The h Bar - 2020-02-06

« first day (3381 days earlier) ← previous day next day → last day (1550 days later) »

5:50 AM

"When an electron is struck by electromagnetic radiation (informational input) the electron jumps up or down into a lower or higher energy orbit depending on the wavelength (the type of input) of the radiation it receives. It then outputs (information generation or state transformation) a photon with a wavelength depending on the energy level wavelength of the photon that was inputted into it and therefore the maximal or minimal possible energy it can output (bias).
In this case the bias would be the energy threshold of the atomic system, if it exceeds a certain lower or higher end bound th…

6:03 AM

@Milan WHAT ? WHY?

Hi

Novalium Company

6:59 AM

Is is possible for a China-based company to expand globally?

I mean, as far as I know, China is a closed off country but what if a company that is pretty big domestically wants to expand into other countries. Will the Chinese government allow that?

@NovaliumCompany Like Huawei you mean?

Novalium Company

I guess

Is there some kinda tech war between China and the US of whose companies can capture the rest of the world?

All companies compete with each other. I think calling this war is a bit extreme.

Novalium Company

yeah agreed

What if the next big companies (the next Google, Apple, Facebook) are established in China? What do you think about that?

Chinese tech companies and startups have huge government support and incentive to innovate. The population there is vastly connected and a huge stream of information is constantly flowing. Connectivity = productivity.

Do you think China can become the next tech giant? (especially now when they are hugely AI orientated)

1 hour later…

8:19 AM

@Slereah Sir I got a very tough question for you, should I put it up?

@Knight Shoot

And I am no sir

No gods or kings, only men, etc

8:36 AM

lesswrong.com/posts/rrW7yf42vQYDf8AcH/timeless-physics

intechopen.com/online-first/…

The moment of change only takes 0 seconds

I wonder how will the maths of quantum mechanics is like if we replace time with dispositions

clearly $e^{i\hat{H} t}$ no longer makes sense

@Slereah Please acknowledge the terms and conditions: you will not get angry and you will not say anything until I finish. :)

because there is no longer a time interval

Novalium Company

why does no one wanna discuss china with me :(

We don't want to get arrested

Novalium Company

tf?

8:39 AM

(it's a joke)

(OR IS IT)

Let me start

Today when I went to bed, silence was beating up the night sky

Novalium Company

tefe?

China China China.

And I thought something, something very nice. I thought that after 4 years from now I shall become eligible to work wuth yku as an assisstant in your journey of GR.

Please understand the typos

Then I will work day and night with you, I will listen all your ideas, all your own elegant thoughts on GR. We will work together like a ... like a ... like a farmer

Novalium Company

tf r u talkin bout?

After that I thought, it would take me around one year to knoe al your thoughts. And after knowing your thoughts I will publish them under my name, I will publish all your hard work with my name and become a renowned figure in the whole world.

8:47 AM

does anyone know the form the trace of the extrinsic curvature tensor for the AdS_{4}-Schwarzschild metric?

Now, I'm unable to think what you would do to me. Assume that killing is not allowed in real world :). Please solve this problem, I'm stuck

Novalium Company

@Knight What in the f*cking world are you talking about?

@NovaliumCompany please avoid the cuss words

Novalium Company

@Knight What in the world are you talking about?

@Knight GR is a lot of fun, but you'll struggle to find a job doing research in GR.

8:50 AM

Yes, therefore I thought i would work as an assistant (lol) @JohnRennie

Novalium Company

anyways, i'll just go watch some Ronny Chiang vids.

h bar have been becoming super weird ever since I became inactive

@Knight unless you mean an unpaid assistant, you'll struggle to find a job as an assistanct researcher in GR.

@NovaliumCompany most of us are mainly interested in physics not international commerce.

@JohnRennie Yeah I too think that

I think my joke was super bad and hurting?

I apologize to everyone

I doubt you offended anyone because I suspect none of us had the faintest idea what you were talking about :-)

8:57 AM

Ahhahahahah. Thank you sir

Physics need more alternative conception of time besides linear and circular

What about a triangle

or

a cube

Did anyone ever call you Sammy?

