The h Bar

DanielSank

00:02

@Danu good, you?

T Abraham

@DanielSank u do realize he has left the chatroom... and went to sleep...

T Abraham

00:21

hey @alarge !

why does it seem that some people are trying to avoid me?

alarge

00:36

@TAbraham hey

T Abraham

@alarge I actually am busy... I wanted to talk but have to talk later.. maybe 1/30 afternoon (PST time)?

alarge

You can just write the messages, I will see the pings (@).

T Abraham

@alarge sure!

DanielSank

@TAbraham Yes.

I see chat as an asynchronous communication protocol.

7

dmckee

Wait. @TAbraham's here? I need to go ... uhm ... wash the cat. Yeah. And sort the screwdrivers again.

vzn

02:24

& the sock drawer! :p

David Z

04:27

@TAbraham perhaps because you have a natural tendency to assume that people are trying to avoid you?

infinitesimal

:O I never knew that Roger Penrose was sent one year back in grade school

DanielSank

08:28

@MarkMitchison: I finally wrote up my answer about quantum spectral density in relation to transition rates. Here it is.

Kyle Kanos

14:22

Well look at that:

32

Q: Site graduated! New design launched

As you can see the new design just went live. Which means this site has been officially launched! Congratulations! Thank you for your valuable design feedback. If you see any CSS/styling bugs, please start a new post and tag it with "design" and "bug." We have also updated site's Twitter profil...

Movies.SE graduated last night

Jim

Well good for them

Gaurav

@KyleKanos you a movie buff ?

tpg2114

That's a pretty cool design

Kyle Kanos

@Gaurav I do enjoy a lot of movies

That said, I'm not active on that site

Curiously, the guy who posted that (obviously a SE dev member) has 1 rep

Normally they have 101

ACuriousMind

@KyleKanos Only because they normally have the association bonus

That guy has no site with enough rep to get it, so he's stuck at 1

Kyle Kanos

14:34

@ACuriousMind Yeah, which means this guy is brand-new

0

Q: Loosening a brass screw from an aluminium hole

I'm trying (desperately) to remove a screw from my espresso machine but it won't budge. The screw is a dark yellowish colour making me think it's brass (or maybe bronze). I've been told that the rest of the metal in the unit is mostly aluminium. I know that metals contract and expand and diffe...

thermodynamics everyday-life metals

Anyone think that Home Improvement might be the place for this question?

Maybe Life Hacks?

tpg2114

Either one, but we don't migrate to beta sites

Kyle Kanos

Home Improvement isn't beta

tpg2114

Really? Learn something new

Jim

@tpg2114 In the last 90 days we've migrated 21 questions to various beta sites

tpg2114

It's not really on topic here anyway

Kyle Kanos

14:42

That's what I said in my comment

tpg2114

@Jim Per user request or by community/moderation decision?

Jim

both but mostly moderation decisions

Gaurav

@KyleKanos oh, never knew that devs start with 101 rep... but I see that he's a product designer for the SE network. So maybe he's not a dev.

Anyways, I just figured that I still haven't understood online chat room etiquette as opposed to face-to-face conversation. In a face-to-face chat, we normally finish a chat on a particular topic in a single sitting. When I join chat, I normally do so when I have five minute breaks or so. When a conversation stretches on beyond that time limit, a confusion brews in my mind whether to continue chatting so that the other person doesn't get insulted, o…

ACuriousMind

@Gaurav I think the heavy starring of DanielSank's "I see chat as an asynchronous communication protocol" indicates it is not considered rude at all.

Kyle Kanos

@Gaurav Devs, like everyone else, would start at 1 rep. Once you get 200 rep on one site, any new site that you register to gives an "association bonus" of 100 rep

I second ACuriousMind's quote of DanielSank's comment

Gaurav

15:01

@ACuriousMind Oh, okay. Most of the times I found it tough to apply the same etiquette as we do while conversing normally. For example, we can edit what we just said a few moments ago. In a real conv. that'd be really useful.

