The h Bar

alarge

00:02

@TanMath Just running the initialization phase gives you a warning on trap and even if it didn't, I don't think the assignment does what you think it does. But you have been told this before, so I am guessing that you think that that is not the source of the unexpected output.

TanMath

@alarge oh no...that shouldn't be there...it is an old version...still has a problem i think...

@DanielSank i didn't understand...

what am i to test? there isn't any functions to test...

Emilio Pisanty

00:29

@TanMath Can't be. Profiles are still the old version.

But it does seem to be coming on line slowly.

DanielSank

@TanMath Either give me access to the code via a github repository, or write yourself a simpler example using a smaller dimensional Hilbert space, like maybe a two level system.

TanMath

56 mins ago, by TanMath

Correction - "almost converted":

DanielSank

@TanMath Well ffs put up the same code you're actually using.

TanMath

@DanielSank the gist wasn't enough?

@DanielSank huh?

DanielSank

@TanMath It's fine but then you said to @alarge that it's an old version.

Emilio Pisanty

00:31

@TanMath I don't really care that much about the redesign

but get me my new profile!

TanMath

@EmilioPisanty it is almost there...observe...

Emilio Pisanty

Also fonts that don't get into a fight with the MathJax on every other page

DanielSank

I don't want to spend time helping you on a version of the code that's not what you're using.

TanMath

@DanielSank yeah, i edited the gist...i just needed to change the int to complex...

@DanielSank and i do not want help for a code that is an old version too!

TanMath

01:11

@DanielSank you got the code?

ACuriousMind

01:58

0

Q: What's the force required to cause a 5lb object to go airborn and fly 20ft?

Trying to figure out the speed of a car from the moving free body within. Anyone care to help a cat out? i'll run the math if you've got the formula rotational movement of the car in reverse (3pt turn fail and hit an immovable object squarely with the car) Laptop dug into the passenger pota...

free-body-diagram

Um

TanMath

02:18

@DanielSank everything went well with the code?

0celo7

02:30

@ACuriousMind poor guy

0537

02:48

@ACuriousMind What did I read...

Kyle Kanos

@ACuriousMind More curiously: how did that get not one but two upvotes.

GBeau

03:05

no response yet for my request for a refund for the "nonrefundable application fee" for the graduate school

the graduate department inquiry email might just be swamped with emails, though :P

Kyle Kanos

If I ran a grad program that stated that an application fee was nonrefundable and someone asked for a refund, I'd probably just stick it in the trash

GBeau

@KyleKanos they double charged my card

(I stated this up front, and in the title)

only asking for the second charge to be refunded

Kyle Kanos

Well that changes things slightly

GBeau

I guess we'll found out how nonrefundable it really is

Kyle Kanos

I guess that changes things, but I'd be putting that sort of thing on the back burner or pawning it off to someone else

GBeau

03:08

It wouldn't be as bad if the application fees weren't ~$100/school

Kyle Kanos

Of course, worst case scenario is you just dispute the charge on the card

dmckee

@GBeau Apparently there is a population of no-hopers that apply widely every year, and serious school like to keep the slush pile down

GBeau

@dmckee "but I made a 410 on the PGRE"

Kyle Kanos

I thought I was going to be a no-hoper, but I got in

Then they gave me the doctorate

GBeau

I wish the school that double charged me wasn't also one of my top choices...

dmckee

03:11

@KyleKanos You have to see the application I mean to disbelieve them. Think about what a crackpot post looks like on Physics, and imagine what they would write on an application...

@GBeau Sure, but they don't know who the kooks are. They figure that serious candidates can scape it up for the one year they need.

Kyle Kanos

@dmckee Hmm, I suppose that's true.

dmckee

It's not like the people on the admissions committee see the money or anything.

Kyle Kanos

Does Duffy have a doctorate?

dmckee

@KyleKanos He doesn't admit to having anything, does he? But it doesn't matter as such.

GBeau

@dmckee Well, it does make me hesitant to attempt to, say, dispute the charge

Lest the graduate school inform the physics department my application is no longer to be under consideration...

Kyle Kanos

03:14

@dmckee I mean it as he's fairly crack-pottish, so I'm just wondering how he managed to eke through the process

dmckee

@GBeau No, dispute the second charge. You didn't get two bites of the apple. And if they object to that you don't want to be indentured to them anyway.

@KyleKanos Well, crackpottery can appear after admission. Or even graduation.

Kyle Kanos

Hmm

dmckee

Which is the SMBC that gets posted here so often?

