The h Bar

Danu

20:01

@JamalS Told ya, linearizeeee :D

JamalS

@Danu ??

"Linearize"? That's what you always do in GR perturbation theory...

Danu

@JamalS well then there you go, that's already telling you how small you are taking the perturbations, no?

JamalS

I'm not 100% satisfied, but I'll drop the question... Maybe it's not that important...

By the way, do you know how to use xTensor?

I can't get it to work.

I've looked at their guide, but I don't know how to get it to expand the perturbation equations explicitly, component by component.

Danu

I don't know

I REALLY REALLY REALLY want to be able to manipulate tensors in Mathematica though!!!

JamalS

Yeah, me too :(

It's just not practical doing GR perturbation theory without Mathematica

I'm sure all the papers I've read on it in the past decade were done with at least some aid from Mathematica

Danu

20:15

lol

have you looked at Weinberg's book on Cosmology?

Trying to read it convinced me I cannot deal without mathematica

JamalS

Nope

Have you seen Mannheim's book on brane-localized gravity?

There was a paper by Regge and Wheeler on perturbations of a black hole, and they evaluated the full perturbation equations by hand!

Danu

It'll tell you, Weinberg does similar things

JamalS

Maybe there's a way to do it in the Cartan formalism that's faster

Kyle Kanos

Pffft.......You guys and your gibberish

"mathematica" "perturbation"

JamalS

Professor Mark Wise told a story about Weinberg in a lecture. To paraphrase, he said whenever he had large, complex computations he had to evaluate, and he was flustered, he would give it to Weinberg.

And he'd get it back in a day...

@KyleKanos: Lol, what's the matter?

Kyle Kanos

20:21

I'm just messing, that's all

JamalS

@Danu: Now you know what I look like :)

Danu

Just messing? :)

Kyle Kanos

Though, I often feel left out because I don't do GR & half the people in here do it.

Danu

@JamalS not really, but you know what I look like

@KyleKanos Wait, weren't you doing some cosmology stuff?!

Kyle Kanos

Nope, I do Supernova Remnants

Danu

20:22

How do you... not need GR?!

Kyle Kanos

The other Kyle does cosmology stuff, I think

Because plasmas are diffuse enough material to not warp space

JamalS

I like GR most because of it's connection (pun intended) to diff. geometry

Kyle Kanos

If I wanted to model the actual explosion of a supernova, I'd probably need GR

So messing = brass?

Danu

@JamalS Har har har

@KyleKanos IDK, I'm Dutch

JamalS

::giggles::

Danu

20:24

@KyleKanos You could quite easily learn GR, if you tried... At least the basics (I don't know more than that either)

Kyle Kanos

@Danu The German Wiki translated to brass

Danu

Are you familiar with tensor notation in the context of SRT?

Kyle Kanos

@Danu It is one of my goals after I graduate

Danu

@KyleKanos then, sure :D

@KyleKanos :D cool

JamalS

@KyleKanos: I'd recommend 'Applied Differential Geometry.' Once you learn that, GR comes as a very straightforward application.

Kyle Kanos

20:25

@JamalS: Burke's book?

JamalS

ISBN 978-0521269292

Yeah

If you want a quick course

Otherwise, it's Spivak

Kyle Kanos

Added to my Amazon wish list

JamalS

Pirate it :)

Kyle Kanos

I have Joel Franklins Advanced Mechanics and General Relativity

That's what we used for my Mechanics + GR course a number of years ago

But the course was awful

So I didn't really get much out of it

JamalS

By the way, you program in Fortran, right?

Kyle Kanos

20:29

That I do

Danu

@JamalS Wrong. The way to go is Lee's Introduction to Smooth Manifolds, for basic DiffGeo

Trust me, it's based on experience

JamalS

Is it dense?

Danu

Spivak is too high-level if you're not a mathematically oriented student (i.e. have 2-3 years of mathematics - for mathematicians, under your belt)

JamalS

That I agree

Have you seen Burke's?

Danu

Lee's book starts with some high level stuff in the first half of the first chapter but then builds up reasonably simply - and he has good appendices that cover all the basics you need

@KyleKanos have you ever taken formal math courses (Analysis, Topology?)

Kyle Kanos

20:31

@Danu Nope and nope. Calc I-III, Number Theory, Diff Eq I & II, and Linear Algebra were the requirements for the Mathematics Minor I got in undergrad.

Danu

PS I don't think you really need to understand DiffGeo at all to do basic GR

@KyleKanos OK, so you're more like me

JamalS

Yeah, but, it's interesting!