9:20 AM

Why would they

9:32 AM

triangle time hmm...

that will mean there are singularities

because it will mean we are dealing with a time that is a conifold

I have not thought of conifold time geometries before...

Obviously not, since all 1-manifolds are homeomorphic to the line or the circle

Are non hausedoff 1 manifolds homeomorphic to the line. I think the line with two origin isn't

All connected, hausdorff, paracompact 1-manifolds are homeomorphic to the line or the circle

if you prefer

right

hmm...

Novalium Company

@JohnRennie I thought you liked learning about unfamiliar fields to widen your view on the world :P

9:36 AM

If you consider one dimension of time, that doesn't leave a lot of topologies

1) the line
2) the half-line
3) the segment
4) the circle
5) the long ray
6) the long line
7) some unholy non-Hausdorff traintrack

drools on unholy things

Of course, experimentally speaking, any time structure you consider will just be a segment

That was actually part of the point of Reichenbach

true, we only ever get intervals by the use of start and end point measurements

(also giving a topology to time only make sense if your spacetime can be foliated by spacelike hypersurfaces)

Foliated uniquely, I should say

At its core time is more of a partial order

9:57 AM

yeah, I think that's how they treat them in background independent theories like causal sets if I recall, where it ends up with a lot of directed graphs to represent which direction causality is pointing at each point

in the bizz it is called an etiological space

hmm... so as some toy model I am trying to investigate the foliation of the line of two origins. The naive idea is I can "extrude" it into a cuboid by for every point on the 1-manifold of time, I define some spatial manifold that are open rectangles. However I am still trying to work out whether it is differeomorphic at the origin, as the foliation chart from open sets on the 1-manifold to the collection of open rectangles does not seemed invertible at the two origins

10:28 AM

ok, Let the spacetime $S$ be $\Bbb{R}^2 \times X$ where $X$ is the line of two origins. Then the leaves of $S$ has the form $\Bbb{R}^2_{x} \times \{x\}$ for all $x \in X$. Since each point is mapped to its own $\Bbb{R}^2$ copy for some $x$ (in particular $\Bbb{R}^2_0$ and $\Bbb{R}^2_{0'}$ for those two at the origins), the map is thus invertible. Furthermore, $X$ inherits the standard topology of the real line, thus any open sets are intervals, so it is differentiable. Hence it is a

differeomorphism and a foliation on $S$ is defined

still, that left the question. If a system evolves in time pass through the origins, which origin will its worldline will pass through

A foliation is defined by the fact that you defined it as $\mathbb{R}^2 \times X$

That is already a foliation

ok

link.springer.com/article/10.1007/BF00414289

hmm... so we cannot predict which origin is responsible for the past of a given event, nor we can predict which origin will an event evolve into

How do measurements work in branching classical spacetimes, is it that we measure outcomes within a time interval, and then we conclude the worldline of the system passes through the branch consistent with the measured outcomes?

10:49 AM

Branching spacetimes as they are usually defined are such that every observers remain in a single maximally Hausdorff submanifolds

ergo you won't be able to measure any differences from simply inhabiting a Hausdorff spacetime.

ah...

There's a reason only philosophers seem to really care about it

there was like a one year period where it was mildly popular in GR circles and then everyone dropped it

Actually I am surprised that philosophers care because based on what you said, these branching spacetimes do not really reproduce what is naively expected for things like parallel universes and hence play no role on the debate of indeterminism

and from what I remember, metaphysicians are usually interested in these concepts because of the possibility to use them to ground debates on freewill and choice

but GR branching spacetimes seemed to do neither

Branching spacetime is fun but outside of a very narrow field, it's not really that interesting

it doesn't really offer much new concepts or help with many world in any non-trivial way

It does permit the implementation of modal logic on a geometrical level, if you care about that sort of thing

11:18 AM

I see

3 hours later…

2:43 PM

@Slereah Is there any chance that you can forget my joke (today's one) and we can be like just the way we were before?

