The h Bar

heather

14:01

@Bernardo thanks for the link

Slereah

You can very much foliate a spacetime with CTCs

ACuriousMind

@Slereah It is good

First thing I open when I need to know something about that part of string theory

Slereah

Good 2 know

ACuriousMind

And his QFT script is the first thing I open when I need to know something about basic QFT

heather

oh, i forgot to say this @Bernardo - iowa state has a close connection with a national laboratory (Ames) that I believe has a group working on quantum computing, though I don't know how active they are. the university itself also has a lab working on molecular computing (among many other labs) called LAMP

Slereah

14:04

Though I'm starting to lose the grip on why RQM doesn't work

I'm not quite sure why quantizing the Polyakov action works but not RQM

I mean I know specific details of why RQM doesn't work

But the deeper ones are starting to elude me

Bernardo Meurer

@heather Are you trying to get me to transfer to Iowa? :P

heather

@Bernardo yeah =P

Bernardo Meurer

Hehehe

I have a very good incentive to stay in CA though

heather

Iowa State's got fabulous engineering/science programs, it's rated among some of the top in the country y'know

@Bernardo wait, you've already moved?

Bernardo Meurer

No, but I will

I just gotta finish my exams

and then hope the US doesn't deny my visa

Cause I need that F-1

heather

14:07

good luck ::crosses fingers::

Bernardo Meurer

I'm spoofing my phone's GPS to use Tinder pretending I'm in SB already

A true use of what I learn in uni at last

I must say, I am very excited already

Very good, ehm, education there

Kenshin

@BernardoMeurer CA will be first hit by NK

Bernardo Meurer

@Kenshin I'll die smiling

I had a really good answer

But If I say it I will get banned

And heather's eyes will burn

So I will control myself

heather

what is NK?

Bernardo Meurer

North Korea

heather

14:16

uhm...@Kenshin what was the point of that comment?

Slereah

Hm

Maybe the problem is just that I think Malament's theorem is stronger than it is

Kenshin

@heather Bernardo stated he had incentive to stay in CA, I was just pointing out some of the cons of staying in CA for balance

Slereah

Malament's theorem is just "In a compact region, the probability to measure the particle is 1"

It could just be that there's no state of point particles that are of compact support

Though I'd have to prove that I guess

It doesn't so much forbid point particles as it forbids their certain presence within a region

Bernardo Meurer

@JohnRennie What up

John Rennie

@ACuriousMind yes, that's my understanding. Effectively it's the sign of the vector.

Kenshin

14:18

@JohnRennie have you voted ?

John Rennie

@BernardoMeurer Hi :-)

Bernardo Meurer

@JohnRennie Did you go vote dawg

Slereah

The fact that there's basically no papers on Malament's theorem doesn't help

My PSE questions on the topic are on the first google page

John Rennie

@Kenshin No! I had to drive down to my Mum's to take her to a hospital appointment, then she wanted me to stay another week. My polling card is 200 miles away!

Emilio Pisanty

of possible interest to folks

Watch live: Former FBI Director James Comey testifies before Congress

►

please don't discuss here, though

Bernardo Meurer

14:20

@EmilioPisanty I ALREADY POSTED

Kenshin

@JohnRennie hope her eye feels better

Bernardo Meurer

YOU'RE LATE

John Rennie

@EmilioPisanty That promises to be very interesting :-)

Bernardo Meurer

@JohnRennie You won't drive 200 miles to vote? WEAK

My polling station is across the damn ocean

John Rennie

@Kenshin It's gone really well! I took her for a check up this morning and the doctor is really pleased with how well it's gone :-)

Kenshin

14:21

that's excellent to hear :)

John Rennie

@BernardoMeurer If I'd known I could have arranged a postal vote. But I wasn't expecting to have to stay this long!

Kenshin

Trump was right to fire Comey. The organisation was in disarray and was poorly led

Emilio Pisanty

@JohnRennie @BernardoMeurer I just got some code from a colleague, and the makefile calls gcc-mp-4.8

Bernardo Meurer

@JohnRennie I'll call your mother and complain, don't worry

Slereah

"The following example shows that Malament’s theorem fails even for the very simple case where M is a two-dimensional cylinder"

Emilio Pisanty

14:23

is that Mac for g++?