Danu

You will need to familiarize yourself with some analysis and topology before you can handle DiffGeo (if you want to learn about it at all)

I'm currently struggling quite hard in my DIffGeo course without this background material

JamalS

Ah, I actually have a diff geom question for you Danu

Danu

Calc I-III may be OK in terms of analysis. Do you know the formal statement of the inverse function theorem, for instance?

JamalS

20:34

With a metric connection, it's defined such that it is compatible with the metric structure, i.e. parallel transporting two tangent vectors preserves the inner product. Now, if I choose it to be compatible with the structure of a Lie group G, what does that mean?

Danu

@JamalS no idea.

In my program, Riemannian geometry only comes after a full semester of DiffGeo

Kyle Kanos

@Danu For some function $f(a)=b$ we have $f^{-1}(b)=a$ if both are differentiably continuous & something else

JamalS

e.g. the electromagnetic potential is a U(1)-connection

Ok

Danu

@KyleKanos hmmkay, as long as you have some idea, and know where to look it up you may be OK

But you will definitely need some topology (compactness, Hausdorff, second countability are all very important)

Kyle Kanos

For GR or DiffGeom or both?

Danu

20:36

DiffGeo

not at all for GR

JamalS

Well, it depends how in depth you want to go in GR

Danu

Not really, it depends on which direction in GR

JamalS

Yes, that's better put.

Danu

you can go all the way as far as cosmology is concerned 100% without hardcore math

JamalS

Touché

Danu

20:37

Maybe you need some DiffGeo for really understanding the precise statement of some singularity theorems

but even that, I'm not sure of

...but DiffGeo is pretty cool, and worth learning IMO

it does give some very very neat insights

even something as basic (in that context, at least) as the idea that a vector should be identified with a derivative blows my mind

JamalS

^yes, that was not at all obvious to me

One of my favorite results is still the Chern-Gauss-Bonnet theorem

Danu

that smells of topology

instant dislike grows inside of me

JamalS

haha

The left hand side is differential geometry

And the right hand side is topology

Which is partially why it's a great result

Well, it's getting late... I need to be up early for a flight to London... Wish me luck with my uni interview :)

Danu

20:53

GL GL

you'll be fine

try to make the impression that you're also a chill dude

I don't know whether you are

but you should be, it really helps in life

don't be that angry nerd person ;)

JamalS

Ah, that's what my teacher told me today - 'Jamal, just relax.'

So, what do you mean by 'chill dude'?

So I can act 'chilled' :) lol

Danu

don't be the angry nerd type

hope you know what I mean :P

if not, you may be the angry nerd type

:D

JamalS

Agh, not really

Yikes!

What is it?

Danu

lmao

JamalS

Don't leave me hanging!

Danu

20:56

just messing with you

Try to be social and stuff

t's important to get ahead in life

you've got the academics down

dont worry about that

JamalS

My Weschler test indicated I was below average socially

Danu

well, that usually happens with people like you who work disproportionally hard on their academics :P

it's important to overcome that

JamalS

Isn't that what's required?

To maximise success in that field?

Danu

what?

JamalS

To be the best I can be, I have to maximise the time I spend, no?

Danu

20:59

No

JamalS

Well, I'm still young, I'll probably change in the future

Danu

to be the best you can be, you have to enjoy your life. Get rid of these naive external morale-type things. You're the one who has to live your life, so make sure you love it. Don't try chasing that dream if you're doing it because of some external motivation

Kyle Kanos

10,000 hours

Danu

@JamalS I'm just warning you to not become that guy in social isolation because all he does is work work work

@KyleKanos the greatest bullshit ever heard. Also, I think Jamal is already more than halfway there

@JamalS, In the end, you have to let go of 'being the best' is what I'm trying to say, I guess

Kyle Kanos

While the number might be wrong, I think the intent is correct: if you want to be a master at something, you need to spend a ton of time doing it

JamalS

21:03

@KyleKanos: You can't 'master' physics

Danu

@KyleKanos The intent is to fix a number, and that's ridiculous

JamalS

^

Kyle Kanos

@JamalS Of course not, but you can master your subfield

Danu

We've all spent several thousands of hours, I think

This is not the point (anymore)

JamalS

Hours well spent

Kyle Kanos

21:04

Several thousand sounds about right

Danu

My advice is also specific to Jamal (we talked quite a bit outside of SE)

make sure you have friends. GOOD friends. Ones you can really relate to, and that can really help you on a personal level

Really, you're so far ahead in academics right now that you can afford to invest some more time in being social

and if you don't do it now, it'll only become harder and harder to 'be fun' (I understand that fun is a subjective notion, but I"m talking about the stereotypical notion of social, 'cool', etc)

JamalS

Oh god, my MIT interview may have gone badly then... All I talked about was computer programming, mathematics and physics...

Danu

...and that non-academic part of life often turns out to be more fulfilling (and importnat, at least to me) than you may have thought

@JamalS don't worry about it. You'll get into a top school for sure

...and maybe the MIT people are psyched to see someone as dedicated as you

I'm trying to give you some 'life advice' that I learned the semi-hard way

JamalS

But how do you 'be social.' It doesn't come naturally to me.

Danu

An easy way to start is to start doing your homework together with other people in your program

If you find that you're way ahead of them, just spend time talking to them, maybe you can help them understand some problems - try not to be a showoff or drag overly advanced stuff in it though when you do this

I personally never do a problem set alone.

There's a group of 3-6 people that I work together with

and you get to know each other, talk about physics but also other stuff, etc. You become friends.

Kyle Kanos

21:11

@JamalS Find the largest local/University sports team (whatever sport it is) and start watching them play. One topic that appears regularly amongst my colleagues is sports

Danu

@KyleKanos watch out, this may be quite specific to the US

JamalS

@KyleKanos: Danu's advice I can definitely implement, but sports... Agh, I can't.... I feel sick thinking of it...

Danu

...although football (soccer) is usually OK, I guess in the UK people care about that stuff too

Kyle Kanos

@Danu Soccer with most EU sports fans?

Ice Boy

just be yourself

Danu

21:12

@IceBoy Not if that means total seclusion!

Kyle Kanos

Ever play Dungeons & Dragons (or other related dice-based role-playing game)?

JamalS

I like Chess

I play against some people at my school, during breaks

Danu

@KyleKanos this is also an option; I hear a lot of people like it in the physics scene (although I personally am not into that kinda thing, fantasy and all...)

JamalS

But of course we're in total silence for that...

Danu

@JamalS this is also an OK, though not GREAT way to make friends

Ice Boy

21:13

@JamalS who cares what people think

JamalS

No, I'm not into the fantasy thing either.

Danu

@IceBoy I would love to do chess-boxing heh :D

Ice Boy

you can do both

JamalS

That looks like fun... if it weren't painful...

Danu

@JamalS Anyways, just make sure you don't become lonely. It's really not that hard; as you'll notice, many people will be looking to make new friends

if you show some initiative to hang out/do stuff with people, it'll be relativetly easy

JamalS

21:15

I play the clarinet, I could join a band :)

Danu

@JamalS there's definitely uni-related orchestra's and stuff. Good idea

Ice Boy

my point is just be yourself, don't change because of peer preasure

Kyle Kanos

Don't be afraid to say "Yes" to the question, "Hey, we're gonna go check out this move X this weekend, you in?"

Unless it's a really bad movie

Danu

@KyleKanos no, ESPECIALLY if it is.

JamalS

Yeah, I'm really picky with my movies, and there haven't been any good ones lately :(

Kyle Kanos

21:17

@JamalS Note that you won't be going to watch the movie, you're going to be social :)

JamalS

Anyone here watched Stanley Kubrick's films?

Danu

@JamalS Doesn't really matter, you can have fun watching bad movies beforehand and afterwards, and even during

@KyleKanos exactly

@JamalS sure!

Kyle Kanos

Some of his films, not all of them

Danu

I always found Full Metal Jacket amazing

JamalS

I love 2001

Danu

21:18

so haunting

...although I'm not sure this will be the favorite movie of many freshmen :D Maybe they'd appreciate it though

JamalS

Not so keen on the science fiction, but on the underlying metaphors, symbolism, etc.

Have you seen 'Who's Afraid of Virginia Woolf'?

It was made in 1966

One of my favorite films

Ice Boy

classic

I'm into classics too.

music, not so much...

JamalS

I do like classical music

There's a music group where I live, and they hire musicians from around Europe to perform in a monthly meetup

Ice Boy

Join an orchestra

JamalS

But the majority are old people (at least 75)

Ice Boy

21:21

so

JamalS

I like going regardless for the music

Ice Boy

age means nothing

a friend is a friend

JamalS

No, I know, but I stand out

Danu

@IceBoy Sorry, but this is too 'easy'.

@JamalS Orchestra's are a good thing, as I already mentioned.

Kyle Kanos

My wife and I are around 30, most of our friends are around 50

JamalS

21:23

I love Jazz too

Ice Boy

cool

JamalS

Benny Goodman, Duke Elington

Ice Boy

classics

JamalS

Speaking of music, @Danu, I assume you've seen: youtube.com/watch?v=BipvGD-LCjU?

