« first day (2902 days earlier)      last day (2022 days later) » 
05:00 - 21:0021:00 - 00:00

user280247
5:09 AM
I'll made a very stupid question possibly, but what does it mean physically the 'time reversion' of newton laws?
 
user280247
I understand the math (I believe), but I can't guess what is important/interesting about it
 
@santimirandarp I think that probably refers to the fact that Newton's laws are time reversible i.e. they work the same way for time flowing forwards or backwards.
 
5:26 AM
What is meant by pass axis here?
 
The pass axis is the plane of polarisation of the polariser
i.e. light passing though the polariser is polarised in the plane defined by the horizontal axis and the pass axis.
 
So how can they have same intensities?
with different angles too?
 
What's the angle between 8º and 188º, and between 38º and 218º?
 
180 and 180 ...
so the there the intensities are same...
but...between 8 and 38?
 
Right, so of the four angles you're given 8º and 188º are the same straight line passing through the centre of the polariser, and so ar 38º and 218º.
So you have two lines forming an X
 
5:33 AM
yes...
 
And there are two other lines that bisect the two angles in the X
 
well the other two lines are the axes,right?
 
You have one line at 8/188º and the other at 38/218º. So one bisector is at 23/203º ...
and the other is at 113/293º
Shall I draw a diagram?
 
I dont understand your notation..yes....please.
 
The blue lines are the polariser angles in the question, 8º, 38º, 188º, and 218º
The red lines are the bisectors of the two angles formed by the blue lines.
 
5:45 AM
ok....thank you,,,i now understand,,,,
again i will ask the qsn,,,the red lines are the axes?right?>?
 
Why would you think the red lines are the axes?
The point of the red lines is that if you take light polarised in those directions the component of the light along the two blue lines is the same ...
If you call the angle between the red and blue lines $\phi$ then the component of that that light along both blue lines is $\cos\phi$.
 
iok.
ok,
 
ok now i get...
it,,,yes,,,
thank u
 
Cool :-)
 
5:50 AM
because for both the cases cos phi is the same ,
Our physics teacher always asks us to draw a diagram...i now know why.
 
... so light polarised along the red line gives the same intensity for both polariser angles.
 
user280247
@JohnRennie yes but what's the curiosity about that?
 
It was rather stupid of blaring out the term axes without any context.
 
@santimirandarp well everyday experience tells us that times isn't reversible. It only flows forward. But this idea of a direction to time is completely absent in Newtonian mechanics.
This has puzzled generations of physicists.
Indeed many of us believe time doesn't flow at all and the flow of time is an artefact of the way the human brain works.
 
user280247
Well yes but time can be safely forgotten..
 
user280247
5:57 AM
still would be a problem, one thing follows another and not the inverse...isn't it?
 
user280247
6:15 AM
@JohnRennie did you make the plot?
 
user280247
I'm trying to make nice ones but anything I try is bad or difficult to use
 
user280247
gnuplot and tikz, respectively...
 
@santimirandarp I used Google Draw. It's basic, but it's easy to use and fine for simple diagrams like that one.
 
user280247
@JohnRennie Cool...
 
Anonymous
7:44 AM
@santimirandarp gnuplot is not suitable for anything other than graphs
 
9:15 AM
Has anyone read The Geometry of Physics by Theodore Frankel? Amazon just spammed me with an ad for it, and it looks as if it covers some interesting ground.
 
'fraid not
looks like a big old gauge theory book
 
 
1 hour later…
user351417
10:18 AM
Would opinion-based be a better description for this question?
 
user351417
0
Q: What do researchers hope to learn from the “Bloodhound SSC” rocket-powered car project?

gerritThe Bloodhound SSC is a project to develop a rocket-powered car that can travel at 470 m/s (1690 km/) (with a crewed payload!). See Wikipedia or BBC News. What do scientists and engineers hope to learn from this project? Rockets are a currently inevitable technology for locomotion outside the ...

 
@Chair to be honest, once a question is seven years old there are probably better uses for your close votes.
 
user351417
@JohnRennie Wait, that question was posted 10 minutes ago, right? Gimme a minute
 
@Chair oh yes. Huh? What question did I end up looking at?
 
user351417
@JohnRennie Perhaps one of the linked questions?
 