2:59 PM

I don't care that much

just ask your GR questions if you wish

What mistake did Hilbert make in his formulations of GR?

Hell if I know

Did your friends call you Sam?

3:45 PM

@Knight I don't think Hilbert made any mistakes in GR, but all he did was propose the action from which the vacuum equations can be derived.

Novalium Company

I'm still a bit triggered by the fact that people like Nick Bostrom and Ray Kurzweil attempt to predict what superintelligence will do... I guess they want some quick bucks from their entertainment stories.

4:09 PM

@JohnRennie Two years ago, I was seeing the series Einstein:Genius. In that Hilbert said “I’ll solve field equations independently” this scared Einstein and both embarked on their way. Hilbert completed his work before Einstein and published it, due to this Einstein went into depression (if that’s the right word) but suddenly he realised that Hilbert didn’t take into his account the changing trajectory of mercury.

Was that thing real?

4:31 PM

@Knight That doesn't match the history as I know it.

4:54 PM

@JohnRennie I have more faith in you than those directors whose only aim was to increase the viewers.

@Knight GR is a big complicated theory, and by 1916 most of it was in place and reasonably well understood but there was one last problem.

The exact form of the equation that links the spacetime curvature to the mass/energy distribution was not know.

Is it the due to inadequacy of mathematics or physics?

Up to this point Hilbert had not been involved with relativity and it was Einstein who had done all the heavy lifting (with help from friends). But now Hilbert became interested.

And he and Einstein simultaneously got the correct form of the equation. What we now call the Einstein field equation.

😱

Novalium Company

The Bayes Theorem is the most confusing thing I've encountered in my life.

4:58 PM

Now this is where it gets murky because it isn't known for certain if they both got the right answer independently, or if Hilbert got it first and Einstein cribbed his answer.

Novalium Company

and I've talked to girls.

There simply isn't the documentary evidence to be sure, so different historians have different views and it's unlikely we'll ever know for certain.

What what is beyond doubt is that Einstein had built the theory and this was just the final step. So even if Hilbert got it first the theory is still quite rightly called Einstein's theory of relativity.

yes.

What Kurt Gödel did in GR?

@Knight In a way getting the Einstein field equations was the easy bit. The problem is they are fiendishly hard to solve. Even now, 100 years later, relatively few analytic solutions are know. For most solutions we have to wheel out a supercomputer and solve the equations numerically.

Gödel managed to find one of the few analytic solutions we know about. His solution described a rotating universe with a cosmological constant,

Gödel's metric isn't of any physical significance because the universe isn't rotating and the cosmological constant is the opposite sign to the one in Gödel's universe. However Gödel's metric is of interest because it contains closed timelike curves i.e. it allows time travel.

So, considering whatever you have said this quote by Sommerfeld fits best here “If you want to be a physicist, you must do three things—first, study mathematics, second, study more mathematics, and third, do the same.”

5:10 PM

@NovaliumCompany The main lesson there is that human intuition is terrible at statistics.

@Knight there is a huge difference between the motivations of physicists and mathematicians. For mathematicians it is the maths that is interesting for its own sake. For physicists it's the physics that is interesting and the maths is just a tool. But if you plan to work in theoretical physics then yes you do need to be very good at maths.

Novalium Company

@ACuriousMind I'm watching 3Blue1brown tutorial now, hope it helps

@NovaliumCompany Why do you assume they are only doing it to make bucks from entertainment stories? There are plenty of reasons it is beneficial to consider the possible outcomes of superintelligence. Just because we can't be sure what will happen, doesn't mean it's bad to consider the possibilities and how we might best react. It's just a proactive approach.

Novalium Company

@JMac ok

With locations as $(x_1 , y_1, z_1)$ and $(x_2, y_2, z_2)$

@JohnRennie Sir, please resolve this doubt of mine. Let’s say I have a 3D space with orthogonal axes names as $\varphi$ $\theta$ and $\alpha$. And there are two charges $q_1$ and $q_2$.

