The h Bar

Sir Cumference

12:29 AM

The real struggle

Sir Cumference

1:26 AM

quora.com/…

Seems intergalactic travel will just remain a fantasy :/

dmckee

1:46 AM

Without some fundamental breakthroughs (in physics, not just technology), even sending corporal humans to the nearest stars will require resources significant on a solar-system wide scale. And it is thousands of times that far to get to anything that could be called "intergalactic space" even at a stretch.

vzn

2:33 AM

@Blue miss the old liveliness of this room + others too. suspect SE-wide chat activity is down but dont know any easy way to quantify it. wonder if they (SE) are not running main page links to chat as much as previously (probably the main way that new users find it). speaking of SE features, would like to see the chat rooms more prominently indicated to attract new users.

Sir Cumference

2:47 AM

@vzn I mean imo some of the biggest we are missing are obe, 0celo, Balarka, Heather, Bernardo, Kaumudi, Coopercape, Obliv, etc. Without them the chat's a lot quieter

Most seem to have disappeared when the semester started

dmckee

3:08 AM

My unquantified memory of past events in here suggests that complaints about intro-level homeworky stuff have mostly arisen when there has been a steady stream of it so that it is what you see every time you peak in.

When it is just one or two a week no one seems to say anything.

I suspect JEE became the focus of discussion along those lines in part because it has a season of sorts.

Emilio Pisanty

4:13 AM

@dmckee how do you get humans to intergalactic space without some form of suspended animation?

Or do you just rely on relativistic time dilation?

dmckee

@EmilioPisanty I don't think it is possible. The energy problem and stuff ramming us problem make highly relativistic speeds mindbogglingly hard.

If we had working ramjets it becomes conceivable that some descendant civilization could find themselves on the fringes of the galaxy, but the required time scales are vast and who knows what kind of beings those would be.

vzn

5:24 AM

@SirCumference people come and go all the time over the yrs, as its aptly named... the h bar. do miss the colorful predecessors at times. my concern is also "replacement rate/ conversions". it has to do with an old dotcom concept called "stickiness" (aka recruiting/ engagement). alas it seems sometimes few others fully take it into account...

Emilio Pisanty

@dmckee I was thinking more in terms of intergalactic space being many tens of thousands of light years away, which puts a lower bound on the travel time that's much longer than a human lifetime

but then again if you're going fast enough (assuming the energy and stuff-in-the-middle problems have been magically solved) I guess you can get there in sub-human-lifetime proper time

the ellapsed time on Earth will still be longer than the current human civilization, of course

but you're still alive, so I guess that's a plus?

Anonymous

7:17 AM

> complaints about intro-level homeworky stuff have mostly arisen when there has been a steady stream of it so that it is what you see every time you peak in.

Anonymous

Unfortunately, that's exactly the kind of feeling I sometimes get these days. I'm hoping that it will improve with time, however. :/

Anonymous

@SirCumference Yeah, that's very true. Those guys were pretty fun and knowledgeable.

Anonymous

I wonder if we should hold homework and intro-level questions to the same standard as on the main site, in here (show your effort --- have you done basic Googling at least?, tell us what you researched, use proper formatting (using MathJax) and correct spellings and grammar, don't ask about asking --- just ask, etc.). Otherwise they could be sent off the problem solving room which is more suited for beginners. Anyway, I will write about this on meta someday.

Anonymous

For example:

Anonymous

in Problem Solving Strategies, 20 hours ago, by Arjun

Someone pls pls pls help me with this :- https://www.reddit.com/r/Physics/comments/abt3o0/how_to_calculate_the_distance_w‌ith_only/

Anonymous

7:27 AM

I couldn't really think of a polite way to say - "Your question is nonsense. Please go elsewhere."

Anonymous

Perhaps I'm spotting these things a bit more these days because I recently had to write up the How to write a good question? thread on QCSE, after getting fed up of having to edit every alternate question on the site.

user438666

8:33 AM

Hi I've never been to this chat, how does this work? If I have a small question that doesn't seem to deserve its own post is it ok to ask here, or this is not the purpose of the chat?