Bernardo Meurer

@EmilioPisanty mp means MacPorts IIRC

Secret

A sneak peak of my quantum question currently writing on PSE:

Bernardo Meurer

So it's just MacPorts of GCC 4.8

Secret

Emilio Pisanty

@BernardoMeurer it compiles with g++ but not gcc, though

Bernardo Meurer

14:24

Note that gcc is not g++ thought

@EmilioPisanty Eh?

Well, gcc-mp should yield a true gcc

Slereah

"Consider a free relativistic particle of positive or zero mass and arbitrary spin. Assume that at time t=0t=0 the particle is localized with probability 1 in a bounded region V . Then there is a nonzero probability of finding the particle arbitrarily far away at any later time."

Bernardo Meurer

try running gcc-mp-4.8 --version

Slereah

Argh

Bernardo Meurer

see what comes out

Kenshin

"the fbi is honest" Comey just committed pergery

Emilio Pisanty

14:26

@BernardoMeurer I don't got no gcc-mp

Slereah

I guess what I need to do is try a compact wavepacket for the Polyakov particle and then see how it spreads

Emilio Pisanty

he's running mac, I'm running ubuntu

Bernardo Meurer

@EmilioPisanty Papa bless

You're a man of god

then tell him to run it

Emilio Pisanty

@BernardoMeurer I'm not particularly in a position to tell him things

how do I tell if some given code is C or C++?

Bernardo Meurer

@EmilioPisanty I'll tell him, give me his phone number

@EmilioPisanty Can I take a look at it?

Such for foo::bar syntax

That's an easy spot

search for namespace

Emilio Pisanty

14:27

@BernardoMeurer name.seekp(0,ios::beg); sname << icont;

kinda thing?

Bernardo Meurer

@EmilioPisanty That's C++

Fo' sure

Emilio Pisanty

@BernardoMeurer hmmmm

so no surprise that gcc blocks up there

Bernardo Meurer

Just call g++ I guess

Emilio Pisanty

but why would it compile under gcc-mp?

Bernardo Meurer

@EmilioPisanty Because macports is weird probably

Emilio Pisanty

14:29

@BernardoMeurer huh

Bernardo Meurer

Or he is manually linking stuff

-lstdc++

If you add that flag to gcc it should compile C++ code just fine

Emilio Pisanty

@BernardoMeurer yeah, that's in the makefile

but why is it not working with my gcc?

Bernardo Meurer

Are you linking with that flag?

Also, define "not working"?

Emilio Pisanty

codemainfile.o: In function `void std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::_M_construct<char*>(char*, char*, std::forward_iterator_tag)':
/usr/include/c++/5/bits/basic_string.tcc:216: undefined reference to `std::__throw_logic_error(char const*)'

and then several pages of similar errors

Bernardo Meurer

How are you calling the compiler?

Also, can't you just change to g++ on the makefile?

Also, do gcc --version and see which one you got

Emilio Pisanty

14:33

gcc --version
gcc (Ubuntu 5.4.0-6ubuntu1~16.04.4) 5.4.0 20160609

@BernardoMeurer if I do, it compiles

at this stage I'm just confused as to why it doesn't compile with gcc in the makefile

Bernardo Meurer

@EmilioPisanty Can you share the makefile?

I bet my ass this is Ubuntu's fault

I hate Ubuntu

Emilio Pisanty

@BernardoMeurer dropbox.com/s/eorbv25fu8kk6ie/makefile?dl=0

@BernardoMeurer you just called me a man of God for running ubuntu

Bernardo Meurer

AH

@EmilioPisanty Well, it's GNU/Linux and I like that

BUT

It's a bad distro and uses APT :P

Secret

Ugh, This integral is not making any sense to me:
$$\lvert a\rangle = \int_{\Bbb{R}} e^{-x^2}\lvert x\rangle dx$$
Is that how you write the continuum version of
$$\lvert a\rangle = \sum_{k=1}^n e^{-k^2}\lvert k\rangle$$
?