10:27 AM
@Chair anyhow, now it's closed I'd just move on.
 
user351417
@JohnRennie Yeah, that makes sense, particularly since the premise that the Bloodhound project is research-focused isn't correct. Thanks!
 
@JohnRennie I expect they'll learn more about the aerodynamics of ground vehicles moving faster than sound :P
 
@Mithrandir24601 possibly also make advances in medical care :-)
 
10:50 AM
@JohnRennie yeah, possibly... :P
 
 
2 hours later…
12:42 PM
in The Factory Floor, 3 mins ago, by Secret
So if we exposed something simutaneously to extreme heat and extreme cold, it will be a lot worse, but I don't know how to describe what the thing will become
 
12:56 PM
Why do people say we can know the shape of the earth with local experiments
when it's just its geometry
the topology is something else entirely!
The earth could be a hemisphere
 
1:22 PM
4
A: Who cut the cheese?

samcarterA question about cheese magically attracts the TikZmice. They are part of a LaTeX package named "tikzlings" which is currently developed with the help many TeX.Stackexchange users. In case you want to try it yourself, the package source code is available from https://github.com/samcarter/tikzli...

seriously?
bejeesus
tex.se's madness knows no bougons
I mean bounds
 
does it know any bougons though
 
I mean
they went and tikzed all of that?
 
it's cute
 
it's cute, yes
but it's insane
well beyond @Danu's mild eccentricities, for one
 
2:13 PM
TeX sounds pretty far out there as far as hobbies go
 
@JohnRennie Frankel is the kind of book that will spend a lot of time setting up math tools (especially algebraic topology, differential geometry) for physics in coordinate-free form trying to give pictures now and then (I have been disappointed every time I looked in it for little things but I know it's supposed to be really good if you take it on it's own terms)
 
2:30 PM
@bolbteppa is it similar to Schutz?
 
Yeah that's a good spirit-animal analogue
 
You can tell when a book is serious about coordinate freeness when it uses musical isomorphisms
nothing so crass as an index
 
You mean you don't like $v^\sharp$
Or I suppose it would better be $\omega^\sharp$. I never really understood why you'd use that notation over an expression with the metric though
 
@EmilioPisanty Don't challenge me :P
 
2:47 PM
@danielunderwood expression with the metric would be worse, though
Would be like $g(v, -)$
 
3:06 PM
@bolbteppa Now I find myself wondering what my spirit animal would be
The owl is the obvious one for a bookish insomniac
 
I concur
 
3:19 PM
@Danu I'm saying you're doing better than other people
though if you actively want to be crazier than the others, then to each his own, really
 
3:39 PM
hmmm
 
3:59 PM
folks, some bibliographic help
I was under the impression that Rev. Mod. Phys. papers need to mention all their references in the text
but I can't find where the Choi 2012 paper is mentioned
is that impression incorrect?
no, wait, found it
ctrl+f definitely isn't working today
 
:D
 
@bolbteppa thanks. I found a pdf of the contents on the publisher's web site and it's over my head anyway.
 
4:22 PM
omg
I requested a graphing software a month ago...apparently it's so slow cus it's pending approval from the VP of infosec...why the heck does this have to go up to the VP level loooool...
 
don't want to get those nasty graphing bugs
Also there's this that is supposedly going to offer GPU implementations of sklearn algos. I've never had any performance troubles with the CPU versions though rapids.ai/index.html
It's also filled with nvidia's "Look at how much faster GPUs are for GPU tasks" marketing
 
For vanilla ML I'm not sure how much GPU acceleration one can expect...
GPUs really good for like massive matrix multiplication problems
 
I mean you don't think a GPU company would tell you that you need a GPU if you don't, right?!
3
 
@danielunderwood A simple yet profund insight into human mentality :P
 
4:39 PM
stupid timesheet entry website is bugging out
Finally...took so long to just enter the time -.-
 
Huh indeed
hmmm
 
gonna go ahead and flag it for non-mainstream
 
> You have proven ability with data interrogation
seems a bit aggressive towards the data
 
I finished writing the topic modeling code
 
4:47 PM
::Imagines @danielunderwood beating the data with a wrench to obtain its meaning::
 
now it's more about playing around with it to get good topic models
 
5:03 PM
Hello, I'm in need of assistance
In the Clausius-Clapeyron equation, does ΔᵥH refer to molar heat capacity or specific heat cap?
Wikipedia says it's specific, but I have a weird feeling it may be molar, especially since R is defined for a per mole basis.
I'm referring to: $\log{\left(\frac{P_1}{P_2}\right)}=\left(\frac{\Delta _{v}H}{R}\right)\left(\frac{1}{T_2}-\frac{1}{T_1}\right)$
 
@PrittBalagopal I am 99% certain it is the molar heat capacity. Let me dig out my trusty old copy of Moore and I'll double check.
 
I too felt the same way
 
Anonymous
@PrittBalagopal Long time no see. How's it going? :D
 
Hey Blue, whats up?
Yeah, I completely forgot about SE for a while
And not getting much time to contribute these days :(
 
Anonymous
We're having a 2-week long holiday now, due to the pujas :) I'm mostly sleeping all day and going out for pandal-hopping once in a while :P
 
5:13 PM
Ahh nice
 
Anonymous
@PrittBalagopal I can understand :/
 
Anonymous
Same here with most of the other uni students
 
yeah, hopefully this semester holidays should be very fruitful
 
Anonymous
When are your holidays?
 
Around christmas
 
Anonymous
5:14 PM
Oh, same
 
Anonymous
After semester exams
 
Yup, same
 
Anonymous
How's the Chem eng syllabus?
 
I plan to study my next sem portions in those holidays, so I can get a head start
Oh it's going quite fun actually
 
Anonymous
Like mostly chemistry or mostly physics-y stuff?
 
5:15 PM
I'm loving the current subjects
Ehh its a weird assortment lol
 
Anonymous
I guess tons of thermodynamics ? :P
 
Chemistry, Fluid Mech, Thermodynamics, Electricals, Processes
Yup, there is definitely quite a bit of Thermo in all of them
 
Anonymous
Ah, cool. Not for me though :P
 
Anonymous
Transistors are killing us on the other hand
 
Oh I totally get you man, transistors are so hard to understand
I could get a qualitative idea after watching those Learn Engineering YT videos, but a quantitative idea is harder to grasp
 
5:18 PM
circuit elements are easy, so long as you only worry about the linear ones :P
 
Hahaha yeah
 
Anonymous
Do you have any electives this year?
 
Anonymous
Or any minor?
 
Why does every job posting ask for spark/hadoop experience?? Are these companies all maintaining their own computing clusters and servers nowadays...
 
Yeah, we can choose between two electives for next sem, nuclear or petrochemical
 
Anonymous
5:19 PM
@PrittBalagopal nuclear obvio XD
 
I would have thought they'd just all be on AWS or Google Cloud or something
 
Some peeps tell me to go for nuke, although I'm interested in petro lol
Haha you too lol
 
Anonymous
Oh boy :P
 
Anonymous
It sounds like a lot of chemistry and memorization
 
Yeah, petro also seems more towards what I envision my future to be
Nukes are cool, but nahh
 
Anonymous
5:20 PM
@PrittBalagopal Oh?
 
Anonymous
Like?
 
I dont want to elaborate too much, but I try to think about alternate fuels
 
Anonymous
Sounds cool
 
So I think studying petro can help understand the difficulties of alternate fuels
Yup, sure is.
 
Anonymous
5:20 PM
Yeah, save us from the global warming man ;)
 
I will try :)
 
@enumaris A lot of them are probably hadoop deployments on one of the cloud providers, but a lot of companies still have on-premises stuff as well
 
hmmm
It feels like working w/ Hadoop/spark should be a software engineer's job instead of a data scientist's :P
 
Are AWS and Google Cloud the most used in businesses? I personally use Heroku and have been considering DigitalOcean or Azure.
 
I feel like a lot of companies are working with them when they don't really need to be. Hadoop is a pain to set up when you don't actually need it
 
5:25 PM
Ahh and Heroku is also known for being extremely simple to use too.
 
what's the split in how spark/hadoop works specifically?
 
I use the free plan and do quite alot of stuff btw
 
Hadoop manages the storage and spark manages the computing?
@PrittBalagopal I am not familiar with those
 
Alright
 
AWS is generally most used. GCP and Azure have similar offerings. Heroku is the PaaS part of them and DigitalOcean is essentially VMs. I think AWS/GCP/Azure are more complicated and expensive than the other offerings, but offer things like HIPAA/PCA compliance
 
5:27 PM
I wonder how much infrastructure cost goes into maintaining a medium sized on-prem computing cluster...
 
@enumaris the way that I understand is that hadoop is just batch processing while spark supports both batch and streaming while also having a pandasish library in python. My understanding is that spark is newer and better, but I haven't really had to look at them beyond realizing that they were too much overhead for the amount of data I had
 
For deep learning, I feel like getting a 20k machine...maybe 8 top of the line GPUs should be able to do like... a ton of work...
 
Hadoop does have a clustered filesystem component. I'm not sure if spark has something similar
 
hmmm
@danielunderwood most of the job postings require both lol
 
I'm getting to the point where I may just start ignoring the requirements and apply for the jobs I find interesting lol
I saw a job earlier for "Data Scientist I" that wanted a PhD in machine learning
 
5:31 PM
lol
 
And those jobs seem to just be sitting there. I started casually looking for jobs a few months ago and some have been there the whole time
 
Yeah...
 
I did once see a posting that said PhD or higher lol
The only thing that I could think of would be a postdoc position or something
 
I suppose you could look for a prof in the field of ML
or just say
"Apply only if your name is Andrew Ng and you teach ML at Stanford University"
 
lol
"Sorry we're looking for someone with more Excel experience"
 
Anonymous
5:44 PM
Hmm, I'm wondering whether XORing 24 bit (binary) integers and counting 1's would be the fastest way to calculate Hamming distance or there's something better. XORing in Python would be much slower than in C I guess
 
Anonymous
(there's a few million of the binary integers btw)
 
I think the worst part is that I have to at least skim every "Data (Scientist|Analyst|Engineer)" posting since every company seems to have a different definition of what each one is and what requirements it entails
@Blue benchmark!
I figure for low-byte integers it may be equivalent?
 
Yeah...
 
or the occasional Software Engineer, (Machine|Deep) Learning
 
Data Scientist vs Data Analyst vs Data Engineer is not well understood across the board...
 
Anonymous
5:46 PM
@danielunderwood Yeah, lemme put a timer and check
 
it's all venn diagrams
 
Could apply to the wrong one and get a job dragging & dropping plots all day
 
I suspect that would be cleared up at the interview stage...
but yeah they could pull a bait and switch I guess lol
 
@Blue It would surprise me if bit operations on integers would be much slower in Python than in C. But of course it's once again an implementation detail and not a feature of the language
Nothing prohibits a Python interpreter from implementing an efficient XOR, but nothing forces it to, either
 
Just write your own interpreter
problem solved
 
5:49 PM
I figured the basic types would be using the types in the interpreter language. There's some funkiness somewhere that python allows integers of arbitrary byte length though
I've always been curious about writing my own interpreter
 
do it
Then you will learn the joys of never having to write your own interpreter
220k more comments to process...
 
@danielunderwood It's easy enough for a toy language, it's a hell of a lot of work for a "real-world language"
 
I can't imagine that this is quite as fast as a plain C implementation unless the compiler is doing something truly crazy github.com/python/cpython/blob/master/Objects/…
@ACuriousMind yeah I'd probably have to do it on a subset of a language if I tried it. I did try to create a dictionary implementation in C last night and it completely exploded
 
do it
dew it
 
Also @Blue there's weave that I've heard people talk about. Just note that you'll want to bundle up your array and call a c function on the whole array rather than passing a pair at a time
I'd like to see someone try to implement a python interpreter in MATLAB...just because
 
5:59 PM
sounds legit
 
I found this wonderful thing back in college mathworks.com/matlabcentral/fileexchange/34513-matlabtetris
(it's tetris)
I never understood how people could write entire pieces of software in MATLAB
 
lol
y tho
it was a homework for some class?
 
I refer you above to the convo about mice in TikZ
Some people are just...different
Though I do admit that I find interest in some ridiculous tasks
 
hmmm
 
Speaking of ridiculous tasks, I kind of want to work out electromagnetism in weird spacetimes
 
6:10 PM
"work out"?
Like given some distribution of charges in some given space-time, figure out the motion?
Assuming the charges themselves don't affect the background spacetime?
 
Well I was thinking the kind of Maxwell's equations of different spacetimes, but I guess that's already done in the field tensor formulation and they wouldn't necessarily become a set of vector equations like they do in Minkowski
 
So you mean Maxwell's equations in some given coordinate system?
wtf...why is pandas not saving the last column of this dataframe...
 
My first thought was taking a spacetime like $\mathbb{R} \times S^3$ and getting "Maxwell's equations"...but that sounds like less of a well-posed problem the more that I think about it
Does the last column happen to be an index?
 
no
it's a bunch of numbers
$\mathbb{R}\times S^3$ is not a spacetime, but just a manifold. You have to add a Lorentzian metric on top to make it a spacetime.
 
Anonymous
@ACuriousMind I thought that GCC supports binary literals (beginning with 0b) and thus XORing (using say uint32) would be more efficient on them. Python ints, on the other hand, don't natively support binary literals (putting them in normal ints would be memory consuming). So I'd need to go for something like bytearray (programiz.com/python-programming/methods/built-in/bytearray). But then even a CPython program can't beat a C program if they're doing the same thing.
 
Anonymous
6:18 PM
Also, calls to the CPython program would eat up time (especially when dealing with several million numbers)
 
what the crap is going on lol...so confused
I print the dataframe and I see 3 columns + index
I save the dataframe to SQL and it has only 2 columns
uhhhhhhh
 
hmmm clearly I need to learn more about the general formulation of EM
When you say you're saving it to SQL...are you writing the inserts manually?
 
nope
it's a pandas method
to_sql
oh
 
Sup guys, I was just wondering is YOLO (you only look once) from machine leraning unsupervised? I'm just getting into machine learning and YOLO looked interesting. Since it's a real-time object recognition software I suppose It'll be unsupervised since it's just finding patterns at the moment and not relying on set data to make predictions. But also I guess it kinda looks up google to recongize. Maybe it's semi-supervised?
 
I'm retarded
My select statement is wrong looool
 
6:22 PM
huh that could have saved me a bit of work a while back. I didn't even think to look if pandas would do such a thing
 
of course pandas would do such a thing
it uses sqlalchemy to do it tho
for other SQL stuff I use pyodbc
 
I just used a db connector and manually read/wrote to/from dataframes
 
Is it possible to have a manifold that changes in some sort of way that changes the metric signature?
 
not in GR I don't think
the metric signature is a constant of the manifold iirc
I don't recall why that is though
 
6:28 PM
Maybe YOLO uses reinforcement learning type?
 
Are there theories that allow a dynamic manifold? Related, is there a name for a transition between manifolds? Say you go from a sphere to torus
 
user280247
6:50 PM
I'm trying to understand time reversibility paradox, the problem is I can't see a paradox...Even though newton's laws are time reversible the forces determine the direction of the movement, so, what is the paradox?
 
user280247
There's no problem with the maths, at least in the basics...
 
user280247
The world could be going backwards but only if the initial forces where the opposite...
 
@santimirandarp The paradox is that if you just take the initial conditions at some point in time and solve the equations of motion, you get both the past and the future evolution, and nothing in the equations as such indicates which is which.
 
user280247
But we need the forces...@ACuriousMind
 
user280247
they tell in which direction the body is moving...; as least if we consider V_i=0
 
7:05 PM
@santimirandarp When we talk about "reversing time" or "T-symmetry", it is implied that we also invert all velocities (because velocity is an odd derivative w.r.t. to time, and so changes sign when time does). You are right that for an initial condition with zero velocity time reversal doesn't really do anything..
And then there's Loschmidt's paradox as a follow-up - how do apparently "irreversible" processes arise from time-symmetric, and hence reversible, mechanics?
 
user280247
@ACuriousMind it's fine for me if I understand the first paradox before my lifetime ends...XD
 
To be fair, the problem of why the arrow of time points where it does is not really a "paradox". There's no apparent contradiction, just the question "why does time run forward and not backward?" or "why do we perceive a unique direction of time at all?"
 
0
Q: Why was my self-answer accepted, and why did I get +5 rep?

YlyI just found that my own answer to my own question has been marked as accepted today, and I got +5 rep for it. I didn't do this, and I didn't think it was possible to get rep for accepting your own answer anyhow. Did some moderator do this, or is this a bug?

 
user280247
@ACuriousMind hm...
 
Loschmidt's paradox on the other hand is a true apparent paradox.
 
user280247
7:18 PM
So the question "why does time run forward and not backward?" would be equivalent to "why the velocity of the particles is v_x and not -v_x?
 
user280247
@ACuriousMind
 
That meeting with I.T. was kinda aggravating...
lol
 
@santimirandarp That's part of it, but T-symmetry is more than just reversing velocities. The question is "why do we experience the solutions of physics as we do and not their T-symmetric (=time reversed) versions?"
 
@danielunderwood a change in topology would be quite drastic since the topology of a set is a pretty primitive concept. The problem you might run into though is "what do you mean by transition"? A topology doesn't have a notion such as time, so what does it mean for a topology to transition from something to something else?
 
user280247
@ACuriousMind why we couldn't say: we experience those because of the initial force...?
 
7:23 PM
@santimirandarp Note that forces do not change under time-reversal, so you cannot use the direction of forces to distinguish a solution from its image under T-symmetry.
 
user280247
(which sets the velocity also...)
 
I once asked a question about what prevented the (3-D) hypersurfaces in FLRW from changing topologies...but that was for embedded hypersurfaces in a manifold which is globally hyperbolic so there is a clear definition of what it would mean to transition. I don't know if I got a very clear answer to that one either though...
 
user280247
Well it does change, a force towards the earth is different to a force outward the earth, isn't it? @ACuriousMind
 
@santimirandarp I don't understand what you're saying
$t\mapsto -t$ is a well-defined symmetry. Forces do not change under it.
 
@enumaris maybe a change of genus? I think that would be a mathematically ugly transition, but maybe it's possible in some sense? I'm not terribly familiar with topology though. The other thought would be a transition between spacetimes where the metric signature changes, but I'm even less familiar with the consequences of metric signature
 
7:25 PM
like a portion of the spacetime where the metric signature is one way and another portion where the metric signature is different?
I feel like you would have trouble making that transition smooth
 
user280247
When you say forces dont change, it doesn't seem true about direction of the forces...@ACuriousMind
 
but I don't know of any work in that area
 
@santimirandarp I assure you it is true.
 
Anonymous
It's so annoying that the Stack Overflow chats are dissociated from the Stack Exchange chats. I need to keep two separate tabs open :/ There might be something in Stack Apps
 
See e.g. the Wikipedia article about T-symmetry I linked, which even discusses that the Lorentz force is invariant under it only because both velocity and the magnetic field change signs!
 
user280247
7:31 PM
@ACuriousMind too complicated, but thanks anyway...
 
@enumaris my line of thinking was a time-dependent change. Perhaps you could parametrize such a thing, but I suppose it would be changing in different regions of the manifold if your parameter was one of the parts of the manifold. It seems to me that it wouldn't be a nice transition, but I'm pretty far out of the waters I know
 
"time-dependent" can get pretty tricky in general space times :D
 
@danielunderwood There are far-reaching theorems in GR that forbid the change of the topology in many senses. I don't know any details, but I'm sure @Slereah does and also has some pathological examples for you where it happens nevertheless.
 
changing metric signatures is beyond my level of understanding as well
 
Such things are even beyond my ability to google at the moment lol
 
7:35 PM
@ACuriousMind seems like some of those theorems would have been useful to answer my question w.r.t. topology changes in FLRW
 
7:55 PM
Is this right: A symmetric tensor in 4d spacetime has 10 degrees of freedom, if we want it to satisfy the EFE then must be defined only up to diffeomorphisms, removing 4 d.o.f., but then 4 Bianchi identities remove 4 further d.o.f. leaving 2 d.o.f. (polarizations) total, and since a symmetric traceless tensor in $4 - 2$ dimensions has $(4-2)(4-1)/2 - 1 = 3 - 1 = 2$ d.o.f. we have a symmetric traceless tensor of the little group $SO(4-2)$ describing a massless particle?
 
Are you trying to arrive at the fact that gravitons are spin 2?
 
Trying to justify it depending on 2 d.o.f.
The issue is the Bianchi identity argument I think
Another thing I seen is you can say the EFE put 4 constraints on the d.o.f.
 
The EFEs should have, in general, at the end of the day, 6 degrees of freedom though. The Bianchi identities are built into the EFEs as something the Einstein tensor already satisfies though, I don't think they remove degrees of freedom.
A general metric will obey the Bianchi Identities
there's no further restrictions on the metric needed to make the Bianchi Identities hold
 
I should say to describe a graviton
3
A: Degrees of freedom of the graviton versus classical degrees of freedom

yngablAs i can remember, for the case of classical electrodynamic the d.o.f counting start after the assumption of the Bianchi Identity, and at the end the desired result came all from the gauge freedom. In fact $\partial_{[\mu}F_{\nu\alpha]}=0$ only choose a suitable form for $F_{\mu\nu}$, for example...

 
Anonymous
8:08 PM
LOL
 
@danielunderwood that kind of topology change in GR is called a Lorentz cobordism
It is fairly restricted indeed, yes
Topology change usually means something isn't quite optimal
the big three are lack of time orientability, singularities or closed timelike curves
For instance if you pick the plane with a handle as a spacetime
There's a pretty obvious closed timelike curve
and of course you can always just remove points from a manifold to make another manifold
 
7
Q: Questions about the degree of freedom in General Relatity

346699I'm confused about the number of degrees of freedom in General Relatity. There are two ways to count it. However, they are contradictory. For simplicity, we consider vacuum solution. First, $G_{\mu\nu}=0$ gives $10$ equations and $g_{\mu\nu}$ have $10$ degrees of freedom(d.o.f). While $\nabla^\m...

 
8:38 PM
@bolbteppa looks like I was totally wrong :D
 
So Bianchi reduces from 10 to 6, and we have 6 d.o.f. in general, and then 4 more by fixing a gauge?
3
A: Questions about the degree of freedom in General Relatity

asperanzThe key point in all of this is that general relativity is a gauge theory, and, as the saying goes, "the gauge always hits twice" (apparently attributed to Claudio Teitelboim). What this means is that (1) you have an arbitrary freedom in defining your evolution, corresponding to the ability to m...

 
8:51 PM
4
A: The 10 independent components in the Einstein's Field equations

J.G.The $4\times 4$ tensor is symmetric, so the $6$ entries below the leading diagonal are equal to the $6$ above it. Therefore, there are only $4^2-6=4+6=10$ DOFs before we consider Bianchi; the sum $4+6$, another way of counting the DOFs, is one triangle plus a diagonal. Bianchi is a vector$=0$ con...

So the 4 gauge d.o.f. are the 4 in going from 10 to 6 and 4 in going from 6 to 2 :\
 
@bolbteppa Of course it does - most gauges "strike twice" because the transformations involving the gauge parameters involve both the gauge parameters and their first derivatives, so the d.o.f. eliminated when eliminating gauge freedom are twice the number of gauge parameters.
(Note that this is in no way a universal law if one looks at very generic gauge theories, but happens to be true for gauge theories of Yang-Mills type, and by being able to be cast in the language of gauge fields (Christoffels) and curvature GR is close enough to that for this purpose)
 
05:00 - 21:0021:00 - 00:00

« first day (2902 days earlier)      last day (2022 days later) »