Kyle Kanos

I wonder if I'll go crack-pot since I'm out of the business now

dmckee

smbc-comics.com/?id=2556

Kyle Kanos

03:16

Yeah, that's a good one

dmckee

@KyleKanos No reason to, but it helps to have something else fulfilling in your life. Like you work? Got a person of the apposite sex around?

Kyle Kanos

I love my job & co-workers. I have a wife & 4 kids

dmckee

Should be safe. Until empty nest syndrome hits home, anyway.

Mike Miller

Well, it took Penrose some time to go crackpot. Be patient is all.

Kyle Kanos

The youngest are 2.5 now. Wife's around the time that she's wanting another new baby

dmckee

03:17

@MikeMiller ::chuckles::

GBeau

@dmckee oh my

Kyle Kanos

Ouch:

12

Q: What are the implications of the whole class failing an exam?

I am an MSc student and in my class are 10 people. The module has been taught exceptionally poorly and as a result none of us can see ourselves passing the module. The exam is in two days time. By failing the exam we will have to retake it in the summer because we will not met the threshold to ...

teaching grades lecture-teaching-method

dmckee

@KyleKanos Most teacher have given overly difficult exams and have had poorly taught units. They call for the instructor to make a midstream correction and shouldn't call for anything in particular from the students.

There is an uncharitable side of me harboring deep suspicions that the problem here may lay partly with the students. But I don't have anything like actual evidence of that.

Kyle Kanos

I've never run into the issue OP has, but it seems to me that a good teacher would be able to change course in the middle of the semester

dmckee

The hardest thing about backing and filling is deciding what you're going to shorten or give up later in the course. The earlier you catch the problem the more flexibility you have.

Kyle Kanos

03:29

I can believe that

21 to go....

GBeau

I wonder if I'll have to teach something my first semester of graduate school (like a lab or something)

Kyle Kanos

@GBeau That is very typical of American graduate programs

dmckee

@GBeau Usually you get to TA a lab or two. That is only partially independent, but it is a good place to start developing the skills and coping mechanisms.

Kyle Kanos

My advice: if being a professor is what you want: find the most recent TA of the Year in the department & ask to sit in on how they do it

lol....*Do not attempt. You are not Bo.*

AT&T Commercial 2016 Bo Jackson Strongest of All Time College Football

►

dmckee

03:45

@KyleKanos My advice is to make sure you have a back-up plan and then go for it with all your heart. But you should have a Plan B. A real PLan B that doesn't involve freeway-flying adjunct work as that is no life.

Kyle Kanos

There's also that

I Plan A'd non-academic career and (thankfully) got it

Fortunately there was a research position available to me if the current job hadn't pulled through

GBeau

How competitive are postdoctoral research positions for graduating PhD students in good rapport?

Or does it vary pretty widely by field?

Kyle Kanos

In astronomy/astrophysics, they were good in the recent past. Not sure how it looks going forwards

What field are you considering?

GBeau

high energy experiment

with secondary choices in cosmology and biophysics

but HEP-exp was my first choice in all applications, and clearly indicated as such

0celo7

is it normal for an online thing to ask for my birthday when entering a credit card

Kyle Kanos

03:54

@0celo7 I've never been asked

@GBeau The best particle accelerators are in space

@0celo7 PS: sorry for the name thing

0celo7

@KyleKanos what

Kyle Kanos

Well NM then

My alma mater is up 24-21 over Alabama. Hopefully they can keep the lead this time....

GBeau

@KyleKanos Did you attend Clemson?

well then

Kyle Kanos

6 years for grad school

GBeau

that was faster than I thought

Kyle Kanos

04:04

Never met a player on the football team in my lab classes, but I've met a few soccer, softball & baseball players

GBeau

cool

Is recognition for ability as a TA in graduate school an important factor when applying for academic positions after your PhD?

Kyle Kanos

I can't answer that Q, I don't have an academic position ;)

Also, @ACuriousMind: you suck.

(though congrats on the Suggested Edits badge)

GBeau

04:39

@KyleKanos I have some live betting odds pulled up, and they suddenly shifted

uhoh :(

Kyle Kanos

You can still gamble at this point?

I thought they'd shut down gambling before the game starts

GBeau

@KyleKanos I'm not sure actually, I just always pull it up because they usually have live-odds that get updated

I usually just assumed it was from betting

Kyle Kanos

Hmm, well they definitely hit the over

Oooohh... odds of winning (not gambiling odds) are based on stats

GBeau

probably more likely

Clemson just jumped from 11 to 24%

Kyle Kanos

Hmm, they're down by 4 with less than 10 minutes of game time left

Unless the odds are delayed by some amount, not sure why it'd suddenly jump

GBeau

04:45

field goal?

