Ever see a question that's all messed up, but somehow still gets a couple of upvote?

I just have to shake my head. No idea what people are thinking.

@tpg2114 I dunno if it's GT's Def showing up or just how bad Cole Stoudt really is

However, Alabama beating Miss St was pretty cool to see

And Miami ahead of Shame-us Winston's FSU is nice to see too

@PhysicsMeta The user that asked that question has been suspended. Kinda confusing. I guess the user that posted it was traced back to a sock puppet account (from the same IP maybe?) and suspended?

If I'm getting it right, this is an absolutely crazy idea:

:P

But how does Doorknob know? He has all of 110 rep on this site?

@New_new_newbie Why would his rep help him figure out the statement was vague?

@KyleKanos Yeah, the top 5 are in for a total shake-up again. Too bad FSU won. It would have been nice if the only undefeated team was Marshall!

Sometimes I worry about the serial up/down voting algorithm. There are a few users that I've downvoted almost all of their answers (over a long period of time though) and there are a few like John Rennie where I upvote almost all of their answers (again, over a long period of time).

@BrandonEnright The details are intentionally kept on the QT (from mere mods, too), but it seems to rely on more than just the number of votes you have cast for or against another user.

@dmckee Yeah my main hope is that it's thresholds are set to detect clearly abusive patterns. I don't think my voting isn't abusive so I hope I'm fine.

I imagine they were designed with Jon Skeet in mind...

@BrandonEnright That guy is a mod on another SE but not on Physics. So, access to mod tools here can only be on the basis of rep I suppose. Am I wrong?

@New_new_newbie He would not have mod access here.

Yes. But that comment exchange between him and @ManishEarth appears to be based on an implicit understanding of what happened. As if both know that the OP himself is the culprit. (If I'm understanding right!)

Which raises the question - how does Doorknob know?

@New_new_newbie I don't read the exchange as an understanding of what happened. I read it simply as a comment that Manish was light on details.

No but the moderation policy is - don't divulge details regarding suspensions and warnings. Why else will Doorknob call it "intentionally vague wording"? If the culprit was someone else, that would have just been the usual mod practice - to say everything has been taken care of. It becomes "intentionally vague" when Manish doesn't want to say explicitly - "Dear OP, you've screwed yourself".

OK. I think we can agree to disagree on this. Let's not drag it further.

@New_new_newbie I think either interpretation is plausible :-)

:)

@BrandonEnright as far as I know that's all that's going on

@Shog9 fancy seeing you here

@DavidZ Overall a somewhat entertaining situation if it really was the OP sockpuppetting votes for himself and then pointing it out.

@BrandonEnright haha, no kidding. Well, it wouldn't be the first time we've seen someone do something silly like that.

(if, hypothetically, that is what happened :-P)

@New_new_newbie One may sometimes entertain the prospect of me hypothetically sharing such a conundrum with a fictitious audience of azure-titled comrades, thus piquing the curiosity of said azure rhomboids and inducing commentary from the same. If, that is to say, that conjecture were supposedly true.

@dmckee It's basically telepathy and magic.

@dmckee Ever seen a question not get those pity-votes? :D

@Danu: Are resource recommendations on topic on the hsm site?

@JamalS Of course! As long as your question is about, well, HSM

@Danu: Ok :)

@New_new_newbie I gotta love status-completed on this one

@Danu: I'm sending a screenshot of the question to your Skype before I post it.

Just in case :)

^ @Danu @ACuriousMind: Now physicists have a periodic table too! :)

physicists?

More like group theorists

@Danu: Those groups are important to us too.

@Danu Haha, yeah. But David's comment probably means we are getting it wrong.

How bad. It was a lot more fun with my interpretation. :(

@ManishEarth You know, I'm least inclined to pick up a dictionary right now!

Or refer to an online one, for that matter. That's a horribly complex sentence :(

So I guess, you mean to say - "I'm not going to talk about it and you don't talk about it either."

Accepted.

@New_new_newbie :P

Cartileginous nasal protrusions should be kept out of the affairs of users annotated with azure titles and rhomboids.

(Not that I need to be all vague/complex, it's just really fun!)

Oh Dear Lord!

"Cartileginous nasal protrusions" $\equiv$ Sticking (my) nose.

OK I won't stick my nose here. :P

Exactly

(fwiw azure titles = blue username / rhomboids = ♦ )

Does anyone know how I can (in LaTeX) draw a curly arrow from left to right which is labelled, with another arrow underneath curled going right to left, also labelled?