Kyle Kanos

You have like a 2 or 3 minute window to edit your comment

With an empty text box (where you normally write), just press the UP arrow to get to your last comment

Gaurav

Yeah, if you're on a date in a restaurant, that 2-3 minute window to edit your words will make a big difference ! :)

Kyle Kanos

15:16

Anyone know regular expressions?

tpg2114

How complicated?

I have gotten pretty good at googling for them... but I may have enough already written in various scripts to help you

Kyle Kanos

I've got a bunch of P1, P2, P3, ..., P7 variables that I want to replace as an array: P(1:7)

So in the file there's P1 + Y*(P2 + ....)

I'm trying to use regex to replace any instance of Pn with P(n)

And I don't know how to use the found digit in [P]\d in the replace function

Or if it's possible

(If it matters, I'm using the IDE geany for this and that, I think, uses Perl for regex)

tpg2114

I don't know perl at all. I would do multi-step though (which may not be the right way) -- find [P]\d, parse result to get n then replace Pn with P(n)

And maybe perl could do that as a 1-liner (cause I think every possible thing in perl can be written in 1 line)

I would assume you could nest it actually

Alternatively, since you know it's always Pn

You can find all Pn and insert a ( at position 2 and ) at position 3

Kyle Kanos

Okay, I just found that \0 returns the found item [P]\d in its entirety

So I guess I can insert that in, then search for \(P and replace that with (

Perfect!

tpg2114

Is this a thermodynamic or transport property curvefit?

Kyle Kanos

15:26

?

What I'm trying to replace here?

It's a Bessel function

tpg2114

Yeah

Oh, okay

Kyle Kanos

This guy thinks this code is Fortran 90

tpg2114

Our thermo curvefits (NASA or Chemkin specifications) are things like C0 + T*(C1 + T*(C2 + ...))

Kyle Kanos

Yeah, that is just a faster/more efficient way to do polynomial multiplication

tpg2114

Right. But it only crops up for us in thermo/transport

We don't do polynomial multiplication anywhere else

And that's definitely not F90

Kyle Kanos

15:29

Got it

@tpg2114 That's what I was saying!

I think dude just did a tbessi.f -> tbessi.f90 and called it F90

tpg2114

Not to mention it isn't even really right in f77 either

SQRT(FLOAT(IACC*N)) shoudl be DSQRT(DBLE()) if it's f77

Otherwise it drops precision

But yeah.. that was really just renaming the file and probably changing C to !

And adding & instead of spacing over 7 or whatever it used to be

Kyle Kanos

I think you need a character at column 6 for it to be a line continuation

tpg2114

Back when our code was f77 still, people in our lab used C or ! as a continuation character too

So it was completely unclear whether something was commented or not, or should be or not, especially when they would comment out lines in the middle of expressions by moving the C or ! out of column 6

Gaurav

Sorry to interrupt in your merry conversation about coding which I can't understand a bit. If I want to have a career in fundamental research these days, is knowing programming to a certain level a prequisite ?

tpg2114

How are you defining fundamental research?

Kyle Kanos

15:34

@tpg2114 Yeah, that would drive me insane. Even things like not following 1TB drives me nuts

tpg2114

I would say yes you do, no matter what field you work in, because you will almost always have some ODE/PDE to solve or some data to analyze. Even in highly theoretical work

@KyleKanos 1TB? What's that mean?

Kyle Kanos

@tpg2114 1TB = One True Brace. You do for(i=0; i<10; i++){ instead of the the { on the next line under for.

tpg2114

Ah, okay. I do that for for and if but not for functions

I don't know why, I prefer the function bracket to open on the next line

Or class/struct/etc

Anything other than for/if

select

etc

Gaurav

@tpg2114 I define it as science research opposed to engineering, or to do with activity of commercial nature.

Kyle Kanos

@Gaurav I concur with tpg, at least some knowledge of programming is necessary. Some people, like @Danu can get by without knowing any programming because of the field they are getting into.