Loong

> Don't ask about asking, just ask.

user438666

Understood. About anti-commuting spinors: how come we can make sense of an expression such as $\overline{\psi}\psi + \psi \overline{\psi}$ when it seems to me that the first term is a scalar while the second is a matrix?

whoops. The chat doesn't seem to render math as I expected to

John Rennie

@user438666 see meta.stackexchange.com/questions/97938/…

user438666

Yes, thank you, I saw it in the sidebar just after sending the message, should've looked beforehand

David Z

@user438666 That's generally fine. Here we have very loose restrictions about what can be posted - pretty much anything goes as long as other people in the chat don't object, but physics discussions take priority if the room gets busy. (Of course the SE network code of conduct is still in force.)

Physics Meta

10:46 AM

0

Q: How can we improve the participation of subject experts and enthusiastic physics students in chat?

I don't use the main site much, but I have been a regular participant on Physics SE's main chat: The h Bar, for the past two years. Lately, I have been noticing a drastic drop in the number of subject experts i.e. grad students and professional physicists, visiting the chat. I know that some of t...

Avnish Kabaj

12:35 PM

Oh noes

Blue has become a moderator truly

In true might have worked there :/

ACuriousMind

@user438666 In what context have you encountered that expression?

user129412

12:52 PM

Anyone know if there have been groups that combined electrostatic gating and 3D readout cavities? Say for a nanowire transmon device, or a gate defined quantum dot, something like that. Because I am trying to think of a way to feed the gatelines into the cavity.

GodotMisogi

Problem 2-6 in Feynman and Hibbs' Path Integral and Quantum Mechanics has been messing with me. It considers a unit mass particle moving at light speed in natural units, and investigates the possible paths between points $a$ and $b$ on a spacetime diagram such that 'reversals' $c = 1 \to c = -1$ on the path occur on the boundaries of the steps of some time-discretisation $\epsilon$

Slereah

@GodotMisogi Hey guess what

During my master thesis I did the solutions of the first few chapters of Feynman and Hibbs

I can give you that, lemme see

GodotMisogi

So far, I've been able to reduce it to a combinatorial problem on the number of walks from one corner to the other of a rectangular grid of size $m, n$. The question asks for the number of possible paths as a function of the number of reversals (i.e. the corners taken during the walks), which looks $(n)$difficult$(n+2)$me to derive

@Slereah you're a lifesaver

Slereah

It's in French but there it is

GodotMisogi

Trés bien, je veux pratiquer mon français

user438666

1:02 PM

@ACuriousMind my QFT lecture notes, when taking anticommutators of Dirac field to quantize it. However, I think it's just a shorthand notation for the anticommutator component by component.

ACuriousMind

Ah. Yes, it is.

user438666

It took me some time but I figured it out after having already asked it, thanks

GodotMisogi

@Slereah Merci beaucoup. Looks like I was on the right track with the binomial coefficients.

I wish I could count

In other news, I spoke with @Blue and I think I should share my experiences and (noob-level) understandings of talks taking place during the Black Holes, Inflation and Gravitational Waves workshop at HKUST here. There's also a conference on High-Energy Physics taking place next week, so I can spam interesting talks here, if anyone's interested in any of the speakers' topics (which I can only share once the schedule's released)

Slereah

if you need more feynman hibbs stuff I've got the solutions up to 5.3

GodotMisogi

@Slereah Oh, thanks for the offer! But I think I've done most of the others (up to Chapter 6 anyway). I'd left Problem 2-6 halfway for over a year, and just remembered it, so I thought I'd ask here

So one of the talks today was by Ping Gao from Harvard about creating traversable wormholes in your kitchen: arxiv.org/pdf/1608.05687.pdf

Slereah

1:25 PM

I don't have any phantom scalar fields in my kitchen

2

GodotMisogi

I understood mostly nothing from it, due to my lack of experience, but does anyone know what type of conservation laws arise from these kinds (wormholes, observers outside of the universe, etc.) of considerations? The first idea that troubled me was stress-energy tensor conservation breaking