Kyle Kanos

Clemson has 51 straight wins when holding the lead in the 4th quarter

So maybe it's based on that?

GBeau

\o/

@KyleKanos What types of software do computational astrophysicists use

Kyle Kanos

And Alabama just ran a kickoff return for TD

GBeau

For example, my undergraduate thesis is on a multivariate study with the ATLAS experiment, and we get ROOT (lovingly) shoved down our throat

Kyle Kanos

@GBeau Fortran or C or C++

GBeau

04:47

@KyleKanos I mean software packages (eg, libraries, etc)

Kyle Kanos

Then I guess it depends on what you're doing.

GBeau

are there big ones like ROOT that you use all the time?

Kyle Kanos

Root, Geant4, and Python modules are what I've seen...I'm sure there's more

GBeau

it says Alabama 96% chance

Kyle Kanos

I never used libraries

Just pure Fortran

GBeau

04:49

98% chance after the extra point :<

Kyle Kanos

Yeah, disappointing to say the least

Still 7.5 minutes left

GBeau

@KyleKanos 99% :\

@KyleKanos my school had a 1% win chance at half time earlier this year

now that's a bad feeling

Kyle Kanos

Motivation in the 2nd half: win the 2nd half, it's like winning half the game

Clemson scores a TD

GBeau

16% chance

will probably change after extra point

Kyle Kanos

Nope, went for 2 point and failed

GBeau

04:56

they go for 2 here?

yeah

I was about to say, I think it's better to go for 2

Kyle Kanos

Going for 2 would make it a 3 point game

Still sits at 4 now

GBeau

wait wht

I think the odds are broken?

clemson win chance 59%

Kyle Kanos

Force a 3 and out, takes about 2 minutes off with 2 minutes to win it

GBeau

oh no, it got fixed

alabama win chance is 83%

weird bug

I have to say, you've improved significantly from 1% :D

Kyle Kanos

Lol...the lineman who blew his assignment gave a frustrated sigh when the guy made the play

Well that sucks

30 yard gain, Alabama now in the redzone

GBeau

05:02

97% win chance alabama

Kyle Kanos

30 seconds and it'll be 99

GBeau

99%

good guess

Kyle Kanos

I'd love to see a Pick 6, but they're not going to pass right now

GBeau

I think this odds predictor as a max of 99

supremem...what is wrong with me

my uncle was telling me about his game over Thanksgiving

he has a national championship ring in football from the 80s

sometimes interesting to hear the perspective

In what scenario does Clemson still win this?

Kyle Kanos

Please fumble

GBeau

05:08

Oh, right

Kyle Kanos

None

GBeau

well

the max is definitely 99 because it didn't change

sorry ;(

Kyle Kanos

I've lost a $1, not a big deal

GBeau

the scenario where clemson wins is a touchdown, two point conversion, then a safety or something?

oh wait, forgot the extra point

so you'd need a field goal so there's not enough time :<

Kyle Kanos

They need 12 points, so a TD + 2 Pt + FG would do that

GBeau

05:11

not a bad game though if it stayed interesting into the 4th

Kyle Kanos

But they'd also have to get an onsides kick

Putting up 40 on Alabama is damn good though

GBeau

they didn't go for 2?

I have some delay

Kyle Kanos

Wastes time

GBeau

o

Kyle Kanos

Miss, you need 6, make it you need 4

Either way, you need a TD

Take the PAT, run an onsides kick and hope you recover

GBeau

05:17

yeah I was thinking they needed 3 if they made it

but I was wrong, adding is hard

Kyle Kanos

Alabama ball

GBeau

in principle, they can't screw this up

on paper

Kyle Kanos

AFAIK, no one has fumbled a kneel down in victory formation

Oh well

Bed time now

GBeau

night

sol0invictus

05:46

People did you'all read the Grav-Waves rumors

Nogueira

I see in the motl's blog

sol0invictus

aye,

Slereah

06:02

What rumors

Did we detect a thing maybe

DanielSank

@JohnRennie I think it was you who asked for that special relativity illustration program.

I'ts on github now.

github.com/DanielSank/physics-se

@TanMath ^

Oh, and @TanMath, that's how you write python code.

dmckee

06:52

@GBeau Being a good teacher as a TA helps if you applying to a teaching school. If you're trying to secure a research based position then no; not really.