@tpg2114 See now I'd say that's kinda weird and inconsistent.

tpg2114

15:38

@Gaurav Well, then there's people like me who do fundamental studies of turbulent combustion (with no commercial or engineering applications) and I do massively parallel computing

So if you're looking that broad of a definition, the answer ranges from "Very little" to "Intimate knowledge of hardware, compilers, operating systems, architectures, etc"

@KyleKanos I dunno, emacs will tab over the { after for and if but leaves it flush with the previous line for all the rest

So emacs seems to agree with me

Kyle Kanos

emacs is also highly customizable and will allow you to change that

So I can make emacs agree with me

tpg2114

Sure, but by default it thinks I'm right :P

Kyle Kanos

Maybe

Gaurav

I'm contemplating such a career after my undergrad studies. Since I'm really not sure which field I'll specialise in research, I just wanted to know what I need to know about programming in general so that I can better prepare.

tpg2114

@Gaurav If you've never programmed at all before, you could probably survive by learning something like Python and/or Matlab

That will get you through all of your undergrad work and prepare you for generic data analysis that would come up at the lower-end of the programming needs later

And will set up a decent foundation for more advanced programming if you do something that will need it

Fortran would also probably be good to learn for almost all fields out there

Kyle Kanos

15:43

^ Agreed!

tpg2114

We tried to get some C++ libraries installed on a cluster I use (run by the DoD) and they said that 95% of the codes run on their machine are Fortran and so they don't have much experience with C/C++

Gaurav

I'm actually good at C++ or java.. I heard they literally have no use in the research field.

tpg2114

And this is a machine in the top 10 of the top 500

Java has almost no use

Except as writing a GUI/utilities for real codes ;)

Kyle Kanos

Java is useful for web-applications. For research purposes, I don't know anyone using it

tpg2114

C++ can be useful, there are some codes in C++ but most scientists never learned it. They just know C or Fortran and are reluctant to learn anything new

Kyle Kanos

15:45

C++ is popular among the younger generation of people I know, those who are trying to get away from C

tpg2114

i want to say @dmckee has worked with large C++ code in his research

Kyle Kanos

Chris White works with C++ on Athena

tpg2114

I'm using it for my structural mechanics solver

Gaurav

:) c++ knowledge helps in python too ?

tpg2114

I can't say I would choose it again if I started from scratch...

Only in that you already know OOP concepts

Python isn't all that helpful for research for most things. It's a good Matlab replacement

And good for writing utilities to do stuff

I use it for all my visualization, initial conditions, etc. But I'd never run a solver for anything in it

Kyle Kanos

15:48

I'm not too familiar with reading/writing syntax like BoxLib ::BF_init::~BF_init () :(

Really the whole OOP paradigm is new to me, so I have to basically re-wire my brain if I get the finance job I (sorta) interviewed for on Tuesday

tpg2114

@KyleKanos It took me a few seconds to realize there was a frowny face at the end and that wasn't part of the syntax...

JamalS

C++ is a multi-paradigm language, you don't have to use OOP

Kyle Kanos

Python is great for its interface with LLNL's VisIt

tpg2114

^ And ParaView and MayaVi

And Blender

Kyle Kanos

@JamalS C++ w/o OOP = C code.

Gaurav

15:49

Yeah, I've learnt OOP concepts. Anyways, thanks to the both of you. :)

tpg2114

@KyleKanos Not true

JamalS

@KyleKanos No, I couldn't disagree more.

tpg2114

C++ is C compatible

But it really is it's own language

JamalS

@KyleKanos Take template metaprogramming for example, that's not OOP; is it C? :)

tpg2114

You can use templates without a drop of OOP

JamalS

15:50

^

tpg2114

Templates are awesome

But horribly confusing sometimes, especially to debug

JamalS

Of course, many C++ idioms do rely on OOP, e.g. resource acquisition is initialisation (RAII)

Kyle Kanos

That's not considered OOP?