Slereah

1:50 PM

Conservation laws are the same as in GR in the case of wormholes and whatever

Are you talking about the... conformal anomaly

or whatever

it pops up a lot in that kind of business

GodotMisogi

x.x so sleepy. IIRC, he considers two interacting CFTs, and the quantum matter stress tensor, so...I guess that's relevant?

bolbteppa

Renormalization is madness

GodotMisogi

@bolbteppa That's what I thought too, even with an introduction from a Wilsonian perspective

Slereah

The conformal anomaly is a thing where

Typically, in a CFT, the trace of the stress energy tensor is 0

$$\mathrm{Tr}(T) = 0$$

But this may not be the case in a quantum version of a CFT

bolbteppa

Renormalization even shows up in electromagnetism, in mass renormalization arxiv.org/ftp/arxiv/papers/1310/1310.5533.pdf we already have infinities :'(

Slereah

2:05 PM

Where you get $$\mathrm{Tr}(\langle T \rangle_\psi) \neq 0$$

This quantity is called the trace anomaly

and it's a device used quite a lot in QFT in curved spacetime because it's usually very easy to compute

You can get the VEV of that spacetime in that theory easily enough

which is nice to show divergences or whatever

bolbteppa

I wish I seen this basic example a long time ago, man

Slereah

Mass renormalization for EM stuff is old shit man

That was the pre-special relativity theory

with electromagnetic mass

I mean not quite renormalization yet

but it was hinted at

bolbteppa

Apparently they basically just copied that old idea when they applied it to qed, I think that's the whole game, just copying this example in some form

Slereah

well you know

bolbteppa

In one way it makes sense, in another it's sheer madness

Slereah

2:13 PM

That's why I don't really mind theories that don't go nowhere

people always think it's a bad idea to waste time on theories that turn out to be false

But there's always ideas you can pick up from there to reuse later on

I'm fine with funding whatever stupid pet theory

as long as they're proper physics

ALso @bolbteppa that paper is awful

Was it typed in Microsoft Word?

bolbteppa

I think it's just a weird font, seems like he's using latex equations

Slereah

nah

it looks like Word equations

something is off

bolbteppa

That example in section 2 is actually unbelievable, I can't believe that's where renormalization came from

Slereah

There's plenty of stuff you never really learn about that was like

bolbteppa

This is good in concert with that arxiv.org/pdf/1403.5622.pdf sec I and IV

Slereah

2:16 PM

the huge thing of the era

Which is like

It has pros and cons

don't need to clutter classes with a lot of theories that aren't useful anymore

But on the other hand it's nice to know

GodotMisogi

@Slereah Definitely looks like Word, and it's terrible

bolbteppa

Yeah it depends on the topic, sometimes it's insanely useful, other times it's silly

Slereah

you know what I'd like to learn more about?

19th century theories about curved spaces

Apparently there's a bunch of 'em and nobody ever talks about them

bolbteppa

Like what

Slereah

Like Schwarzschild was doing $\mathbb{R}P^3$ cosmology 10 years before GR was a thing

for some reason he thought the universe being $S^3$ was ridiculous but projective space was entirely reasonable?

bolbteppa

2:24 PM

They all loved projective space for some reason

Slereah

like there was a short period before GR when people had the vague idea that maybe curved space would solve the Olbers paradox

bolbteppa

"Gell‐Mann and Low [25] give the following physical interpretation of charge renormalization:

A test body of "bare charge" q₀ polarizes the vacuum, surrounding itself by a neutral cloud of electrons and positrons; some of these, with a net charge δq, of the same sign as q₀ , escape to infinity, leaving a net charge ‐δq in the part of the cloud which is closely bound to the test body (within a distance of $\hbar$/mc).