John Rennie

@DanielSank Thanks. There were a few of us who expressed an interest. My interest was mainly to see how it was done as I've never used Python for scientific work.

Installing numpy seems a bit of a hassle. I'll have a look at the Python builds that have it built in.

TanMath

07:11

@DanielSank huh?

@DanielSank i never felt the need to write like that since I never thought i needed to share my code...

Plus, my code does not have functions...

and i still do not get the purpose of the spacing...

@DanielSank so what will you be doing with the code for my project? also, what am I supposed to write?

some testcase?

DanielSank

@JohnRennie Woah woah no no no don't do that.

I can offer some advice about how to use python and numpy to achieve nirvanna.

Other methods will lead to you Bad Things.

@TanMath Yes I noticed that. That's not so good.

@JohnRennie what platform do you use?

matplotlib and numpy are the two "hardest" popular libraries to install, but neither is actually hard.

If you're on Linux or Mac, definitely use python's virtual environments for all personal work. Your system uses python for its inner workings, so you really don't want to mess with the system python's packages.

Virtual environments allow you to completely separate groups of installed packages for python.

Basically, you want one virtual environment per projects. In your case, since you're getting started, I'd recommend setting up one called "main" or something like that, for general use.

My specific advice depends a bit on platform. I await your response.

TanMath

07:40

i used a Linux VM to install numpy, scipy, matplotlib, and also qutip...it doesn't reslly go well on Windows!

@DanielSank but there isn't much to put in a function!

@DanielSank so there isn't anything you want me to do?

Secret

Sometimes when shopping, we came across products like ionic hair dryer, ionic hair shampoo etc.

Is there any scientific basis on how they work (I cannot find a wikipedia article on that topic)?

user54412

08:04

@GBeau Only if you're applying to positions that will involve teaching. For instance, postdocs never teach in astro, so no one cares how well we do it. On the other hand, mathematicians do teach as research postdocs, so it can be important.

DanielSank

@TanMath Did you read my special relativity code? I could have put it all in one big chunk. I didn't, because that makes the code hard to understand and hard to maintain.

It also makes the code less useful because it's harder to re-use the pieces.

I suggest you invest some time in reading a lot of code written by other people who know what they're doing. I think you will learn a lot and improve your life.

user54412

@GBeau Of course, if you do a non-teaching postdoc and then apply for a faculty position, it would be good to convince them you don't suck at teaching. Mentoring grads/undergrads and doing volunteer teaching can go a long way in this.

DanielSank

@ChrisWhite Didn't know that.

Help me convince @TanMath that writing modular code with test cases and adhering to style guidelines are good things. He seems determined to avoid these things and then complain that his code doesn't work.

user54412

@DanielSank It's weird, but I guess there's just not enough demand. The physics department gets saddled with all the intro courses, and everyone forgets that astrophysicists can probably handle blocks sliding down inclined planes too.

DanielSank

Haha, right.

user54412

08:08

I might not say much, but I have been reading the chat transcript... ;)

DanielSank

@TanMath Ack! No! First of all, a VM just to run python is major overkill.

Second, on Windows it's perfectly easy. You just use pip to install everything.

Where do you get off saying that it's hard on Windows?

pip is the standard tool for installing python packages.

user54412

I once ran a workshop for the undergrads in the department where I got them all to install python/matplotlib, tex, and a real text editor. I felt like I did something good and productive that day.

Bass

@0celo7 I don't mind pedantics (have learnt a great deal from them). But I doubt that my (1) from Nakahara has certain requirements for the metric. Please find a counter-example.

John Rennie

08:31

@DanielSank I use Win 7 x64. In the few moments when I'm not trolling the PhysicsSE I pay my mortgage by working as a computer nerd so I have loads of test PCs around.

I'll reimage a test VM and put ona copy of Anaconda or something similar.

I'm not tampering with my main workstation - that path leads to madness :-)

0celo7

@Bass It's not wrong if one extends the log to negative numbers

I just avoid imaginary outputs entirely

I guess I'm just used to GR where one always has $|g|$

or $-g$

Bass

08:50

@0celo7 I don't understand:

10 hours ago, by 0celo7

@Bass ok, I'm being super pedantic: your (1) is not true for all pseudo-Riemannian metrics, only those with an even number of negatives

DanielSank

@JohnRennie Well, whatever floats your boat.

That's way heavier than you need, fyi.

If you ever do get annoyed by anaconda telling you what to do, remember virtualenv.

And pip

Those two things are your friends.

