardo Meurer

@JohnRennie So, I'm rebuilding my audio stack now that I moved

And I wanted to add a music server

John Rennie

OK?

Bernardo Meurer

But I want it to double as a distcc client and I'd like it to have a GPU for some compute stuff I'm doing

Mostly hashing

I was thinking getting a used dell optiplex b/c I can lay it sideways

and literally stack it with the amp, cassette deck, etc

Any recommendations/ideas/etc?

John Rennie

The Optiplex will be noisy

Bernardo Meurer

Any alternatives?

John Rennie

19:23

I don't think you should try to combine the two roles.

Bernardo Meurer

$$$

I guess

I guess I can have a raspberry pi

John Rennie

Buy a cheap laptop to use as the music server.

Bernardo Meurer

the cheapest of laptops :P

With SSH

John Rennie

Even a cheap laptop will sound great with a decent DAC, and laptops are silent and sip electricity

Bernardo Meurer

(walking to library, be right back)

John Rennie

19:25

If you're building a machine with a GPU to use for number crunching then even the GPU will cost more than a decent laptop.

My point is that once you've budgeted for building a number crunching PC adding a laptop won't bump up the cost very much.

Jasper

My new laptop was very cheap, but I notice all kinds of problems with it.

John Rennie

@BernardoMeurer: Just look at these

@Jasper what laptop is it?

Jasper

@JohnRennie It's a Lenovo Ideapad 110. Optical disc drive: need to put CD in a few times to detect. Wireless: goes off for a minute every hour. TeXworks editor: takes 14 seconds to start up after clicking icon. HD videos: sometimes the picture freezes. LOL

John Rennie

@Jasper Lenovos are basically fine if rather cheaply built. Remove the hard disk, put in an SSD and do a nice clean install of Windows 10 and that will transform your machine. Cost will be about £75 for a 256GB SSD.

Sid

@JohnRennie Is it bad to connect a laptop to the charger all the time?

Jasper

19:36

@JohnRennie Ah, so you think that SSD is better than HDD?

John Rennie

@Sid No

Bernardo Meurer

@JohnRennie Am I not better suited by a Raspberry Pi (or some single-board PC of the short) connected to an SSD and then buying a proper DAC

John Rennie

@Jasper The speed increase when you put in an SSD is unbelievable. SSDs are expensive but worth every penny.

Bernardo Meurer

Because a laptop won't fit my stack

Sid

@JohnRennie My friend claims it will damage the laptop's battery.

How wrong is he?

John Rennie

19:37

@BernardoMeurer can do I guess. I don't have any experience of working with the RPi and I have lots of experience of Dell laptops :-)

Bernardo Meurer

@JohnRennie It's just because there'll be a turntable on top of the stack, so I can't have the laptop there

Although disabling the sleep-on-close it could work

John Rennie

@Sid if you leave the laptop on the charger the laptop will charge the battery to 100% then stop charging it. In effect the battery is then disconnected. So by leaving the charger connected all the time you are effectively storing the battery in a fully charged state.

Bernardo Meurer

But at that point I'd rather spend 30 bucks on an ARM pc, and invest in a good DAC

@JohnRennie That's true on Dell, but not across the board AFAIK. Even in my Dell I can disable that and tell the computer to be dumb

Sid

@JohnRennie He claims it overheats the laptop.

Is that supposed to be bad?

John Rennie

@Sid in theory storing a battery fully charged can reduce it's life. You're normally recommended to store batteries with about 2/3 charge. However with a modern battery the effect is likely to be small.

It doesn't overheat the laptop. That is, to use a technical term, complete bollocks.

@BernardoMeurer I'd probably go with the RPi. It's $30! The DAC you can reuse if/when you do buy something else to use.

Bernardo Meurer

19:42

Yeah :)

And I can run Arch Linux on it which I have a good mastery of

Just this week I managed to get BTRFS working with full-system LUKS encryption + LZO compression

John Rennie

@BernardoMeurer my main point is that you will normally want to keep the number cruncher off and only fire it up when you need it. You can use WOL to turn it on so you don't even need to walk over to it :-)

Bernardo Meurer

Even my boot partition is encrypted!

John Rennie

That's what I do. I have a mofo Xeon based server in the spare room and I wake it when I have some crunching (usually video resampling) to do.

Bernardo Meurer

Does it have GPUs?

John Rennie

No.

Bernardo Meurer

19:44

Dammit

John Rennie

But then I don't have any software that could use a GPU.

Bernardo Meurer

The makerspace has a nice machine with a 1080, but they don't know how to setup SSH ::eyerolls::

John Rennie

I have to say I would use RDP. It's supported on *nix. For a remote graphical session it's the best protocol I know.

Bernardo Meurer

I don't want a graphical session :(

I'm sshing to Brazil I want as few bytes as possible

John Rennie

We all pretend to be command line demons, but a GUI is quicker and easier. That's life :-)

RDP is astonishingly good at handling low bandwidth connections. I've used it over ISDN lines and it's still usable.

Bernardo Meurer

19:48

For some things, I like a GUI for editing my GRUB2 configurations for example

I hate using CLI for that

Sid

@JohnRennie So, basically I don't have to worry about anything and my friend is simply speaking nonsense.

John Rennie

@Sid I don't know what model your laptop is and how it's charging circuitry works, but as a general rule you needn't worry.

Bernardo Meurer

@JohnRennie But to be honest, I am much faster doing things on the CLI than I am on a GUI generally, just because I know the commands by heart and I type fast. The time to figure somoething out, however, is much longer (going to man, --help, googling, etc)

on a GUI you just go clicking around till you find what you need

John Rennie

The CLI and GUI both have their place. The point is that with a GUI you can open a terminal any time you want. I use the Windows command prompt all the time, but ina GUI session.

Bernardo Meurer

Yeah, but my thing is I just really, really don't need a GUI

I just run one command and it takes several hours to complete

And the output is a file I automagically send to the host over rsync

dmckee

19:52

@JohnRennie I recall speeding a bunch of my advisor's dollars on a 10,000 rpm drive for my grad school computer because I had concluded that our Monte Carlo and analysis programs spent an inordinate amount of time waiting for disk IO.

John Rennie

@BernardoMeurer: Look at this - $50 for a 3GHz i7.

Bernardo Meurer

That's pretty good

dmckee

In principle that 400 MhZ PII desktop was about one-quarter the horsepower of the HP minis that formed our main analysis machines. IN practice if could rum MC 50% faster and analysis almost twice as fast.

Bernardo Meurer

and I have an 8GB RAM stick laying around

John Rennie

@dmckee to be fair I shouldn't make blanket statements about speed because you really need to monitor to identify the bottlenecks. But with modern laptops the disk is almost lways the bottleneck.

Manufacturers put in 5,400 rpm disks to save cost, but it really hobbles the laptop.

@BernardoMeurer no disk, but you'd want to junk the disk and fit an SSD anyway.

dmckee

19:56

@JohnRennie Well, those HP just ran several MC or analyzers and swapped out when waiting, so they got more done overall. They were impressive machines. But my analysis finished faster by doing it on my desktop and using the fast disk.

John Rennie

Note that laptop doesn't have the special disk connector that E6230s need, but you can get those to $5 on eBay. It's called an interposer.

Bernardo Meurer

@JohnRennie If I do get a cheapo laptop I'd like it to be a thinkpad, they look the best closed down (all black, brick-shape)

John Rennie

Not a Dell! Heretic! Burn him!

Sid

@JohnRennie ....I have got a Lenovo

dmckee

@JohnRennie Isn't a thinkpad lenovo now? USed to be IBM but they sold it.

Sid

19:57

(Lenovo G50)

Bernardo Meurer

@dmckee yeah, it's Lenovo Thinkpad instead of IBM Thinkpad

John Rennie

The g50 is a nice laptop.

Bernardo Meurer

Still the great looks though

Sid

Yeah, It hasn't given up on me yet

John Rennie

Thinkpads are very nice laptops. It's just that I have no experience of them so I can't advise on buying them.

heather

20:00

hello

John Rennie

Anyhow, I have to drag my sorry carcass off to bed. I need to be at work in 8 hours. See you all tomorrow.

heather

good night @JohnRennie =)

Sid

20:18

I am sleeping <4 hours for the last week.

Is that bad?

Jasper

@Sid Yes, very bad.

Sid

I do take naps sometimes

heather

@Sid yeah, you need more than 4 hours of sleep over 7 days

Jasper

If you are not sick, try to avoid naps and sleep 8 hours every day.

heather

^

Sid

20:23

I am surviving okay so far. 8 hours is too much. I can't sleep for 8 hours

heather

hey, 3k+ peoples - there's 24 cvs in the review queue.

please head on over to do some reviewing =)

Jasper

There are some people in the world who do not need to sleep.

heather

@Sid the key words being "so far"

David Z

@EmilioPisanty no problem. I didn't realize it was mod-only on this site.

Jasper

And there is one who I think does not need to eat. An Indian holy man who seems to be able to convert sunlight into energy.

Of course, people won't believe this is a genuine case, but I think it is certainly a possibility.

Sid

20:25

Yeah, well I definitely need to eat.

ACuriousMind

@Jasper Eh, uninterrupted ~ 8 hour sleep is a modern invention - our ancestors were fine with two ~4-hour periods with some awaking at night. Unless you have some medical condition, just sleeping when you feel tired works.

Jasper

@ACuriousMind Hmm, I think you are right. That's what I find myself doing these days actually.

Sid

If I add naps to 4 hours of sleep.. I might get it to around 6 hours.

Jasper

Prahlad Jani

Prahlad Jani (Gujarati: પ્રહલાદ જાની), also known as "Mataji", (born Chunriwala Mataji, 13 August 1929) is an Indian breatharian sadhu who claims to have lived without food and water since 1940. He says that the goddess Amba sustains him. == Early life == Born Chunriwala Mataji, Jani grew up in Charada village in Mehsana district. According to Jani, he left his home in Rajasthan at the age of seven, and went to live in the jungle. At the age of 11, Jani underwent a religious experience and became a follower of the Hindu goddess Amba. From that time, he chose to dress as a female devotee of Amba...

That is the man who does not need to eat.

heather

@Sid for the whole week?

and i thought i was tired in the morning...=P

Jasper

20:30

There are also some people with plant-like structures growing out from their bodies. I watched a documentary on this before.

0ßelö7

@BalarkaSen the coarse geometry seminar was really interesting today

Sid

@heather ~4 hours of uninterrupted sleep per day. Around an hour and haf of naps

0ßelö7

noncompact index theory.

heather

@Sid oh per day, phew

that's a lot better.

0ßelö7

@ACuriousMind Can you roughly tell me what "Witten's Morse theory" is?

Jasper

20:34

I have gone for about 30 hours without sleep before. That is the maximum for myself.

ACuriousMind

@0ßelö7 That refers to this paper, I believe

Sid

@heather 4 hours of sleep per week? That would be my roommate 's sleep schedule

ACuriousMind

I read it once but I don't recall enough from it to give a confident summary

Sid

Anyway, I should try and sleep for now

Jasper

Yes, you should. Zzzzzzz.......

Sid

20:37

It is 2 AM. So..4 hours should be enough for me

0ßelö7

@ACuriousMind What is "It is shown that the Morse inequalities can be obtained by consideration of a certain supersymmet
ric quantum mechanics Hamiltonian" supposed to mean?

Like from physics? That sounds like deriving $\pi$ from quantum mechanics

Jasper

Pi can be derived from the supermarket.

ACuriousMind

@0ßelö7 Nah, he's just using techniques usually applied to supersymmetric systems to analyse the Laplacian, which gives a new proof of the Morse inequalities

0ßelö7

@ACuriousMind what the heck does the Laplacian have to do with physics?

ACuriousMind

@0ßelö7 That's a silly question :P (since it appears in plenty of physics). In any case, for a simple supersymmetric system, you have two supersymmetry operators $Q_1,Q_2$, and $Q_1^2 = Q_2^2 = H$. If you take $Q_1 = \mathrm{d} + \mathrm{d}^\dagger$ and $Q_2 = \mathrm{i}(\mathrm{d} - \mathrm{d}^\dagger)$, you get $H = \mathrm{d}\mathrm{d}^\dagger + \mathrm{d}^\dagger \mathrm{d}$, i.e. the "Hamiltonian" is the Laplacian. For more details you'll have to read the paper.

0ßelö7

20:49

@ACuriousMind I've never seen it in the context of compact Riemannian manifolds and SUSY

0ßelö7

21:32

@JohnRennie whoever came up with this language was an insane person

DanielSank

21:56

@EmilioPisanty possibly worth a bounty:

2

Q: Noisy limit of sinusoidal modulation of the frequency of a two level system

I've been reading the paper Motional Averaging in a Superconducting Qubit by Li et al. from 2011, in which they study what happens when they modulate the frequency of a two level system. Specifically, they study a system with the Hamiltonian \begin{equation} H(t) = \frac{\hbar}{2}[\omega_0 + \d...

quantum-mechanics noise open-quantum-systems

user228700

Hi, everyone :-) Is anybody free to help me with an incredibly dumb question about electric power that I have had for ages?

user228700

Do ping me if you do!

Bernardo Meurer

22:32

@ACuriousMind How are you love?

Transcript for