If we observe the body from a distance much greater than $\hbar$ /mc, we see an effective charge q=q₀ ‐ δq, the renormalized charge. However, as we inspect more closely and penetrate through the cloud to the core of the test charge, the charge that we see inside approaches the bare charge q₀ concentrated at a point at the center"

The fact one can get really good pictures of renormalization with the Dirac sea is unbelievable as well

Slereah

Well you know

There's always like a dozen different interpretations and formalisms for every theory

which is nice to have

bolbteppa

Yeah, but the sea picture is just crazy haha

Slereah

That's why I'm never too swayed by people talking about a theory and saying this is the NATURAL IDEA FOR THIS

There can be NO OTHER

Jake Rose

2:49 PM

With the uniqueness theorem for eldtrostatics, the proof often requires that the charge distributions be the same, so why can we use the uniqueness theorem on a dipole in a uniform field and a sphere in a uniform field? They clearly have two different charge distributions

Or is it only the charge distributions in the region that we’re talking about that matter?

danielunderwood

3:37 PM

@Slereah yeah, check out the page title "Microsoft Word - A Critical History of Renormalization.docx - 1310.5533.pdf"

Slereah

3:53 PM

Bad move

Word papers raise immediate crank alarms

dmckee

In principle using Word shouldn't be problem. But it suggests that the author is (a) from outside the field and therefore not familiar with the culture, (b) an iconoclast, or (c) technologically incompetent enough to be unable to learn the more usual tools.

Each possibility offers a different reason to worry:

(a) Not familiar with the culture suggests a real chance that the author is also unfamiliar with some important part of the background knowledge of the field.

(b) Iconoclasts are willing to stand up for their own opinions in the face of widespread disagreement. This is good thing when they are right, but it also means that they may be unwilling to recognize when they are wrong.

(c) Unless the block here is very domain specific, you find yourself wondering just how smart this joker really is.

None of those should be damning, but ... yeah, I take word as a cue to look at other meta information before investing any time in the document.

bolbteppa

4:44 PM

The guy is not a crank haha

It's this guy

amazon.com/Statistical-Mechanics-2nd-Kerson-Huang/dp/0471815187

Kerson Huang

Kerson Huang (Chinese: 黃克孫; pinyin: Huáng Kèsūn; 15 March 1928 – 1 September 2016) was a Chinese American theoretical physicist and translator. Huang was born in Nanning, China and grew up in Manila, Philippines. He earned a B.S. and a Ph.D. in physics from the Massachusetts Institute of Technology (MIT) in 1950 and 1953, respectively. He served as an instructor at MIT from 1953 to 1955, and subsequently spent two years as a fellow at the Institute for Advanced Study. After returning to the MIT faculty in 1957, Huang became an authority on statistical physics, and worked on Bose–Einstein condensation...

Slereah

I know

red flags does not mean he is

bolbteppa

Renormalization is cranky

Physics Meta

-2

Q: What is the general age of users on Physics SE?

I want to what is the age of users in general. I want to ask this because age determines so many things. A student of age 30 can be far better than a student of age 17. I know it’s personal information which you will not want to share but I just want a rough idea.

discussion users

vzn

@bolbteppa lol! not an expert on this but apparently feynman nobel prize winner + other elite physicists had a lot of involvement in renormalization ideas. think it is a "highly questionable verging on suspicious" part of modern theory/ standard model. btw the charge cloud ref reminds me of density version/ formulation of QM...

bolbteppa

@vzn try read section 2 of the link I sent above

To get a sense of it, doesn't matter where the equations come from or why they work, just look at what they do

vzn

5:00 PM

@bolbteppa the Huang book on amazon?

bolbteppa

This: arxiv.org/abs/1310.5533

vzn

@bolbteppa reminds me a little of the bohr model. probably formulated not too far apart in time/ history.

bolbteppa

The failure of these formula's on a quantum level (e.g. the Larmor radiation formula) led to the Bohr model

vzn

ch.8 of the book, something youve expressed aversion to: quantum statistical mechanics...

bolbteppa

Never had any aversion to it haha

vzn

5:09 PM

@bolbteppa its in the transcript. dont feel like digging it up right now.