(the template metaprogramming)

JamalS

@KyleKanos No, it's not.

tpg2114

Nope, it can be used without classes/structs

Kyle Kanos

15:51

Oh

tpg2114

You can template regular functions

Kyle Kanos

Well you learn something new every day

JamalS

@KyleKanos It seems you may have some preconceived ideas about C++, I would recommend looking at the 'modern C++' talks on YouTube

Kyle Kanos

Wonder if F03/08 has that...

tpg2114

We wrote a template metaprogramming thing with CPP macros to add templates to Fortran

Because F03/F08 doesn't have it

Kyle Kanos

15:51

@JamalS I don't program in C++, so my knowledge is rather superficial

I wonder if F15 will add that

JamalS

Fair enough

But yeah, if C++ minus OOP = C, then one is not writing C++ properly.

Kyle Kanos

I'm kinda familiar with newer programming concepts, but because I don't use them, I'm not as familiar as someone, like yourself, who has used it

tpg2114

I guess we'll find out in 2020 when F15 is released, and then play with it in 2040 when F15 is mostly supported by the compilers

3

JamalS

^ Unfortunately, even then, some companies may not use it since updating a codebase to a newer version is troublesome

tpg2114

Yeah, been there, done that

JamalS

15:54

Thankfully I'm not a programmer

tpg2114

It's even worse when the compilers are all at different stages of implementation

Kyle Kanos

Though, to be fair, the Cray is pretty much entirely compliant with the F08 standard: fortranwiki.org/fortran/show/Fortran+2008+status

JamalS

The C++ ISO committee try and communicate with compiler vendors and other people that have a vested interest to try and prevent some of these issues

tpg2114

The cray compiler also takes 96 hours to compile our code

Which makes it absolutely useless

Kyle Kanos

@tpg2114 !!! Serious!

tpg2114

15:55

Yup. It's disgustingly bad at what it does

Kyle Kanos

Is it fast when it finishes?

tpg2114

No different from Intel or GNU

Kyle Kanos

Well that is useless then

tpg2114

And I refuse to use PGI because they release their documentation once every year, in January, and in it they list all the things they plan on having available by year-end but don't make any mention that it may not actually be the current compiler version

And sometimes they actually release the feature in the compiler that year, sometimes not. But the documentation is only written once

Kyle Kanos

Interesting

Never used PGI

JamalS

15:58

Since Kyle recommended the Intel compiler I've been using that, and it's performing well thus far

tpg2114

Intel is killer

Kyle Kanos

The Palmetto Cluster has it, but I haven't tried it because that Cluster sucks

tpg2114

I love their compilers and their tools like VTune

The only reason I'd play with PGI is the whole CUDA-Fortran thing they have going on

JamalS

In addition to all that having a disassembler like IDA can be useful

tpg2114

But even that, I'd rather just use C

@JamalS VTune and IDB will show you the assembly for an executable

Kyle Kanos

16:00

10

Q: To what extent is generic and meta-programming using C++ templates useful in computational science?

The C++ language provides generic programming and metaprogramming through templates. These techniques have found their way into many large-scale scientific computing packages (e.g., MPQC, LAMMPS, CGAL, Trilinos). But what have they actually contributed to scientific computing in value that goes b...

performance c++ programming-paradigms

JamalS

@tpg2114 Do they have a graph view like IDA?

Kyle Kanos

The way I read that answer is that Fortran does sorta have TMP if you count the module procedure wrapper and writing the subroutines 3 (or more) times to account for the different types

JamalS

@KyleKanos Yes, that was the elephant in the room

But trust me C++ is more than OOP + C

tpg2114

@KyleKanos I completely disagree with the accepted answer. "Unusable" because the guy claims people aren't smart enough to figure it out

@KyleKanos That works, but it leads to code duplication and errors propagate

Kyle Kanos

@tpg2114 I cannot judge, but I do concur that "people are too dumb to use it" is a bad reason.

tpg2114

16:02

The module overloading is more like C++ function overloading

Than it is templates

Templates are (kinda) like duck-typing

Which you find in Python

Kyle Kanos

@tpg2114 Agreed, and it'd be neater to do it without resorting to types

tpg2114

A template will work with the exact same code so long as the objects support the operations you want to use

Combining templates with auto results in a lot of C++ code that looks/feels more like python