They pave the way to happiness.

virtualenv makes isolated environments, pip installs packages into them.

TanMath

09:05

@DanielSank well, qutip isn't available for windows...

@DanielSank i still do not understand what to make into a function...

i guess there would be an initialze function to define all the matrices...

then a run function to run the simulation

then a plot function to plot it...

is that what you want me to do?

then write some testcases for the function?

TanMath

09:21

now i am confused what the inputs for each of these functions would be...

David Z

09:35

@TanMath it occurs to me that, given how long this discussion has been going on, it would probably have made sense to put it in a separate chat room so it'd be easier to follow. Probably too late now though :-P (I could migrate messages to another room, but it seems like a lot of trouble to go through the history and find them all)

@DanielSank since you mention it, can you give me a quick TL;DR of how to get started using virtualenv? I've never really needed it but it'd be handy to know how to use it

TanMath

@DavidZ i guess we should have done that... oh well...

I will make a seperate room now...

@DanielSank chat.stackexchange.com/rooms/34188/tanmaths-project

David Z

09:49

There's nothing wrong with having the discussion here, of course.

But putting it in a separate room does make it easier to follow.

@TanMath I can migrate some of the recent messages to that room if you like

DanielSank

@DavidZ What OS?

David Z

Oh, does it matter? Maybe I should just go read something, then

I use both OS X and Linux frequently

DanielSank

Ok.

Do you have a rough idea of the problem virtualenv solves?

David Z

Well, I'm not sure. I guess that's what I'm confused about. I know it's meant to create isolated Python environments where you can install and upgrade packages or so on without affecting other Python environments (e.g. the system one), but that's never actually been a problem for me.

DanielSank

I see.

David Z

10:02

I guess if you can at least recommend the first thing for me to read if I want to start playing with virtualenv, that would be handy.

DanielSank

Well, I suppose the official docs are usually the right place to start.

It may well be that your system python doesn't need many packages, and that because of this you've never accidentally clobbered something the system cared about.

Basically, you do this:

1) Install pip either in your system's python by following the instructions in that link, or via your OS's package manager.

If you use the package manger option, it would be something like sudo apt-get install python-pip.

2) Install virtualenv in your system python: sudo pip install virtualenv.

From now on, you never install anything else into your system python, ever

Suppose you want to start a project which will use python packages. Heck, it could even use a different version of python (py3 baby!).

Step numero uno is to make a virtualenv for that project:

$ cd path/to/your/project

$ virtualenv env

^ That makes a new directory /path/to/your/project/env which contains a virtual environment.

In posix systems, you turn on the virtual environment by doing this:

kevin Tah N.

aahh yes, the god old python days

David Z

OK I see

DanielSank

$ source /path/to/your/project/env/bin/activate

Now, everything you do with python in that shell is in the virtualenv.

You can pip install stuff, or whatever.

It's all contained and can't clobber any other environment.

It even conveniently modifies your shell prompt so you know which environment you're in!

type $ deactivate to turn it off.

David Z

I suppose part of the reason I haven't experienced conflicts is that as a rule, I don't install things into the system Python using pip - only the package manager

DanielSank

10:16

Once you get the hang of it, I'd actually recommend using virtualenvwrapper, which is just some slightly nicer shell scripts to ease management of virtualenvs. It's just sugar, but good sugar.

@DavidZ Ah, interesting.

David Z

mmm, sugar :-P

DanielSank

@DavidZ Yeah, it really becomes necessary to use this if you have more than one project and they don't want to share libs.

That came up when I started deploying pythong powered websites via Google App Engine.

I didn't want all this random web server framework crap all over the rest of my python environments.

It's actually also nice if for some reason you need particular versions of libs.

And again, super duper nice if you want to use a different version of python.

That's it, really.

David Z

Gotcha

I will keep this in mind - though I note that one thing I really like about having a centrally installed Python is that I can upgrade packages once and those upgrades are reflected in all my projects. It seems like virtualenv would negate that.

Slereah

11:03

So Weinberg says that decomposition into ladder operators is important for the cluster decomposition principle

But on the other hand, interacting theories are not made of ladder operators, fundamentally

Does it mean experiments are influenced by things at infinity

John Duffield

13:50

@KyleKanos : no, just a BSc Hons in Computer Science.

@KyleKanos : I'm not some crackpot, I refer you to Einstein and other good stuff, and tell you why time travel and the multiverse is woo.

@MikeMiller : He was already crackpot from the off. Parallel antiverse, LOL!

Transcript for