(lol) the mainstreamers in here arent gonna like this guy. translations of the I ching lol ... but !!! wow look at this, off his wikipedia pg (which doesnt mention it directly!) A Superfluid Universe books.google.com/books/about/…

ACuriousMind

@Slereah The trick is finding the natural framework in which the natural idea is also the correct one ;P

vzn

> This interesting book provides the physical and mathematical background for a theory describing the universe as a quantum superfluid, and how dark energy and dark matter arise. Presenting a novel theory spanning many different fields in physics, the key concepts in each field are introduced.
> The reader is only expected to know the rudiments of condensed matter physics, quantum field theory and general relativity to explore this fascinating new model of dark matter and dark energy as facets of a cosmic superfluid.

bolbteppa

He was an MIT professor who wrote one of the canonical stat mech books, turns out mainstreamers aren't all holding up the pillars of the ivory towestablishment 24-7 :p

vzn

he also worked at IAS. very impressive. never heard of him. ("its a big universe.") thx for bringing to attn, already bookmarked the book for my next epic blog on the fluid paradigm lol wink :) :P

enumaris

fun times

ACuriousMind

5:18 PM

@enumaris Was that a password? :P

enumaris

probably

apparently my mouse didn't click when it should ahve

ACuriousMind

Fun fact: All moderators can see deleted message, so if you're paranoid you should change that one now ;)

enumaris

I'll just add a period to it, it's all good

Sir Cumference

@enumaris Telling the mods how you will change it defeats the purpose :P

ACuriousMind

@SirCumference I think that was the joke :P

enumaris

5:23 PM

:o

Sir Cumference

Yeah I'm slow to the uptake

enumaris

pikachu surprised meme

so the hats are over eh

it's still winter tho

Loong

So we go to Kuusamo and have six months of Winterbash hats each year?

enumaris

It's only winter for like 3 months

Loong

Not in Kuusamo

danielunderwood

5:37 PM

I've almost put in a password both here and Discord...luckily I caught them just before pressing enter

user438666

isn't it winter somewhere in the world for 6 months during the year?

enumaris

somewhere

@danielunderwood I type too fast for my own good

newUser

Hello everyone. Is there a way to send a private message to a user of Physics SE?

ACuriousMind

@newUser No.

newUser

@ACuriousMind Neither with a private chat room?

@ACuriousMind A user posted a long answer to my question some days ago. I wanted to read it, but today he/she removed it. I would like to know why

ACuriousMind

5:49 PM

@newUser There are no private chat rooms for ordinary users. Moderators can set up private rooms that only specific users can see, but this is exclusively reserved for moderation activities

Sir Cumference

(feature-request)

ACuriousMind

(status-declined)

5

Sir Cumference

:(

ACuriousMind

Really, this has been declined on mother meta at least half a dozen times :P

newUser

what?

danielunderwood

5:52 PM

I feel like having private chats may cause the bigger users to be pestered

Or another avenue for crackpottery

Sir Cumference

Oh speaking of meta, have thoughts on Blue's post @ACuriousMind?

newUser

@danielunderwood You are right, but everyone could choose to not receive private messages. And the problem gets solved.

Anonymous

@newUser If by "private", you mean "not visible to other users", then no. But there's definitely an option to create one-to-one chat rooms (simply find that user's chat profile and you'll get the option in there). However, keep in mind that most main site users hardly ever visit chat.

newUser

@Blue user's chat profile? what is this? do you mean the network profile?

Anonymous

@newUser For example: this is your chat profile.

ACuriousMind

5:56 PM

@SirCumference I like everything in it that's not related to formalizing chat further: E.g. If people have good ideas to revive the chat sessions, I'm all for it.

Anonymous

@newUser You can search for the chat profiles of specific users from here.

ACuriousMind

I have reservations about formalizing chat with more rules which I'll post on meta as soon as I've figured out how to express them :P

newUser

@Blue if you have a SE profile doesn't mean you have a chat profile, right?

@ACuriousMind great ;)

Anonymous

@newUser That's true. But often they do.

newUser

not in my case :) thanks anyway

Anonymous

6:03 PM

@ACuriousMind I'd love to hear! :) At the moment, I'm leaning towards creating a "room-rules" list though, like Chair suggested.

enumaris

fun beans

danielunderwood

wooo I just cleaned out/organized 3 years' worth of random poorly-named downloads...now to the other machines

enumaris

o.o

u write a script fo dat?