5:17 PM

Why the weird names for the axes?

Why not just $x$, $y$ and $z$?

Because that’s the main problem, I don’t know how but differentiation wrt to x has caused so much pain

Deleted

What calculation are you trying to do?

@JohnRennie "Vargamma, Axis of Weird" sounds like a Magic card ;P

:-)

@Knight I deleted the post with a typo in the MathJax, since you had posted a corrected version. I don't know why it invited you to the Trash room.

Novalium Company

pff :( I watched like, 5 videos on Bayes Theorem and I'm still struggling to understand it. I feel like my brain is being a d*ck.

5:21 PM

@NovaliumCompany if it's any consolation I have never understood it either.

The names of the "axes" would further more suggest they're actually Euler angles, not axes

@NovaliumCompany What do you not understand about it?

@ACuriousMind Ah yes, that certainly sounds probable.

Novalium Company

@AaronStevens everything. I just know the formula and that's it. I know that the theorem allows people to make better judgements and predictions or something like that.

Our estimates about stuff are stinky.

Now, Coulomb’s law will look something like this $$ \mathbf{F} \left (x_1 , y_1, z_1 , x_2 , y_2, z_2\right) = k \frac{q_1 q_2}{ \left(x_1-x_2\right)^2 + \left(y_1-y_2\right)^2 \left(z_1 -z_2\right)^2 } \hat r$$

Ah! Done

How is that?

5:24 PM

@NovaliumCompany I find the examples on Wiki very helpful in understanding its application, much more so than many Bayesian explanations that involve unhelpful (for understanding the actual theorem) digressions about priors.

$\hat r$ is just that denominator with unit length

@NovaliumCompany Yeah, I think I started reading some sort of machine learning thing about 4 or 5 years ago. I remember they covered Bayes' theorem in the simplest scenario. It made sense. And then they suddenly ramped it up to be more general and that is when I got lost haha

I have had that 3blue1brown video in my "watch later" list for a while; I haven't gotten around to it yet

Novalium Company

@AaronStevens I'm actually learning Bayes Theorem to understand Naive Bayes xD

@Knight yes that's correct.

I have always liked the general interpretation of it though: That probability is really just a sense in how much we know about the system

I am sure it can be described more eloquently than that though :P

Novalium Company

5:26 PM

@AaronStevens i like that interpretation

@JohnRennie Now, would $$ \frac{\partial \vec F}{\partial x}$$ gonna be zero?

@AaronStevens That's the Bayesian vs. frequentist interpretation of probability, but Bayes' theorem applies regardless of what interpretation you take. That people think the theorem is somehow dependent on the Bayesian interpretation is a marketing success of Bayesians, but not really true

@ACuriousMind Yes, I agree with that. But I feel like the motivation of that interpretation follows from the theorem

@Knight $\mathbf F$ is not a function of any variable $x$ so you cannot usefully differentiate it with respect to $x$.

@JohnRennie Sir that’s the main problem, what is $x$ after all? It’s a just nickname of an arbitrary point, isn’t it?

5:31 PM

You are mixing up two different things.

Novalium Company

youtube.com/watch?v=FveF-we6lcE ...why

When we are locating things in space we can use coordinates $(x,y,z)$ to identify their positions. But the space in which the function $\mathbf F$ is defined is not this space. It is a six dimensional parameter space $(x_1, y_1, z_1, x_2, y_2, z_2)$.

0

Q: Serial downvoting or duplicate or unclear often on my questions

This is only an example, Ohm's second law: step by step to obtain this law but my profile is always targeted by the same users who each time either give negative reviews, or my question is either unclear, or because my English is incomprehensible, or because it is a fake duplicate. What action...

feature-request close-reasons flagging duplicates vote-to-close

When we differentiate F we are in effect finding its gradient in the parameter space not in our real 3D space.

Novalium Company

So as far as I understand it, Bayes theorem basically says that the absence of additional information is what makes our intuitive probability guessing incorrect?