Bernardo Meurer

@EmilioPisanty I'll fix this and send it back, one second

I see what's up

Slereah

14:38

I think I got the issue with Malament and point particles

It is not strictly forbidden

they just can't be certainly within a compact volume

and that is true of the Polyakov in 1D

Emilio Pisanty

@Secret that's about right

what's wrong with the first integral?

Secret

Probably because I have been in the maths chat for too long in the past 3 weeks, somehow it looks weird to me. That might be just an illusion

well, I rarely saw $dx$ placed next to a basis vector, usually that stuff before the $dx$ I typically see expressions of the form $\lvert\rangle\langle\rvert$

Emilio Pisanty

@Secret no, it's just fine

Secret

ok

Emilio Pisanty

write $$|\psi\rangle = \int_\mathbb{R} \psi(x) |x\rangle \mathrm dx$$

Bernardo Meurer

14:43

gist.github.com/bemeurer/3464ad9b980d5160e3660e9f183f0275

@EmilioPisanty Check this out

Emilio Pisanty

and then you can show $$\langle x|\psi\rangle = \psi(x)$$

Bernardo Meurer

See if it works

Emilio Pisanty

@BernardoMeurer it's better but not yet there

runs fine on the cpp

stumbles on the .o

Bernardo Meurer

AH

I KNOW I KNOW

ONE SECOND

Emilio Pisanty

codemainfile.o: In function `void std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::_M_construct<char*>(char*, char*, std::forward_iterator_tag)':
/usr/include/c++/5/bits/basic_string.tcc:216: undefined reference to `std::__throw_logic_error(char const*)'

and more pages of that

Secret

14:47

$$\langle x|\psi\rangle = \langle x \rvert\int_\mathbb{R} \psi(x) |x\rangle \mathrm dx$$

Hmm, I guess I need to revise my quantum to justify how the $\langle x\rvert$ can move into the integral since it is a function of $x$ and the thing in the integral is also integrating wrt $x$. I am guessing since the stuff in the integral is a dummy variable, while the $x$ in $\langle x\rvert$ is not thus that term can bubble into the integral and then form the delta function as required

Emilio Pisanty

@Secret No

don't repeat dummy variables

$$\langle x|\psi\rangle = \langle x \rvert\int_\mathbb{R} \psi(x') |x'\rangle \mathrm dx'$$

Secret

ah that makes more sense

Bernardo Meurer

Hmmm

Emilio Pisanty

@BernardoMeurer what's up?

Bernardo Meurer

I'm trying to figure out why it's not working

There's a sketchy flag there

I updated the gist, try again

AHA! That -c flag makes it not run the linker

Emilio Pisanty

14:56

@BernardoMeurer it's stumbling again

Bernardo Meurer

Hmm

Same thing?

Emilio Pisanty

@BernardoMeurer same thing as before

your first version starts with warnings (which are OK) and moves on to errors

the second one goes direct to errors

what are you changing?

Bernardo Meurer

There was a -c flag on the

Ah

I understand what's happening now

But why is this guy doing this

: |

Emilio Pisanty

@BernardoMeurer doing what?

Bernardo Meurer

Linking stuff in a funny way :)

I hate Makefiles

everyone should use CMake

Emilio Pisanty

15:01

@BernardoMeurer then again, you hate about half of all the stuff you can do on a computer

Bernardo Meurer

More than half, by far

This is a many-file project, right?

Emilio Pisanty

@BernardoMeurer not particularly

three .h files and one .cpp

Bernardo Meurer

Do you want to use CMake? Give me the names and I'll write you a CMake file

You'll be happy forever

This Makefile is doing some linking brouhaha because I think this guy is not doing .hpp/.cpp separation correctly

Emilio Pisanty

@BernardoMeurer it's working with g++ up top, and I'd rather not rock the boat

Bernardo Meurer

Alrighty then :)

Emilio Pisanty

15:05

I think I'll just accept the mystery =)

thanks for the help

Bernardo Meurer

I want to email him about why you should separate .hpp and .cpp files

No worries :)

Emilio Pisanty

@BernardoMeurer why should you separate .hpp and .cpp files?

also, what are .hpp files?