2

The song's called 'finite simple group of order two' :)

(Of course, it describes a romance in terms of group theory.)

Danu

This is pretty funny.

Is there any way to set up a reminder for myself to post this on Valentine's day?

JamalS

21:28

I'll remind you :)

Kyle Kanos

@Danu There was an invention a long time ago called a "calendar" ;)

JamalS

What? What's that?!?

Oh, you mean a Google Calendar? :)

Kyle Kanos

Calendar

A calendar is a system of organizing days for social, religious, commercial or administrative purposes. This is done by giving names to periods of time, typically days, weeks, months, and years. A date is the designation of a single, specific day within such a system. Periods in a calendar (such as years and months) are usually, though not necessarily, synchronized with the cycle of the sun or the moon. Many civilizations and societies have devised a calendar, usually derived from other calendars on which they model their systems, suited to their particular needs. A calendar is also a physical...

Danu

Kyle Kanos

@Danu Aww, what an epic movie

JamalS

21:29

What movie is that?

Danu

Jumanji

haven't seen it - LOL

JamalS

Surprisingly, I've watched it, in Spanish class!

When I was 10

Danu

Oh shit! It's Mrs. Dunst

whoa, she looked like that only 20 years ago?

Kyle Kanos

Yep

she was also in Interview with a Vampire:

JamalS

That was a good movie, for a vampire movie...

Kyle Kanos

21:34

Back in 1994.

Danu

...but seriously guys, how do I set up something that'll alert me

Kyle Kanos

Google Calendar can send emails

JamalS

Can't you put a notification on your phone?

I can do it on my iPhone

Where are you going to post it?

Ice Boy

@Danu just go to your calender on the wall and write it on Feb 14

seriously :-)

Danu

@IceBoy Don't have one. Why would I?

JamalS

21:40

Tattoo it on your hand lol

Ice Boy

Orly? sorry.

I thought most people still use paper...

Danu

@IceBoy No apologies necessary, haha. I just don't see any use for a calendar - I've never actively used one and also don't use an agenda or anything like this, I just remember stuff.

JamalS

I've even abandoned paper, I have a whiteboard behind my desk now :)

Ice Boy

@Danu great vid for valentine's day...do you have a YouTube account you can link it there

Danu

@JamalS ... I take lecture notes for sure, on paper. When I do assignments, though, I TeX it.

JamalS

21:43

Sometimes I TeX my notes

Danu

I have 30 hours of classes per week, there's no way I can TeX that much.

JamalS

Yeah, good point

A friend of mine who is a mathematician uses something called 'tex mex'

Or something phonetically similar to that

It's a relaxed LaTeX, where what you type get's converted directly into what it would look like in the PDF

As if you're editing the PDF, rather than the source

Danu

LyX

JamalS

Hmm

He said 'tex mex', agh he has a bad memory anyway

Danu

I'm fast at LaTeX, it's not that

I just don't want to spend all of my time typing up shit

that I already wrote down

The way I do my problem sets is I never write anything down

I do all the manipulations in my head

and write as I think

JamalS

21:47

Out of curiosity, how's the level of your problem sets?

Are they easy?

Danu

Not to me

to you - maybe

Example: Prove that $\mathbb{C}\mathrm{P}^n$ is Hausdorff

Other example: Give a counterexample showing that $A\geq B\geq 0$ (where $A$ and $B$ are symmetric operators on a finite-dimensional Hilbert space) does not imply $A^2\geq B^2$.

Yet another example: Prove Kramers' theorem

JamalS

I know a proof of the first using exact sequences

Is that what you would use?

Or how do they want you to do it in your class?

Danu

no ring theory

only topology

the way I did it was using the homogeneous coordinates on projective space

(I'm not entirely sure this is standard terminology)

JamalS

Homogeneous coordinates, as in the point at infinity is mapped to a finite value?

Danu

no

It's sort of the canonical way of representing elements of the projective space by means of coordinates in the original space

JamalS

22:02

Can I see your proof?

Danu

Sure

JamalS

Hang on, let me launch Skype :)

I see it :)

Danu

It's really simple once you know how to do it

Ice Boy

OK guys, a gotta run, see ya later :D

JamalS

How long have you been taking the course Diferenzierbare Mannigfaltigkeiten?

Danu

22:06

3 weeks

the professor is quite horrible - his approach is a total WTF for most people in the lecture

JamalS

Based on your problem set solution, you seem to be doing fine :)

Transcript for