5:36 PM

@JohnRennie Isn’t $x_1$ is just a designation for an arbitrary real number? Does adding a subscript of 1 changes the meaning of it. (I apologise if my questions are silly)

If you have some quantity that is a function of position (x,y,z) then it makes sense to differentiate it wrt x or y or z.

But F is not a function of position (x,y,z). It is a function of the six variables. $(x_1, y_1, z_1, x_2, y_2, z_2)$. So $dF/dx_1$ or $dF/dy_2$ makes sense, but $dF/dx$ does not.

deja vu

@JohnRennie I liked this point very much. Please elaborate this a little more

If you were calculating the electric field or the potential then they are functions of position (x,y,z) so they could be differentiated wrt x or y or z.

i.e. for constant $x_1, y_1, z_1, x_2, y_2$ and $z_2$ calculate the field or potential as a function of position (x,y,z).

Hey! Does anyone know the form the trace of the extrinsic curvature tensor for the AdS_{4}-Schwarzschild metric and the BTZ metric in AdS_{3}?

5:42 PM

They're playing Hawkwind on the radio. Excuse me for a moment while my brain melts :-)

@JMac Now I am an enemy of critical thinking and freedom of thought

@AaronStevens I saw you trying to restrict freedom of thought. You were so close!

Novalium Company

This crappy video somehow made me understand what Bayes Theorem is about: youtube.com/watch?v=j2tNxIaGpR4

Don't judge the book by it's cover I guess

@JMac Maybe one day I will succeed in restricting the thoughts of others. One meta post at a time I suppose

@AaronStevens Wow you get accused of making negative judgments for flagging the question and asking for clarification; but Johan gets thanked for his support for linking to a main meta question about basically the same thing.

5:51 PM

I do sometimes feel we are a little eager to downvote people who find difficulty in expressing themselves as clearly as we might wish.

@JohnRennie It's all about tone for me. If someone is automatically playing the victim and treating regular users as antagonists, even when they just try to clarify things, I lose patience pretty fast.

@JohnRennie I try not to down vote if I can tell it is just a language barrier. On the specific post they link to I only voted to close as a duplicate; I didn't down vote that one

I didn't mean to point the finger at anyone here. It's just that I feel I frequently see downvotes that strike me as unkind.

I guess, although there is never really any way to know the intention of the down vote unless the user comments

I think sometimes we get users who post an unclear question, perhaps because of language problems or perhaps because they really aren't thinking clearly. But then they react badly to criticism and a vicious cycle quickly develops.

They become that awkward user and we all get less patient with them and criticise more, and down the spiral we all go.

5:57 PM

@JohnRennie I think equating votes on questions with "kindness" is a big part of the problem though. People downvoting posts usually do it because of actual issues with the post, as far as I see. And although leaving a comment might be a sort of "kindness" in that situation, it can also have the exact opposite effect where users see it as a personal judgement, so it can be hard to win for the person who thinks there are issues with the post.

Either you d/v anonymously and are "rude/unkind", or you explain it and then the person considers you their enemy.

@JohnRennie We don't downvote people, we downvote posts. The problem is that the people are often inseparable from the post (both in the minds of the regulars who get exasperated with the posters, and in the minds of the askers, who feel personally attacked).

The way I see it, a big issue is getting users to stop associating votes on posts with judgement of the person.

In an ideal world, we would all be wise and enlightened creatures that never conflated "the post" and "the person who wrote it", alas, we are not.

I usually don't even look at the user who made the post until I am done reading it, making votes, etc. I actually didn't realize that I was familiar with this user until after I voted to close as a duplicate and started looking at comments on the accepted answer.

And then that is when I down voted :P jk

It's impossible to be perfect and silly to try, but there's a large spectrum between perfect and ... erm ... whatever the opposite end of the spectrum is. Perfectly imperfect?

6:01 PM

I think the best solution is to make sure one is eating a mango while on PSE

I tried that and the keyboard gets sticky :-)

You haven't uploaded your consciousness to your keyboard yet?