Bernardo Meurer

@EmilioPisanty .hpp is what some people choose to name their .h in c++

just so it's not all .h :)

Emilio Pisanty

@BernardoMeurer as in, if you have a resource I can read on, then I can ask him what he's doing regarding that

Bernardo Meurer

And you should separate them so that you can properly link all your .cpp files individually and then link the .hpp's with the main file separately

Oh, I think I have a bookmark

One second

Emilio Pisanty

15:07

he knows c++ and I don't so I'm kinda not in a position to press

Bernardo Meurer

Well, theoretically you don't have to separate the header and the implementation

that does make writing Makefiles easier IMO

However no one should be writing Makefiles

Kenshin

doesn't sound like Trump made unreasonable requests to Comey at all

Bernardo Meurer

You should just be using autotools (::gag reflex::) or CMake

Emilio Pisanty

@BernardoMeurer tutorial for cmake?

as in, for someone who's learning c++ and might otherwise start using makefiles

Bernardo Meurer

@EmilioPisanty These are the resources I used:
https://cmake.org/cmake-tutorial/
https://tuannguyen68.gitbooks.io/learning-cmake-a-beginner-s-guide/content/chap1/chap1.html

Lozansky

15:16

Page 22 in this lecture webs.um.es/bussons/GW_lecture_KG.pdf shows that the time for a binary system to coalesce is is $\dfrac{5}{256} \dfrac{c^5 a_0^4}{G^3 \mu M^4}$. But the dimension for this is $[TM^{-2}]$. What's wrong?

Emilio Pisanty

@BernardoMeurer thanks, will keep them on hand

also

heather

@Kenshin, I'd disagree

Emilio Pisanty

@BernardoMeurer what are best practices for handling makefiles &c when managing a git repo where different people use different OSs?

Bernardo Meurer

@EmilioPisanty Well, first of all, you want to enforce that everyone is using compatible stuff. If some use Windows, and other's POSIX that get's hard. If everyone is on POSIX it's mostly fine. So make sure you have guidelines that make people write portable code, always.

Secret

Hmm, trying to find the representation of an operator that transform the following two particle state into an entangled state where all (x,y) coords are correlated in the way (x1,y1)=(x2,y2) seems very messy...

$$\lvert x_1y_1\rangle \otimes \lvert x_2y_2\rangle = \frac{1}{N^2}\int_{\Bbb{R}}^{(4)}e^{(x'+d)^2+(y'+d)^2}e^{(x'-d)^2+(y'-d)^2} \lvert x'y'\rangle\otimes \lvert x''y''\rangle d^4\mathbf{x}$$

Emilio Pisanty

15:23

@BernardoMeurer sure, but that's over the c code itself, right?

Bernardo Meurer

Second of all, specially in an environment like this, I'd stay away from traditional Makefiles, they can yield a lot of non-portable stuff with includes and so on

@EmilioPisanty Yes

And then autotools is also not a choice

Emilio Pisanty

@BernardoMeurer so, then what?

cmake?

Bernardo Meurer

Yep, CMake is the only portable alternative that I know

Emilio Pisanty

how does that handle a situation where I use g++ and other folks use gcc-mp-we?

(-we standing, obviously, for -whatever)

Secret

I kinda have some idea what it look like by generalising from the finite basis set case, the challenge is how to slot in the delta function $\delta (x',x'')\delta (y',y'')$ into the integral

Bernardo Meurer

15:25

It doesn't care, CMake will use the system defined compiler unless explicitly told otherwise

42

A: How can I make CMake use GCC instead of Clang on Mac OS X?

CMake doesn't listen to CC and CXX. Instead use CMAKE_C_COMPILER and CMAKE_CXX_COMPILER: cmake -DCMAKE_C_COMPILER=/usr/bin/gcc -DCMAKE_CXX_COMPILER=/usr/bin/g++ ... See also the documentation. Alternatively, you can provide a toolchain file, but that might be overkill in this case.

You can also do stuff like this:

20

Q: OS specific instructions in CMAKE: How to?

I am a beginner to CMAKE. Below is a simple cmake file which works well in mingw environment windows. The problem is clearly with target_link_libraries() function of CMAKE where I am linking libwsock32.a. In windows this works and I get the results. However, as expected, in Linux, the /usr/bin/...

WHICH IS A BLESSING

Emilio Pisanty

@BernardoMeurer yeah, that looks pretty powerful

thanks for your help

Bernardo Meurer

Any time man :)

What are you using to write C++?

Emilio Pisanty

@BernardoMeurer atom

but I'm open to suggestions

Bernardo Meurer

I can give a shinning recommendation of CLion

I left Atom for CLion when writing C/C++ and I couldn't be happier

Emilio Pisanty

@BernardoMeurer not free, though

Bernardo Meurer

15:29

@EmilioPisanty Free as in Gratis or freedom?

Emilio Pisanty

@BernardoMeurer free as in the institute IT will be willing to install it

Bernardo Meurer

jetbrains.com/clion/buy/#edition=discounts

There are education discounts

I get mine for free with those

Emilio Pisanty

@BernardoMeurer we're not primarily an education institution

Bernardo Meurer

@EmilioPisanty I won't tell if you don't

Emilio Pisanty

@BernardoMeurer question is whether they'll validate the domain or not

but in any case, it's an institute-managed machine

Bernardo Meurer

15:36

@EmilioPisanty I'd give it a try, it truly is a wonderful IDE

And I'm generally not a fan of IDEs

but it just makes using GDB so easy

heather

15:51

@ACuriousMind/@dmckee/@modwhoiscurrentlyintheroom: would you mind adding Bernardo's AMA to the chat event schedule?

Sha Vuklia

hey @Bernardo question; where you do find your online books actually?:P I kind of feel bad asking you every time, if I could do it myself too!

dmckee

@heather Er ... well, I'll have to figure out how that works first. Give me a minute.

heather

thanks =)

Sha Vuklia

I tried bookzz.org (which moved to b-ok.org recently), but they only have djvu files:(

and I'm not sure how to convert those, because the first couple of conversion sites I tried didn't really do anything, or I couldn't even find the djvu file in the first place:P

Bernardo Meurer

@ShaVuklia I buy all of them of course!

I have never done anything illegal ever

With that said, what book do you need?

Feeds

15:54

dmckee has added an event to this room's schedule.

Sha Vuklia

huh okayy, well the book I was looking for is Optics by Hecht @Bernardo

ACuriousMind

@ShaVuklia The easiest way to convert djvu files to pdf I've ever found is to print the djvu to a pdf printer.

heather

@dmckee thanks!

Sha Vuklia

if you fix that book, you'll get three coffees if you come to Amsterdam :P @Bernardo

dmckee

No problem. The interface is actually very clear.

Bernardo Meurer

15:55

@ShaVuklia You sure know how to convince me

@ShaVuklia Got it, emailing it to you

Sha Vuklia

whutttttt

how can you be so fast?

in that time you can't even finish one coffee :P

but many, many thanks of course!

Bernardo Meurer

That's what she said

Sha Vuklia

I'll try not to misuse your service too often:P only in times of desperate needs

Bernardo Meurer

As I said before, feel free to ask, really

Secret

Ok, this question of mine sounds VERY dodgy. I guess I need some advice...

Bernardo Meurer

15:59

@ShaVuklia Sent!

Sha Vuklia

@Bernardo thanksss~ :) (I got it!)

Secret

I will post it later...

bolbteppa

Bosonic part of 11d sugra is $\int d^{11}x \sqrt{-G}(R - \frac{1}{2}|F_4|^2) - \frac{1}{6} \int A_3 \wedge F_4 \wedge F_4$, where tf does this come from?