ACuriousMind

The download folder is a terrible place best left undisturbed

enumaris

everytime I visit the download folder I feel an intense shame

2

danielunderwood

6:11 PM

hmmm I suppose a script for it would be helpful. I just poked through everything and opened up code/pdfs

Sir Cumference

I get nostalgic since I still have downloads from 10 years ago in that folder :P

danielunderwood

Though pdfs were more of just making sure they were in mendeley. Tons of things of the format 1234.5678.pdf from arxiv

I don't think I have an os install that's that old, but I do have files that have been floating around on flash drives and such for many years

enumaris

write a script fo dat

danielunderwood

That's probably a good idea

Doing the downloads on my Windows machine is going to be interesting...the folder takes a good 20 seconds to load contents for whatever reason. I suppose maybe they don't index the download folder or something

enumaris

dunno

You're more tech savvy than me

danielunderwood

6:32 PM

Maybe they don't expect anyone to actually go to the downloads folder

@enumaris have you ever heard of anyone doing an RNN for noise cancellation? I've been wondering how it would compare to whatever the "classical" way of doing it is

enumaris

I have not heard it

you'd need a way to simulate noise

newUser

What do you think about the extremely excessive length of the papers on arxiv?

enumaris

And then you'd need a way to simulate how to cancel that noise

danielunderwood

Yeah I didn't even think of what the training data would be...seems trickier than I had originally thought

enumaris

it's a bit different than standard ML problems in that you could build a system to "cancel" the noise in a recording

but you want to build a system that actively outputs sound to cancel ambient noises for a user

which is a different problem than canceling noise from a recording

de-noising is pretty standard ML fair though

I'm not sure how much of the process is really ML and how much is just acoustics

ACuriousMind

6:55 PM

@newUser The lack of length restrictions on arXiv is a blessing and a curse. It makes it possible for people to actually say as much as is need for a good explanation, but it also allows them to waffle endlessly. Some long articles are good, some are bad

newUser

@ACuriousMind I think instead that long articles, except few exceptions, are harmful for research

For example, the paper by Nima about spinor helicity amplitudes for any spin and masses is too long and written very badly (in my opinion)

enumaris

if you like short, try physical review letters

newUser

this does not help the research going on.. or am I wrong?

enumaris

there's a 4 page maximum there iirc

:D

ACuriousMind

@newUser Just that you've found one long bad paper doesn't mean all long papers are bad, does it? :P

newUser

7:00 PM

@ACuriousMind I just gave an example of a potentially milestone paper :D I wanted to know your opinion, not just complain eheh

enumaris

going to work while being sick since I already used up my 2 sick days...feelsbadman

Anonymous

7:49 PM

@DanielSank There are a few unanswered superconducting-quantum-computing questions (and some without accepted answers) on our site. It would be great if you could consider answering some of them. :)

enumaris

8:45 PM

mmm had pho for lunch, pretty good

danielunderwood

10:08 PM

Sounds much better than my instant ramen

Curio

silly question

Can I write k = 30,7552N/m +- 3,0965N/m?

Shouldn't I write 31N/m +-3N/m?

I read that the error must have only one significant figure

Sir Cumference

10:27 PM

@Curio I was taught in lab that the error should have the same number of decimal points as the measurement

rob

@Curio Some people work very hard to get two significant figures on the uncertainty.

But that's usually it.

Curio

@SirCumference @rob But at the same time writing for example 5,3174+-1,2 doesn't make sense

rob

Any digits in your central value that are insignificant figures in your uncertainty are, by definition, insignificant in your central value.

Curio

In fact I should write 5,3174 +- 1,2111 (if this was the original number)

rob

Right: it's 5.3 +- 1.2.

Curio

10:30 PM

I mean, they should have the same number of digits after the comma

Sir Cumference

@Curio Yep, this is what I meant to say

rob

@Curio Yes, but it's the uncertainty which determines how many.