@JamalS I don't know, I've never used IDA

Or really played with the assembly features in the Intel tools, aside from me just reading the assembly

Kyle Kanos

I try not reading the assembly from my code since it's ~60k lines of Fortran, I cannot imagine how big it really would be

Though I've looked at smaller (~50 line) codes Assembly

JamalS

@KyleKanos Is your code public/opensource?

tpg2114

@KyleKanos Everybody needs some bedtime/bathroom reading

Kyle Kanos

16:08

@JamalS The magnetohydrodynamics code I use is semi-open source (you have to request it, but it is GPL'd). All of my codes are entirely open source (though I haven't really put many of them online)

@tpg2114 I have Liars Poker for that

tpg2114

@KyleKanos When do you find out about your semi-job interview?

JamalS

Oh, yeah, how did it go?

Kyle Kanos

In about 2 weeks they're going to get back to me, though the guy said I can "bother him" if he hasn't gotten back to me by then.

It went pretty well.

tpg2114

Where was the company?

JamalS

What did you get asked?

Kyle Kanos

16:11

@tpg2114 Charlotte NC

JamalS

Is it like a university interview? They give you problems?

tpg2114

The guy I used to share an office with just got a job with a company outside Charlotte doing aerodynamics for Indycar/NASCAR

Kyle Kanos

@JamalS Stuff about my future (why finance vs academia/industry), my knowledge of the finance field, things like that

There was no math involved

tpg2114

Knowledge of the financial field... I know I need money, I know I don't have money, and I know this job will give me money.

JamalS

I mentioned the mathematician from Cambridge that went insane at Goldman Sachs, right?

tpg2114

16:13

@JamalS To be fair, I'm an engineer at Georgia Tech and I'm going insane... still at Georgia Tech

Kyle Kanos

This guy is a bit higher up in the company. He feels that the strict "answer X number of questions on this phone interview" method that they currently use is wrong and should be a bit more relaxed. The issue he sees is, yes they get the smartest kid in the lot, but they're often terrible employees

@JamalS You have before

JamalS

How do they get the smartest if there are no math/science problems in those questions?

Kyle Kanos

@JamalS In the normal interview, there is. This wasn't a normal interview

JamalS

So you're most likely going to be doing that next?

tpg2114

I think he bypassed that round

And went above the normal HR interview crap

Kyle Kanos

16:16

^ there's that too

JamalS

Ooh, how did you bypass it?

tpg2114

@KyleKanos I know a bunch of people who got 4.0's here who couldn't get a job anywhere because companies would look at them and say "That's great, but you don't know how to do anything other than study or how to interact with people"

Kyle Kanos

But the guy I talked with basically feels that the aforementioned number X is too high and should be lowered to talk to more people.

@JamalS Nepotism. My brother in law works there.

JamalS

Ahhh

Kyle Kanos

@tpg2114 Pretty much. At Clemson here, we have a few courses in the Physics Ed department (which, for whatever reason, is separated from Physics) that help on the socialization aspect of the graduate life

JamalS

16:18

Why on earth would employers care about that (the social thing)?

tpg2114

@JamalS Was that sarcastic? (Can't tell on the internet)

JamalS

No...

Should it have been? oO

tpg2114

Because you can't put 20 people in a room and ask them to work together constructively if they have absolutely no social skills

3

And no business can operate by having individuals work alone

JamalS

I don't think you need great social skills to do that though.

tpg2114

At some point, the engineers/scientists need to talk to clients, or managers, or accountants, or HR

I disagree... Poor social skills lead to horrible dysfunction in business/industry

JamalS

16:20

What exactly are we talking about when we say 'social skills'?

tpg2114

Academia is about the only place where people can sometimes get away with it because you can be sitting alone, working at a desk all day

Kyle Kanos

It's probably important in academia too, since you have to deal with students in your classes

JamalS

If I can communicate a scientific idea to someone, and they to me, that should be it, no?

tpg2114

Ability to hold conversations, empathy, explain yourself to non-technical people, understand different viewpoints and cultures

JamalS

Different cultures? Why?

tpg2114

16:21

@JamalS I don't think so, because not everything you do in a job is scientific

JamalS

As long as you don't let it get in the way of work, it should be fine

tpg2114

@JamalS Because you have to understand people's motivations and backgrounds in order to work well with them

JamalS

So I need to know that employee X wanted to become a scientist because... to work with him?