@JamalS perhaps using \limits on \arrow? Not sure

@ManishEarth: Doesn't seem to work.

hm

mhchem has built in magic for putting stuff on arrows, but it's not enabled here

@ManishEarth: Oh, it's not for the SE, it's in my own LaTeX document :)

@ManishEarth: I'm writing up lecture notes.

Ah

Oh, \overset will let you do whatever you want

@ManishEarth: This is explicitly what I'm trying to do.

oh

shrug

TeX - LaTeX

Getting a single arrow isn't the issue, it's getting two curly ones, above and under :)

I hope the solution isn't TikZ, it seems very intimidating and esoteric.

Will post the answer here when I have it :)

@New_new_newbie I think you're reading too much into my words

@JamalS TikZ isn't esoteric, actually it's amazingly well-documented and not hard at all once you get past the initial learning curve.

Yeah, in your specific case you just need three named text nodes, and then it's not hard to connect them with curvy arrows (and labeling arrows is easy)

I forget the sytnax though

Whenever I sit to do TikZ I always keep the last TikZ document I wrote open to the side

@DavidZ: Okay, I'll peruse the manual, but a priori it seems I need to learn a lot of prerequisites before I can draw my diagram.

If you just need it quickly:

 \begin{tikzpicture}[every to/.append style={bend left},raise/.style={edge node={node[anchor=south]{$E_+$}}},lower/.style={edge node={node[anchor=north]{$E_-$}}}]
  \node (lowered state) at (-8em,0) {$\ket{\vec\mu - \vec\alpha}$};
  \node (starting state) at (0,0) {$\ket{\vec\mu}$};
  \node (raised state) at (8em,0) {$\ket{\vec\mu + \vec\alpha}$};
  \draw[->] (lowered state) to[raise] (starting state);
  \draw[->] (starting state) to[lower] (lowered state);
  \draw[->] (starting state) to[raise] (raised state);

But of course this would be a good project to go through and learn some TikZ. You'll need to look at the section on nodes (which allow you to place text at a specified position in a picture) and the section on "to paths" which let you draw lines from point to point with some sort of custom styling (in this case, curving instead of drawing a straight line)

@DavidZ: Thanks, David. I'll work through TikZ, it's certainly worthwhile.

@DavidZ: I received a notification: "Package pgfkeys Error: I do not know the key '/tikz/edge node' and I am going to ignore it. Perhaps you misspelled it" At least TikZ has nice error messages!

@JamalS that means you need to upgrade to the latest version of TikZ/PGF

@DavidZ: writelatex.com needs to upgrade...

@DavidZ: I'll send them a request, and meanwhile compile it in my machine

ah, yes, this is why I could never get myself to use writelatex, because I can't control what packages they have

@DavidZ: Most of the time I don't need to use anything too fancy so I stick with writelatex for the benefit of having everything saved each time in the cloud. They also have git source control built in.

Fair enough

The other thing about it for me is precisely that the documents are in the cloud: I wouldn't be able to work without internet access

@DavidZ: For me, downsides are 1) loading PDF previews takes too long, 2) Have to constantly upload files to add to document, 3) auto-complete for latex requires a subscription

Thankfully, I type quickly, and make use of \newcommand a lot.

That helps for sure

and I think it's a good practice under any circumstances

@DavidZ: Regarding internet access, I have a surge protector on my computer and router. I also have a private electrical generator, and for when the router doesn't work, I have one of those USB dongles. If that fails, I have hotspot from my iPhone.

Wow you're well prepared

Unfortunately for me, I'm in China. There are no good internet connections.

@DavidZ: The closest I've been is to Hong Kong and Singapore.

@DavidZ: Detested Hong Kong, too hot, too humid, and way too many people. On the other hand, I really enjoyed Singapore.

I think both of those are more "wired", i.e. probably likely to have better internet

and also not subject to a national firewall

@DavidZ: You're in Wuhan?

yep

Oh, I know, the firewall is a pain.

To get around that, I used a remote desktop connection, and essentially used my router at home.

The internet connection quality here isn't good enough to support remote desktop, and they might block it anyway. I've already encountered blocks on VPN, SSH, and HTTPS connections.

@DavidZ: Blocks on HTTPS? That's atrocious. How do you access e-mail?

Or are some HTTPS connections allowed, e.g. to GMail?

The blocks aren't consistent, so e.g. I can generally access Stack Exchange over HTTPS, but not Google.

For email I have to go through the VPN or an SSH tunnel, when it works. (Sometimes it does, sometimes it doesn't)

Occasionally they seem to let a Gmail connection slip through.