Curio

But for example:
5.2 and 1.4839

Should I write 5.2000+- 1.4839 or 5.2 +-1.5

The first I guess

Sir Cumference

Sigh I need to pull out my old lab book

rob

Absolutely the second: 5.2 +- 1.5

Curio

10:32 PM

The second? Why?

Sir Cumference

The former implies you are very precise with 5.2000

And that your uncertainty is very precise

rob

When I teach undergraduates about uncertainties and significant figures, I like to tell them to use the "anger management" method to estimate uncertainties.

Curio

anger management?

rob

For example, if I ask someone how tall they are, they reply (in the US) to the nearest inch: five foot ten.

enumaris

@rob they'd reply 6' for sure

Curio

10:34 PM

Never used feet

rob

Suppose I challenge them: I think they are really five foot, 9.8 inches tall.

Curio

:P

rob

They shrug. Not a big deal.

Suppose I challenge them: I think they are really five foot two inches.

This makes them mad.

So the implicit uncertainty in a height measurement is apparently an inch or two (or a few centimeters).

enumaris

probably they'd be mad you said they are <6' tho

rob

If you really try to measure a person's height, you run up against all sorts of little piddly things that most people ignore.

Curio

10:36 PM

I mean

If the error has 5 digits after the comma, the other number must have 5 digits too

rob

Shoes or no shoes? Soles are typically an inch.

Tall haircut?

Time of day? The squishy parts of your spine compress a little if you spend a whole day upright.

Curio

5.0000 +- 0.0001
5.2712 +- 0.3189

rob

We elide all of those errors by only reporting a height to the nearest inch/centimeter.

And without a lot of sophisticated analysis, we don't know the third significant figure on the uncertainty.

Curio

If the first is right but the second not, then the error must have only 1 significative digit

But I've seen even +- 0.24

rob

@Curio What kind of a problem are you doing where you are getting uncertainties with four significant figures?

Curio

10:38 PM

That should have been +-0.2 then

rob

@Curio There's decent reason for that:

enumaris

errors are generally reported with 1 or 2 significant figures. Don't think I've ever seen 3+

rob

Suppose you're computing an uncertainty and you get +- 0.09

I do the same computation with similar but not identical data, and I get +- 0.11

Curio

I see

rob

If we round to "one significant figure," then I lose precision.

So in one-significant-figure calculations, it's a good idea to include a "guard digit" if your first/only significant figure is a one, or maybe a two.

Curio

10:41 PM

OK

So I have this mass 0,050kg+0.001kg
Its force is 0,49050N +- 0,00981N isn't it?

rob

The Particle Data Group (for whom uncertainties are the main currency) reports uncertainties to two significant figures, with a third guard digit if the first two digits are smaller than 35 (I think --- it's in the introduction to the big book)

By the PDG rules, 5.2712 +- 0.3189 would be recorded as 5.271+-0.319, but most people would write 5.3+-0.3.

@Curio That seems problematic, because you have increased the significance in your uncertainty.

Curio

@rob can't I keep that value?

I mean, I can't write 0,01

rob

@Curio Why not? You started with 0.01.

Curio

With 0,001

rob

Here's another anger management technique: suppose you're planning a trip to the grocery store.

Curio

10:49 PM

I see

rob

In the US, most shops add a "sales tax" at the cash register, which is a fraction of the purchase price that varies by locality. Mine is 9.5%, so if I get a candy bar that say the price is \$1, then I'll pay \$1.10 at the register.

You and I are planning a party. We have a shopping list. You ask me what it will cost. I say, "about \$100."

You think in your head, "that means the sales tax will be \$9.50"

Curio

So +-10

Yes

rob

You get to the register and the total is not exactly \$100, and the sales tax is not exactly \$9.50, and you are not mad.

Curio

Am I not mad?

rob

If you start with a crummy estimate, you can't add significant figures for other things that you know well.

@Curio Well, if you looked at a shopping list and thought it would cost \$100, but the actual price was \$94.32 or \$106.23, would that make you mad?