Slereah

@bolbteppa science

bolbteppa

Polchinski simply states it, so do tons other sources, most justification I've seen so far was 'lorentz invariance' :\

Slereah

16:05

I'm not up to SUSY I'm afraid

So I don't have a clue

ACuriousMind

@bolbteppa The derivation of the SUGRA actions is horrible, and nowhere really done in detail; the original paper that constructed the 11d SUGRA action is "Supergravity theory in 11 dimensions" by Cremmer, Julia, Scherk.

bolbteppa

Have the paper, the action isn't there :p

Slereah

Shrek himself wrote the paper

Maybe it's one of those "The reader can show" paper

bolbteppa

Yeah, or could be old notation, paper is short and crazy

Slereah

some proofs are pretty hard to track down

ACuriousMind

16:09

@bolbteppa The action is there

Second page, second column: "The Lagrangian we find is the following:"

It's missing all of the explicit calculations to arrive at it, though :P

Slereah

For all we know it could be wrong

ACuriousMind

I also wanted to know how the SUGRA actions are derived and there's really no good reference on it, everyone just copies the action from everyone else :P

What's well-established is the possible field content, first shown by Nahm

Slereah

Here's the equation I arrived at : $$\Gamma = \large{☎}^{-1} \Psi \bar \Psi$$

2

Proof left to the reader

"Due to the conformal symmetry of the super G-string, the second-quantized G-string is the same as the first-quantized one [22]. The only difference is that more parentheses are needed: e.g., Φ[X(σ)]."

bolbteppa

haha fermionic terms confused me

I think there is a derivation in West chapter 13 spread over two sections I'm trying to make sense of

Secret

gah, once the downvote starts, more downvotes will follow...
https://physics.stackexchange.com/questions/338208/simple-entanglement-experiment-probabilistic-entanglement

it also does not help that as of edit 1, I cannot seemed to see what other problems it have

Emilio Pisanty

16:37

@Secret That post is riddled with problems.

Secret

I really should have stick to using electron spins as observables if not because of that heisenberg uncertainty misconception...

That is how I end up choosing position as my observable for this question

$E$ is supposed to be an operator that correlates separately the x positions of the two beams and the y positions. However I cannto seemed to use a delta function to make the x' and x'' equal without making the y" and y' to become x" x'

I am still figuring out how to fix that integral

Emilio Pisanty

@Secret The problems start rather earlier than that. The fact that you're using $|x_iy_i\rangle$ for the states before $E$ and the states after it is deeply problematic.

I suspect that what you actually want is the state $$|\psi\rangle = \int_\mathbb{R}\int_\mathbb{R} |x,y\rangle\otimes |x,y\rangle \mathrm dx\mathrm dy$$

though it has some normalization problems

Secret

16:52

uh actually I have not really wrote down the entangled state. I only have wrote down the initial state (which is a product state), and the operator $E$ that does

yeah, trying to get to the above state from that product state is what I am trying to do with $E$

Ok, I think this expression of $E$ will get me there when I act it on the product state:

$$E=\int_{\Bbb{R}^4} \delta(x''',x'''')\delta(y''',y'''')\lvert x''' y'''\rangle\langle x''' y'''\rvert \otimes \lvert x'''' y''''\rangle\langle x'''' y''''\rvert d^4\mathbf{x}$$

Emilio Pisanty

@Secret that's still meaningless

no, wait

what is $d^4\mathbf x$?

is it $dx'''dy'''dx'''' dy''''$?

if so, it should be specified explicitly

Secret

yes

Emilio Pisanty

and more importantly, quadruple primes are past the edge of readability

triple primes are already out, to be honest

Secret

I don't know any better symbols that can tell people I am still dealing with positions

Emilio Pisanty

change to $x^{(3)}$ etc

or drop three primes

there's no reason to be using all those primes to begin with

they're dummy variables, and there's no other primes anywhere. $x$ and $x'$ will do just fine there.

Secret

17:07

oops forgot the supergaussian explosions, fixing them now

Emilio Pisanty

@Secret it's much more grave that you're using $|xy\rangle$ for states that are not position eigenstates

Secret

So xy are generally reserved only for position eigenstates?

Emilio Pisanty

@Secret absolutely reserved

it makes your post entirely incomprehensible

... as I pointed out 45 minutes ago

Bernardo Meurer

17:34

@JohnRennie Black flag is damn good!

Bernardo Meurer

17:49

Sometimes I find the weirdest functions on the Standard Library strspn

Transcript for