What's really weird is that they blocked SSH to my new web server but not my old one

although it's probably a matter of time before they catch on

Wow - I'm speechless. The censorship reminds me of Orwell's 1984.

yep

I mean, for what it's worth, they don't really make a secret of it

and the firewall is not truly effective against any computer user with even moderate technical ability

It's more annoying than anything else

Can somebody please recommend me an introductory undergrad textbook on special relativity?

@picaposo most of the books you'd find cover special and general relativity, so see this

@Danu hey man, let me know when you're around, got one small question on Riemannian curvatures :)

@Phonon: What's your question?

@JamalS it's along the lines: when does the curvature (riemannian) of a manifold changes sign?

e.g. when that manifold describes the phase space of a physical system

@Phonon: What specific curvature are you referring to? The Ricci scalar?

Or the Kretschmann scalar?

@JamalS no the curvature tensor

as used in the Jacobi equation

@Phonon: Ahh, ok

In V.I. Arnold, mathematical methods of classical mechanics, in the first Appendix (riemannian curvature), the example of divergence of two geodesics is solved, for the case where the curvature is negative in all 2-d directions containing the velocity vector of the geodesics

@Phonon: I'm reading the book now, and it seems he is not taking the curvature to be negative.

He is setting the second derivative equal to minus the curvature

@JamalS see page 311, "In particular we consider..."

@Phonon: Oops, I was page 310 :)

:)

@Phonon: So, can I rephrase your question as: "What is the significance of the sign of the curvature of a phase-space manifold?"

@JamalS yes

@Phonon: Hmm, that's a good question. Geometrically, it's clear what it means, but I can't relate that back to the properties of the physical system itself. I'll have to think about it. I'm sure Danu will be online soon.

@Phonon: Why would the phase space have a Riemannian curvature? What's the metric?

@ACuriousMind well the phase space of a chaotic system e.g., trajectories diverge exponentially, just as do geodesics on a manifold with negative riemannian curvature,

^ Indeed, the phase space may not necessarily be endowed with a metric, since you haven't specified a specific system.

@ACuriousMind I know that for a regular system, (integrable-invariant torus) the curvature has to be 0, but I was just thinking that for a chaotic one in contrast, where we know trajectories always diverge exponentially w.r.t.o (and never linearly e.g.), the corresponding manifold has to have a negative curvature,

I still don't know what the metric is - are you just taking the "Euclidean" metric on it?

@ACuriousMind no the riemannian

There is no "the Riemannian metric"

A Riemannian metric is any metric with positive signature

And phase space carries no natural metric, only a symplectic structure

(at least, I don't know what the natural metric on a phase space would be)

@ACuriousMind well doesn't the curvature here just describe the deviation of the geometry from that of an Euclidean? sometimes called "deviation from Euclideanness"

yeah but I see your point

You can't talk about curvature if you have no metric

And yes, curvature describes how "non-flat" the metric is

Although, it should be stressed, it is how intrinsically non-flat the metric is. For example, a cylinder is curved, but it has zero intrinsic curvature. On the other hand, it's extrinsic curvature does not vanish.

@ACuriousMind you're right. I'll have to mull over this one for a while

@JamalS the question actually came at me, when I saw the exponent he calculated for the geodesic divergences (page 312, \lambda=...) and noticed its resemblance to the Lyapunov's exponent, and looked for a link. But there may very well be none!

Hi guys

looks like your question was already resolved, @Phonon?

@Danu well not entirely,

As @ACuriousMind said, you need a metric

@Danu it got stuck at the point of defining a metric for the phase space, of non-integrable systems

@Danu well assuming it is an euclidean, for our purposes, then we can define the curvature tensor, right?

@Phonon Think about this more carefully

I'm not sure a naive guess like that is sensible

(I'm also not sure of the contrary)

yeah, true

webusers.imj-prg.fr/~nicolas.lerner/ch2booklerner.pdf

thanks :)

llooks very technical

@DavidZ I suspect the Great Firewall serves two purposes. Firstly to control the flow of information to the bulk of the people and secondly to let them tag as potential troublemakers anyone who routinely circumvents it.

The fisrt use makes it easier for them to win the local propaganda battle.

I presume the information from the second application would typically be held indefinitely in case they have a future need to apply pressure---pretty much standard practice for totalitarian regimes of all stripes.

Awwwwyis, I finally have all the French algebraic geometry books by Grothendieck! If only I could understand them! :D

@JamalS you can use things like xhookarrow

@DavidZ Right... once you get past the initial learning curve

@JamalS xleftrightarrows should resolve your problem - they are not curved however