And, one can achieve the same outcomes from having empathy through reason

tpg2114

How else can you decide who is best enabled to perform a task you have? If you have zero desire to program ever again, why would a manager want to assign you a tedious programming task?

You have to understand what motivates people if you want them to be motivated and productive

Mark Mitchison

@JamalS You need to be able to inspire confidence and a friendly atmosphere of collaboration, you need to be able to defuse potential arguments when you disagree, you have to make people enjoy working with you, etc.

user3183724

16:23

In the density matrix formalism, the off-diagonal elements represent the coherences, or superpositions. Say I have the density matrix of a two level system. Then the diagonal elements give the populations, but can the offdiagonal elements then tell me something about what kind of superposition I have? Like, equal superposition, or more in one state than the other, etc

Mark Mitchison

"And, one can achieve the same outcomes from having empathy through reason" <- This is a strong contender for the most incorrect statement I have seen written on the Internet.

Not joking

alarge

@JamalS What happens when you disagree? Even technical problems rarely only have one solution.

tpg2114

@JamalS I'd be curious if you posed a question Workplace.SE asking if the empathy/reasoning and scientific communication is all that is needed were true, what that response there would be

Look at our Meta -- you have people who are likely very intelligent but have virtually no social skills who are annoying as anything because they demand things be done their way, without recognizing that there are multiple ways to do things

JamalS

@MarkMitchison Most incorrect? Well, wouldn't something that's factually incorrect be the most?

Mark Mitchison

That is factually incorrect.

tpg2114

16:26

Those people completely corrupt a community, whether it's an internet site or a workplace

Mark Mitchison

Empirically, factually incorrect

Sofia

@KyleKanos Hi ! I see that you are here. Please have a look at this question . The OP has such a low level with trigonometry that he should better be sent to improve himself. I strongly suggest closing the question immediately, because some users may be tempted to solve his exercise, and that would simply a wrong thing to do. Can you intervene quickly?

tpg2114

@Sofia Nobody except a moderator can "intervene quickly"

All we (including Kyle) can do is vote to close

JamalS

@Sofia Voted to close.

Kyle Kanos

It was on the review queue and I already did that (before you came in, I mena)

tpg2114

16:27

Which I did when I saw it posted

JamalS

@MarkMitchison Regarding my sentence, maybe it can't always be done through reason, but surely sometimes?

Mark Mitchison

Sure, sometimes. That's just not enough. People are not perfectly rational beings. Not even scientists. Not even you.

Sofia

@KyleKanos : how many people voted already? To solve the question for him, despite all my good will, would be a mistake.

JamalS

No, we can only strive to be :)

Kyle Kanos

@Sofia It's now closed

So 5

Sofia

16:29

@KyleKanos : Bravo! Very well !

tpg2114

@JamalS I think what it boils down to is no job anymore is a solo endeavor. Things will be multi-disciplenary with people from all different domains, skills, cultures, etc.. What is reasonable to one may be totally confusing to another. It would be awesome if all disputes could be settled reasonably, but then we wouldn't need courts or laws. It would be great if all that was needed was to communicate scientific content but things aren't always scientific. And people are still people

The ability to identify when somebody is overworked, tired, stressed out, having a crappy day, etc. is really important in a workplace. It's important to understand how people are feeling and what motivates them in order to work together efficiently and happily

And there's a lot of very brilliant people who can't do any of those things. And they don't make good employees.

JamalS

@tpg2114 But that's just it, you don't need to feel sorry or identify with an overworked, stressed or tired person to know you shouldn't assign them important work that may be done badly because of that.

tpg2114

@JamalS In general that's true. But what if a 5 minute conversation where you show empathy (which is different from sympathy) could make them feel better and be productive? What if you have something that absolutely must get done and you need to assign it to them anyway? It takes finesse to do that, and it needs to be approached from a position of understanding

Somebody who is a sociopath may be able to fake the empathy enough to get through without actually having any empathy, but they are likely bad employees for other reasons.

The ones who can't fake it or make no attempt, they don't make good employees either

JamalS

@tpg2114 So I should sacrifice my time to make them feel better?

Well, I guess if in the long run it makes them more productive, it could be worth it

tpg2114

If your project and your team and your performance depends on it? Absolutely

Otherwise you're not being a good employee

Really, it could be a 30 minute conversation at 9am that means the person will actually get everything they need to done that day

As opposed to getting little to none of it done because something is wrong

JamalS

16:38

30 minutes?!?

Yikes!

tpg2114

Coffee break, they go by fast

Kyle Kanos

@JamalS: You've been in the room conversing the last 30 minutes

(obviously typing is different than speaking)

JamalS

I'm multitasking, I'm having a French lesson at the moment

We're discussing this in French :)

tpg2114

^ But you can have these conversations online too (if that works for somebody, that's what you would need to know)

JamalS

(It's 5pm here.)

tpg2114

16:41

French huh? I guess if you were in France, spending 30 minutes of your workday would be a giant chunk of the 30 hours a week you're allowed to work

And it would get in the way of the 2 hour lunches, the entire month of July off, and the favorite past time of burning cars and rioting while on strike ;)

@KyleKanos So if you interviewed, are you going to defend soon?

Kyle Kanos

I'm aiming for end of Feb, beginning of March

user3183724

16:59

If we have a two level system with a ground state and an excited state, and the energy difference is omega_eg, can you then excite it from the groundstate into the excited state using a laser with frequency w_c >> w_eg? If so, what happens to the left over energy? This is of course assuming there are no higher lying levels

I suppose I can answer it myself, if it is indeed true that for such a two level system the absorption spectrum is equal to the emission spectrum

Mark Mitchison

17:19

Yes it is possible to excite off-resonance.

Thinking about photons in a laser is usually a recipe for confusion.

Actually the state (and therefore the energy) of the laser is unchanged by the absorption of a photon.

Also, from the point of view of the atom, which sees a time-dependent driving field, there is no energy conservation law

The confusion is resolved when you think about the energy contained in the global Hamiltonian: Field + atom + interaction.

This is of course always conserved. But because of the interaction you cannot split the energy neatly into electronic energy levels + photons

This question has been answered on this site before, I suggest you try searching.

Danu

17:38

Hey guys, is it really correct that the (total) cross section for electron positron annihilation already diverges at tree level?

Danu

17:48

*in the ultrarelativistic limit

Kyle Kanos

18:09

Gamblers make no sense: espn.go.com/chalk/story/_/id/12246876/…

DanielSank

PSA: You can put \newcommand{...} at the top of a Physics.SE answer to define things like kets and save yourself from typing things like |\Psi\rangle over and over :D

2

Kyle Kanos

Someone put a $10 bet that the Seahawks-Patriots game on Sunday would end 4-2

DanielSank

o_O

Kyle Kanos

Apparently no team has ever scored just 4 points in American football

For those unawares of American football: 6 points are scored on a touchdown, plus either a point-after attempt or a two-point conversion; 3 points are scored on a field goal; 2 points are scored on a safety, in which the ball carrier is tackled in their own end-zone

So that $10 bettor is saying that the Seattle team will tackle the ball carrier twice in the endzone and the New England team will do it once and no other offensive scoring will take place in the 60 minute game

o.O

DanielSank

Sounds like a "for lulz" bet.

Kyle Kanos

18:15

Payout is good, the odds were put at 5000:1

And apparently the color of the hoodie that Bill Belichick (coach of the New England Patriots) is up for bets as well

Gamblers continue to amaze me at the random things they'll bet on

tpg2114

18:58

If I had $10 to spend on something other than a lovely beverage at the bar, 5000:1 odds might be worth it

Almost infinitely better odds than lottery tickets

Kyle Kanos

It could be worth it

tpg2114

That'd be a fun(ny) game to watch too

glance

@Sofia sorry, I don't get the point of this edit. Isn't it a little "useless" to edit the post for such a thing (which arguably makes the post better btw)?

tpg2114

@glance I would agree, that's too trivial of an edit to make

Especially since there are a few other things to edit in the post at the same time

glance

@tpg2114 exactly what I was going to say

Kyle Kanos

19:10

Since Pete Carroll became coach of the Seahawks, he's had losses where his team has scored 0 (1x), 3 (3x), and 6 (2x). In that same span, the Patriots have had only one game under 10 points: a 6 point game vs the Bengals in 2013

vzn

20:14

hey KK. sounds interesting, is the job in computational finance? ie "quant"? have studied that field a lot. theres an se site for that, not too busy it seems

lots of links on high freq trading (TMachine blog) which is very algorithmic & has been in the news...

Kyle Kanos

Yes, it would be a quant job

vzn

stocks? (trading)

Kyle Kanos

I've looked at Quant.SE and have thought about joining a few times

But haven't yet

vzn

re that book liars poker. the author recently came out with a book anti-high frequency trading.

iirc...

Kyle Kanos

It'd be capital markets, for the most part

Sofia

20:17

@glance : I'll tell you. When looking at the list of question, the title of the question, *Why is probability proportional to $e^{E/kt}$ looked ugly, with the "e" of the exponential immediately following the "o" of the "to", with no space between them. So, I placed a small space.

Kyle Kanos

@vzn I think everyone in "the know," so to speak, are anti-high frequency trading

vzn

ok, back l8r gotta eat

its a fascinating area... a good documentary out now on that page. they interviewed semifamous blogger "mathbabe"... etc...

Sofia

@glance But, some two days ago you sent me a comment about answering my own question. Which question? Do you have the possibility to repeat the comment and indicate the question?

Kyle Kanos

20:34

physics.stackexchange.com/questions/119034/…

(requires 10k rep)

But for those that can: huh?

ACuriousMind

lol

alarge

@vzn I don't think there's really a problem with algorithmic trading, but it's specifically HFT that is sometimes troubling. Having read Lewis' book, the biggest issue he raised was front running, not so much algorithmic complexity (e.g. the flash crash, although he did mention it too). I didn't see this aspect mentioned on the blog post.

tpg2114

Hey mods -- how did this get migrated here when it's so clearly off-topic?

-1

Q: A tank of the shape of a right circular cylinder $5$ feet across the top and $9$ feet deep is full of water.

A tank of the shape of a right circular cylinder $5$ feet across the top and $9$ feet deep is full of water. How much work is done by pumping the water out of the tank, over the top edge? I need your kind help. I have no idea even to start it. Please help me.

homework-and-exercises work

Doesn't seem right/useful to move a question for it to be immediately placed on hold

@KyleKanos Way to make us <10k'ers jealous

Kyle Kanos

@tpg2114 It's a particular user calling someone out for flagging a non-answer for deletion saying, Do you see? <user> has soon so many reputation, that he has an administrative GUI, where he can put this message to you with a single click. :-)

tpg2114

20:49

Sounds pretty gibberish.

And I can probably figure out who the commenting user is

Kyle Kanos

Yeah

And it's hysterical that the guy who flagged it for migration was basically told: Don't. Migrate. Crap.

tpg2114

Yup. The Math.SE folks knew it was off-topic here but it ended up anyway

I guess it's possible to be made on-topic here, and that's why it migrated

Kyle Kanos

Lol

6

Q: Did the migration rules change?

A few minutes ago, I cast the fifth close-vote on this question. It had three votes for "belongs on another site", and one for "This question is not about mathematics, within the scope defined in the help center". Since I know that "Physics.SE is less fond of homework dumps than Mathematics.SE" a...

support migration

tpg2114

Wait, so 3:2 to migrate will move it over whether the receiving site wants it or not?

Assuming the pipeline is available in the VtC/Migration options?

Transcript for