@JohnRennie The problem is that the only alternative to "downvote and make the asker feel personally attacked" is "not downvote". Which then makes votes less useful as a signal of post quality.

@ACuriousMind not really. It just shifts the origin a bit i.e. zero votes becomes an expression of disapproval rather than neutrality.

Novalium Company

youtube.com/watch?v=IlVINQDk4o8 yes.

when do u guys think this coronavirus will go away?

6:06 PM

Who knows, though with the precautions being currently taken My guess is it will peak soon.

@JohnRennie Not really. With downvotes, there is a clear difference between a bad post with 100 views and -3 score, and a meh post with 100 views and 0 score. Without downvotes, there is no signal to distinguish the two

Novalium Company

@JohnRennie peak?

I thought we're stopping it?

oh nvm

i'm stupid

just... i need sleep

No, right now the change in the number of infected people per day is still positive.

Novalium Company

the slope of the curve? hehe, calculus baby

@ACuriousMind Isn't a three vote difference close to statistical noise anyway? The good questions and answers get tens of votes.

6:10 PM

@JohnRennie In the tags with much traffic, yes. In the smaller tags, such low vote counts are often significant in my impression.

But even if three upvotes are statistical noise sometimes, three downvotes are definitely not (unless the question is HNQ)

three downvotes mean there is something at least controversial about the question

@JohnRennie 3 votes different might be statistical noise when comparing +6 to +9, but a -3 score answer seems to almost always have obvious issues.

Oh well. I'm going to put this burning issue behind me and go read some science fiction. And I might even have a cold beer.

Novalium Company

@JohnRennie siims cul

bowchicawowow

Novalium Company

that's what ma mama said

6:19 PM

@AaronStevens Sir, you are not a welcome user to me. — Sebastiano 2 mins ago

oops

@AaronStevens That's the type of comment I really struggle with. Is he trying to say you're not welcome to participate in his posts, or your haven't been welcoming to him?

Presumably it's the latter, given how much you want to stifle free thinking; but I really can't tell.

Well, I think he is making fun of the fact that my application to be a "welcome user" was declined.

@AaronStevens Is that like a SE Walmart greeter?

@JMac Yeah pretty much. The role consists of sitting at the computer waiting for certain users to create posts, comments, etc. Whenever you see such activity from such users you are supposed to welcome them to the site

I also find it odd to be upset about being closed as a duplicate. Being closed as a duplicate is the best way to get closed because you still get your question answered. All other closing reasons don't give you an answer

Which makes me feel like this user isn't asking to learn, but asking just to have upvoted posts

6:48 PM

I'm removing a problematic comment discussion, including an apology that wouldn't make sense out of context. Take a walk outside, friends. — rob ♦ 17 mins ago

"Including an apology" what happened? I was not online, who apologized to whom?

@Knight The point of removing such discussions is so that they are not publicly visible.

2

@ACuriousMind Yes, I too think that

Then why ask?

curiosity killed the cat

(I hate that phrase)

@enumaris I mean really, who is the cat? I've seen more than one cat, I'm not sure which one that saying is even talking about!

6:56 PM

for real

@ACuriousMind I thought something ground breaking had happened, but your cool reply told me that things were quite usual

Novalium Company

7:08 PM

AAAAAAAHAHAHA

I wanna chat with someone

please

I'm bored

Do you Know me, I'm your Nearest Neighbour.

i'm fun

knn is fun :)

2 hours later…

Novalium Company

8:51 PM

@enumaris im funner than knn

9:05 PM

sounds legit

3 hours later…

11:53 PM

@NovaliumCompany Only one thing is certain, a super intelligence will be even more goal oriented than us

eh...

@Secret For some of us, that's not a high hurdle to jump

« first day (3381 days earlier) ← previous day next day → last day (1550 days later) »

Transcript for

Feb '206

The h Bar

General chat for Physics SE (physics.stackexchange.com). For M...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile