The h Bar

ACuriousMind

00:05

I...might have gotten a bit carried away answering that

skill patrol

+1

or, if you want to get carried away +0.999... :P

Bernard Meurer

C++ is killing me

skill patrol

Try C+– it's more balanced

:-)

dmckee

@BernardMeurer YOu asked the other day if I do C++. The answer is yes, but not really by choice.

Bernard Meurer

00:21

@dmckee I feel your pain, I'm forcing myself to learn it

I'm just too much of a Python kid, all the types are boggling my brain

dmckee

@BernardMeurer Oh, I've used it for so long that much of it is second nature (through the '08 standard and parts of the '11 standards, but not so much the '14 stuff).

But it is neither an elegant language nor an easy one, so I use it because it is the language of ROOT and Geant4.

Before that I used fortran 77 because it was the language of CERNLIB and GEANT3.

Bernard Meurer

I have NO clue how my stuff is working

it just works

and that worries me

dmckee

::chuckles:: It should.

Bernard Meurer

also arbitrary precision is a pain to implement

dmckee

@ACuriousMind I think it is good. Not only because you emphasized the fact that there is policy, but also because you tell the readers that they can be the site if only they will put in the effort.

@BernardMeurer Er ... never done it by hand and wouldn't except as an exercise or if I had already tried all the available libraries. There must be some to chose from for C++, no?

Bernard Meurer

00:33

Using Boost right now

it's good, I'm a mess though

dmckee

Waaaaaay back I wrote a little package in pascal, but even implementing the trig functions bogged down.

Bernard Meurer

I feel like once this project is over I'll be past the initiation

dmckee

And I looked at writing one in m68k assembly, but it was such a big task that I chickened out.

Because the m68k has BDC instructions that would have made it soooo sweet.

ACuriousMind

@dmckee I just tire of having to agree or disagree with the same ~10 reviewers ;) I've also resolved to leaving a link to the specific close review in the comments everytime I see users arguing about the on-topicness of a question. It might nudge a few into doing some reviews.

So far, on a sample size of one question and two users, I got one to do ~10 reviews. Not a bad start.

Bernard Meurer

@dmckee If I get another error regarding my pointer not pointing right I'll kill myself

dmckee

00:40

Yeah. With my mod-hammer and no recourse to a lesser weapon I try not to weight in except as the last voter or in the most clear-cut cases, but I recall that feeling from the second year of Stack Overflow.

@BernardMeurer I grew up on pascal and c: I have pointer handling powers that lesser mortals can only dream of.

And I still shoot myself in the foot with them

Bernard Meurer

@dmckee Will you look at my code if I promise you a beer someday?

c:

ACuriousMind

@dmckee I just realized that a hammer is really the appropriate metaphor because unlike a sword you can't dull its edge or something similar.

dmckee

BTW: those pointer handling powers actually consist mostly of drawing lots of boxes and arrows on pads or whiteboards. There really isn't a substitute for walking through the code by hand when it comes to seeing how your pointers are breaking.

It's tedious, but it is the only way.

Sir Cumference

01:20

How come when we calculate the energy of a neutrino, we don't use $E=mc^2$

But rather $E=\frac{mc^{2}}{\sqrt{1-\frac{v^{2}}{c^{2}}}}$?

Joshua Lin

Because $E = mc^2$ is an approximation from special relativity for slow-ish things, $E = \frac{mc^2}{\sqrt{1-v^2/c^2}}$ is used in general

Sir Cumference

01:39

@JoshuaLin I thought $E^2 = m^2c^4 + p^2c^2$ was the full equation?

So what is it now, $E^2=\frac{m^2c^4+p^2c^2}{1-\frac{v^{2}}{c^{2}}}$?

Jesus

dmckee

@SirCumference No they are both correct. Note that they apply when you have different information.

Sir Cumference

So, don't use that huge combination I just posted?

dmckee

$E(m,p) = \sqrt{m^2 + p^2}$ ($c=1$ units).

$E(m,v) = m/\sqrt{1 - v^2}$ (likewise).

Sir Cumference

So that equation I posted wouldn't work?

dmckee

But in many cases momentum is a more sensible variable than velocity (which people just hang onto because they are used to it from Newtonian mechanics).

Sir Cumference

01:47

$E^2=\frac{m^2c^4+p^2c^2}{1-\frac{v^{2}}{c^{2}}}$

dmckee

@SirCumference Do you have the momentum? Or just the velocity?

Sir Cumference

Let's say we have both

dmckee

@SirCumference Uh, no that is completely and always wrong.

Sir Cumference

K good

So could I use $E=\frac{mc^{2}}{\sqrt{1-\frac{v^{2}}{c^{2}}}}$ to find the energy of a photon?

dmckee

No.

Because a photon has no mass. You'd get zero.

Sir Cumference

01:49

I would have to use the other, right?

dmckee

The relationship $E^2 = m^2 + p^2$ is fundamental. It come from the geometry of space-time.

Sir Cumference

$E^2=m^2c^4+p^2c^2$

dmckee

The other one is derived and the derivation depended on having a massive particle.

@SirCumference Yeah. But I'm a particle physicist, I live in a $c =1$ universe.

Sir Cumference

But will $E^2=m^2c^4+p^2c^2$ work if I know the momentum of a neutrino, but want to find its energy?

dmckee

By understanding what dimension each quantity has you can stick the $c$'s back in later if you really need them.

But mostly you don't.

@SirCumference Yeah. Sure.

Sir Cumference

01:51

Cool

dmckee

But neutrinos are mostly ultra-relativistic, so you can approximate $E \approx pc$.

Sir Cumference

True

dmckee

The only notable exception is relic neutrinos.

Sir Cumference

So ${\sqrt{m^2c^4+p^2c^2}} = {\frac{mc^{2}}{\sqrt{1-\frac{v^{2}}{c^{2}}}}}$

Am I right?

dmckee

For massive particles, yeah. Though it'll be algebraiclly easier if you square both sides.

Sir Cumference

01:54

So then ${m^2c^4+p^2c^2} = {\frac{m^2c^4}{1-\frac{v^{2}}{c^{2}}}}$

Cool

@dmckee Wait, you said the first equation is used if you know the momentum, and the second equation is used when you know the velocity, right?

user54412

@dmckee We all do as far as I care. Some people are just in denial ;)

Sir Cumference

But couldn't I just use the velocity to find the momentum, through $\gamma(\mathbf{v}) m_0 \mathbf{v}$?

Then why would I need that second equation?

dmckee

02:35

@SirCumference Because the relationship between energy, mass and momentum is fundamental and the one between energy, mass and velocity is not: it only applies to massive particles.

By calling $E^2 = m^2 + p^2$ the "second" equation you get the priority the wrong way 'round.

$\frac{m}{\sqrt{1 - v^2}}$ is less general and less useful because the other one (written in the proper form $m^2 = \mathbf{p}^2 = E^2 - p^2$) tells you that "mass" is a Lorentz invariant.

I'm going to get on one of my (many) hobby horses here.

The treatment of special relativity that you find in pop-sci sources is not helpful, because it blinds you to the powerful tools at your disposal and instead focus on the things that will "wow" you (meaning confuse you).

You're much better off using a treatment that focuses on giving you the tool that you will be comfortable with first and showing you the odd consequences second.

Tatsu Takeuchi's book is really good: amazon.com/Illustrated-Guide-Relativity-Tatsu-Takeuchi/dp/….

Sir Cumference

Huh, all right, thanks

vzn

03:40

in theory salon, 5 mins ago, by vzn

PM Justin Trudeau gives reporter quick lesson on quantum computing during visit to Waterloo

elsewhere, nice article on google qm computing/ martinis lab in this mos technology review but alas, not free online (yet?)... not even listed either (???)

user116211

04:10

0

Q: Why entropy is additive?

If we have two identical isolated macroscopic systems both with energy $E$. The number of accessible states of each of them is $\Omega(E)$ and hence the entropy is $\ln\Omega(E)$. Now if we put them in thermal contact to form a larger isolated system with energy 2$E$ (suppose there are weakly int...

user116211

OP says the total entropy is $$\ln\sum_x\Omega(x)\Omega(2E-x) $$

user116211

But how could it be? Why is he considering the microstates of every accessible energy states?

user116211

The entropy should be $$S^*(x)= \ln\Omega(x)+ \ln\Omega(2E-x)$$

DanielSank

04:27

@vzn Which article are you looking for?

Bernard Meurer

05:07

heya

John Rennie

05:20

@Awm0121 the conduction electrons in a metal behave remarkably like a gas of free electrons. The speed at which electrical signals flow is effectively the speed of sound in this gas.

The free electron gas is highly incompressible so the this speed is very, very high. However it's significantly slower than the speed of light.

Secret

06:41

@ACuriousMind request noted. Sorry for the inconvenience

Slereah

07:50

welporino

I guess I'd better go back to Peskin for now :p

user116211

Damn, I missed the word 'total' ;(

Slereah

Also I think I know why I don't approve of the use of the propagator in the axioms of AQFT

How are you supposed to find the propagator for an interacting theory

Secret

08:08

Infinity always strikes me with this property
As if there are a unlimited number of things, but in some sense bounded by a finite sized "object"

I always think that the reason why infinity has the strange properties it has is all because of this 'tail'

I am not sure however how to mathematically describe this 'tail'

Slereah

Don't get too bogged down in representations

At its core it's just formal systems

I'm sure you have axiom systems that forbid such infinite shenanigans

Slereah

08:23

0

Q: I'M CONFUSING ABOUT SPACETIME

As what i studying about linear motion at high school, there was a graph which presents a velocity by a Time axis and a Distance Axis. I wonder Does spacetime is really just a graph that have Space axises and a Time axis? It is really funny if it is true because everybody on internet using that ...

quantum-mechanics classical-mechanics gravity mathematical-physics spacetime

Secret

While I am more tended to use representations, I do understand that in the core mathematical objects are just things that obeys some rules

For example here, we simply define the inaccessible cardinal as something that replacement axiom cannot produce

That's the abstract approach to mathematics, people don't see me use it often, but I do use it when drawing representations become counterproductive to the understanding

I am guessing we could have just axiomise a new number such that any conceivable iterrative process cannot reach it (not even via something like "define a new number such that it cannot be reached from below by any previously defined numbers"

But then the question is, is this really needed...?

https://en.wikipedia.org/wiki/Supertask

Do you find it weird that whever we generalise something in maths (e.g. from naturals to rationals, from rationals to reals, from reals to complex, OR e.g. from scalars to vector spaces, from vector spaces to manifolds etc.) it is always in some form of super or hypertask?

(i.e. some form of iterative process)

...I wonder if there are exceptions to this...?

ACuriousMind

09:29

@Secret That's just not true. The definition of a supertask there is a countably infinite number of process happening within finite time. There are neither processes nor time in mathematical definitions.

And there is no iterative process in defining the integers, the rationals and the reals from the naturals. Adding negative inverses is not iterative. Adding multiplicative inverses is not iterative. Making Dedekind cuts is not iterative.

Slereah

Really defining the integers and the rationals is barely any effort in set theory

You just need to define an ordered pair

Rationals are just $\langle a, b\rangle$ such that $a\in \Bbb N,\ b \in \Bbb N^*$

And then you define some identification if $\exists n. n \times a = b$

ACuriousMind

@Slereah Modulo the equivalence relation $(a,b)\sim(c,d):\iff ad-bc = 0$.

Slereah

yeah that

ACuriousMind

And that's not specifically "in set theory", it's the standard definition of localization/quotient field in abstract algebra.

Slereah

I forget the definition of integers from $N$

I think it's just an ordered pair too

Well I guess there's several methods to define them

Yeah it's just an ordered pair with the equivalence $(a,b) \approx (c,d) \rightarrow a+d = b+c$

Secret

10:00

great, hopefully by June when I have enough time to knock through some of my reading list, I will have a better knowledge on set theory and stop making mistakes like these

Slereah

I think maybe set theory is not the priority for you

Secret

what subject in the reading list you suggest should i dive through first (For the sake of discussion, just assume any subject in physics and maths is in my reading list)?

Slereah

Calculus and linear algebra :p

ACuriousMind

^

NeuroFuzzy

^.^

Secret

10:10

Linear algebra i am ok with it, analysis on the other hand, I am still struggling with convergences and series

ACuriousMind

@Secret Honestly, if you think that going from scalars to vector spaces is an iterative process as you wrote above, you are not ok with linear algebra.

Slereah

How are you on basic calculus

Derivatives and integrals and limits

ACuriousMind

Haha...I think Valter Moretti was a bit bored this Saturday, never seen him write so much on a rather basic question

Secret

@Slereah
(I have mentioned about my background before)

Derivatives I knew all the basic rules, 1st, 2nd derivative tests, extrema (and other things that I don't remeber on top of my head, but I should have know because I done it on my undergrad

Integrals I can solve the basic ones, trigonometric ones, some common improper integrals such as gaussian integrals, various ways to simplify integrals (partial fractions etc.),

Limits I can solve some trigonometric, rational functions, and some with pwoers of x, but in general I am bad at limits

NeuroFuzzy

@Secret So... a physicist's knowledge of calculus is totally differen tfrom a Mathematician's knowledge of calculus.

I spent a ton of time on Rudin's principles of mathematical analysis

and never once did it come in handy deeply.

Like, just as buzz words to impress people.

So unless you want to be a math grad student... do physics! :D

Slereah

10:23

A physicist's knowledge of calculus is like

NeuroFuzzy

oh is it?

Slereah

$\frac{df}{dx} = \frac{f(x + dx) - f(x)}{dx}$

NeuroFuzzy

it's like?

Slereah

DX IS VERY SMALL

Did I ever tell you my favorite physicist trick for calculus

NeuroFuzzy

but then it's like $\int \mathcal{D}[x] E^{-S/\hbar}$, but don't worry about it

Secret

10:25

We learn $\frac{df}{dx}=\lim_{h\rightarrow 0}\frac{f(x+h)-f(x)}{h}$ back in our undergrad

Slereah

$\frac{df}{dx} = a f \rightarrow \frac{df}{f} = a dx \rightarrow \int \frac{df}{f} = a\int dx = ln(f) = a x $

So $f = e^{ax}$

NeuroFuzzy

or $(1-\nabla^2)G= f \iff G=\frac{1}{1-\nabla^2} f $ :D

Slereah

:D

NeuroFuzzy

:DDD

Slereah

I've also seen square roots of differential operators

Secret

10:30

I am guessing G is a green function to be able to do that?

Slereah

Hamburger moment problem

In mathematics, the Hamburger moment problem, named after Hans Ludwig Hamburger, is formulated as follows: given a sequence { mn : n = 1, 2, 3, ... }, does there exist a positive Borel measure μ on the real line such that In other words, an affirmative answer to the problem means that { mn : n = 0, 1, 2, ... } is the sequence of moments of some positive Borel measure μ. The Stieltjes moment problem, Vorobyev moment problem, and the Hausdorff moment problem are similar but replace the real line by (Stieltjes and Vorobyev; but Vorobyev formulates the problem in the terms of matrix theory), or a...

2

So delicious

NeuroFuzzy

@Secret I didn't mean it as such, no.

Slereah

Obviously not a green function since it is f and not a delta :p

Slereah

11:17

Is the argument that perturbation theory doesn't work because 1) it's basically a polynomial expansion in $\lambda$, so it assumes that the propagator is analytic 2) the theory isn't defined for $\lambda < 0$ 3) same argument as compact support functions not being analytic

John Duffield

13:11

Well then, does anybody want to talk about anything?

skill patrol

13:25

sure

Faheem Mitha

I learned earlier today that David MacKay just died. I'm not sure if one would call him a physicist - he did a bunch of things.

John Duffield

What would you like to talk about? I may have to open another chatroom.

Danu

David J. C. MacKay

Sir David John Cameron MacKay, FRS, FInstP, FICE (22 April 1967 – 14 April 2016) was the Regius Professor of Engineering in the Department of Engineering at the University of Cambridge and from 2009 to 2014 was Chief Scientific Adviser to the UK Department of Energy and Climate Change (DECC). Before being appointed to the DECC, MacKay was most well known as author of the book Sustainable Energy – Without the Hot Air. == Life and career == MacKay was born the fifth child of Donald MacCrimmon MacKay and Valerie MacKay. His elder brother Robert S. MacKay FRS (born in 1956) is Professor of Mathematics...

This guy?

Faheem Mitha

@Danu Yes, him.

I thought he was quite famous. Maybe not.

I read one of his books a while back.

I was quite shocked to learn of his death.

skill patrol

how was the book?

Faheem Mitha

13:32

This one:

inference.phy.cam.ac.uk/itprnn/book.html

@skillpatrol Pretty good. Also quite famous, or so I thought.

The book, I mean.

John Duffield

Oh no. I've exchanged emails with David MacKay. He wrote Sustainable Energy – without the hot air, which is well worth reading.

Faheem Mitha

His father was a physicist too. Question for discussion: how much of a career advantage does that give you?

@JohnDuffield You didn't know, then?

skill patrol

wow only 48

FenderLesPaul

@Danu officially committed to Berkeley! :)

Danu

@FaheemMitha Probably a lot, if you discover you're into science early on.

@FenderLesPaul Congrats!

Faheem Mitha

13:35

I happened to learn about it while looking at Debian Planet earlier this morning. Matthew Garrett of Debian had a post.

John Duffield

@FaheemMitha : no, I've been a bit tied up this week.

FenderLesPaul

@Danu Bousso gave me a project idea to think about

so yay

Faheem Mitha

@skillpatrol Yes, bloody awful, isn't it? Stomach cancer. Hell of a way to go.

FenderLesPaul

head starts are always nice

skill patrol

@FaheemMitha :(

Faheem Mitha

13:36

@JohnDuffield Ok. It could have been months before I learned of it. I'm not exactly plugged into the scientific community (any longer).

He sounds like one of those impossibly cool people one reads about.

Occasionally I run into them online too. Rarely in real life, though.

skill patrol

online doesn't really count :P

imo

Faheem Mitha

@JohnDuffield I've not read his Energy book, though I've heard good things about it.

@skillpatrol heh. I think it does. Real life is better, of course.

I gather he was advising the UK govt about energy policy for a while, though I don't know any details.

Sorry, I didn't mean to butt in here. I've been here before, but rarely.

skill patrol

he did some bbc stuff too, right?

John Duffield

@FaheemMitha : I read it some years ago, it made real sense. How sad that he's died before his time, and so tragically. There but for the grace of God go I. Not that I'm religious mind.

Faheem Mitha

@JohnDuffield Yes, I'm sure it would repay a careful read. Though the Energy news rarely seems good.

And yes, it really makes one think. Talk about the good dying young.

Are you guys all physicists here?

John Duffield

13:44

@FaheemMitha : I fear we're heading for power cuts in the UK. Ferrybridge power station closed down a week ago. Our nuclear engineers all retired, we haven't put any effort into Thorium, and as I speak Drax is burning USA hardwood forests. IMHO that's Kafkaesque.

I'm an IT guy with an interest in physics.

Faheem Mitha

@JohnDuffield Nasty.

If anyone is interested, here is Garrett's post on MacKay.

@JohnDuffield Ok. I guess it's hard to make a living at physics. I mean, professional physicists do exist, but it's a tough job.

skill patrol

@FaheemMitha thanks for sharing

Danu

@FenderLesPaul Topic?

Faheem Mitha

I've heard of Dasher, but I've never used it, and don't really know anything about it.

John Duffield

@FaheemMitha : unfortunately two of my nephews with physics degrees aren't making a living out of physics. I have friends in the same boat.

Faheem Mitha

13:46

@JohnDuffield Are they doing computer stuff?

Danu

@FaheemMitha Don't worry about it. The chat can accommodate newcomers

Faheem Mitha

@Danu Ok. Thanks.

John Duffield

@FaheemMitha : one is, the other one isn't.

Faheem Mitha

@JohnDuffield Oh.

@JohnDuffield You're English, then?

John Duffield

@FaheemMitha : one sees continual talk about a STEM shortage in the UK, but people get a degree etc but then can't get a job in the field. Yes, I'm English. I live in Poole.

Faheem Mitha

13:49

Hmm, Dasher is in Debian, but the stable and unstable versions are the same. Which suggests either the maintainer is asleep, or development is less than active.

@JohnDuffield Yes, I think all that stuff about STEM shortages is crap.

John Duffield

@FaheemMitha : see things like this physicsworld article: blog.physicsworld.com/2014/10/17/…

Faheem Mitha

I think people say that because they want to drive prices down. Similar to the logic of bringing more foreigners in.

People meaning those running big companies.

@JohnDuffield Thanks.

Well, dinner time. Nice chatting with you folks.

skill patrol

cya later

John Duffield

Ditto, I have to go out now. Bye.

skill patrol

later

FenderLesPaul

15:26

@Danu It's on the relationship between BMS transformations and entanglement entropy

Danu

15:47

BMS seems like a hot topic (100+ citations already!)

Faheem Mitha

16:03

@JohnDuffield I think it's possible that all those tech people are needed. The problem is matching them to suitable work. I don't think the current systems do a good job of that. Like, at all.

And here's a complete version of that article.

live.iop-pp01.agh.sleek.net/2014/09/25/…

I'm using the word tech instead of STEM.

I'm not particularly fond of that acronym.

And it's not just about pay, though that's part of it.

vzn

@DanielSank theres a nice new profile in MIT technology review, just saw it at newsstand/bookstore, but its apparently not even listed online anywhere. (yet? maybe/ hopefully it will be available online later.)

vzn

16:24

@FaheemMitha STEM is a very broad field in general and one would not a priori expect all the subdisciplines to have highly simiar employment statistics profiles. for example CS (my field) vs physics seems substantially different. agreed there is probably a lot of misreporting in the area, somewhat even by entities dedicated to supposedly impartially tracking it. nowadays its a highly politicized topic.

Faheem Mitha

@vzn My point was that in some cases people looking for jobs could possibly do good work in the right job, but they don't get it.

I think this is a problem with any job, but the more specialist the work, the worse the problem.

vzn

@FaheemMitha right. JDs ref supports that. my sentiment is concurring.

Faheem Mitha

I'm just spitballing here, of course.

vzn

(am looking up another ref that popped up recently...)

Faheem Mitha

@vzn JDs ref?

Ah, John Duffield.

@vzn The reference doesn't exactly say that.

This is partly based on my own observations.

vzn

16:29

@FaheemMitha its says roughly that.

Faheem Mitha

As far as I know, no sensible mechanisms exist to match people to jobs. (I'm not saying it's an easy problem.)

vzn

yes everyone has anecdotal evidence.

Faheem Mitha

At least in academia, it's was not uncommon for me to meet people who filled me with wonderment, and not in a good way.

I'd be wondering - how on earth did this person ever get that job.

Yet, supposedly, the academic field is super-competitive.

It doesn't add up.

So, I think partly the problem of finding the right people for a job is that it's not unusual to have the wrong people in that job.

Secret

16:55

https://www.youtube.com/watch?v=9rIy0xY99a0
The question on what is true randomness is discussed earlier with Acuriousmind, David Z and Barrycarter with various conclusions (Serious one is that there may be no such thing, not so serious (but technically can be made serious) is the intellectual level of the h bar chat room at any given time)

To me, even quantum is not random enough because their probability amplitudes are determinstic, thus still provide us some means of controlling the outcomes

dmckee

@FaheemMitha Not only are academic hiring decisions made by other humans (which is, of course, the usual case) but they are made by other academics.

The existence of stiff competition does not imply that the judging criteria are either consistent or uniformly well thought through.

Faheem Mitha

@Secret The usual way of generating randomness is to create tests that we think "random" numbers should pass.

Then devise algorithms that produce numbers that pass those tests.

dmckee

On top of which the time-scale for replacing mistakes is too long at first in most place, and then ends abruptily.

Faheem Mitha

One common way in which some people cover up their ineptitude is by getting other people to do their work for them.

In theory this would seem a difficult thing to accomplish. But I've seen a lot of it happen.

Granted, I don't think I was in the best places.

dmckee

And finally, the job has a lot of challenges and pressures (again, the usual case), but if someone turns strange under the pressure after tenure they stay as long as they can meet the basic requirements, odd or not.

Secret

17:01

Actually, this give another interesting question: Given that in quanutm mechanics all we can ever physically interact with are when we measure the system we get observables that follow some kind of probability. How do we know that the underlying rules the govern the probability amplitudes (which in copenhagen interpretation, is just a mathematical construct rather than a physical thing) must be deterministic?

Faheem Mitha

@dmckee Yes, tenure does create a large class of deadwood.

It's supposedly designed so that researchers can get on with their work without worrying about job security. And I'm sure in some minority of cases that's what people do.

But there are many people who don't.

dmckee

@Secret Oh, that one is easy. You rig a bunch of case in which the deterministic rules give overwhelmingly strong probabilities of a fixed outcome and check the results.

Interferometric experiments are the usual way to do that, and QM passes with flying colors.

Secret

make sense.

Faheem Mitha

@dmckee Wow, you're an actual physicist. Color me impressed.

Oh, and a mod too.

Secret

Sometimes I felt I actually have a nonrealist mindset like most classical people

Because the probability amplitudes are deterministic, that is, pretty much the only thing that agrees with the unrealist's view that it exists independent of measurment, I do have a tendency to treat state vectors as physical entities, although I am perfectly fine in accepting that the state is undetermined until a measurement is made (the realist view)

dmckee works in particle physics experiments, thus he has a rich knowledge in particle phenomenology

Faheem Mitha

17:07

@Secret That's a bit metaphysical.

Like I said, in statistics, we consider numbers to be random if they pass tests for randomness.

But maybe you're having a different conversation.

Secret

O that above is just part of the spinoff discussion from quantum mechanics from the randomness discussion, sorry for that tangent.

Back to your discussion, what is the best test for randomness?

dmckee

@FaheemMitha Don't be too impressed, I work at a university so minor that we don't have a research expectation as such, and the project I do with students have budgets measure in hundreds of dollars if I'm lucky.

Faheem Mitha

@Secret Best test of randomness? No idea. I think there are a bunch of them, but details are way out of my area of expertise.

vzn

@FaheemMitha heres a nice graph that showed up recently cited on Fortnows blog. it shows wide variation in different STEM fields such as CS vs engr vs physics etc. cccblog.org/2016/03/31/where-the-jobs-are-2016-edition (esp graph 3)

Faheem Mitha

For the record, I have a math background and a PhD in statistics.

vzn

17:10

so while policymakers would like to group STEM fields for some (good) reasons, employment-wise, it might be somewhat incoherent at times.

Faheem Mitha

I'm sorry, I didn't really introduce myself. As the user name says, I'm Faheem. Mostly I hang out in U&L chat. Sometimes I wander over to other places.

@dmckee Do you get to do research? And how do you like your job? (Don't feel obliged to reply.)

vzn

another common theme somewhat hidden in these "statistics" is that companies constantly say "we cant find qualified [x]" but what they really mean can be highly specialized, or that they want a highly qualified [x] at low-qualified pay [y] etc...

dmckee

I have a couple of toy scale projects going on with students; those are about getting them the research experience and can't be anywhere near the cutting edge, but I hope to turn at least one in to a minor paper.

Faheem Mitha

@vzn Agreed, it's not a homogenous situation.

dmckee

I also do a hours a week on a project for an old boss that keeps me in touch with the "real" research community.

Faheem Mitha

17:14

@dmckee Would it be fair to say that getting a "good" research job in physics is brutally hard these days?

dmckee

And despite the stress and the lack of time for anything else I enjoy the teaching, so yeah, I'm mostly happy.

Faheem Mitha

@vzn Yes, pay is part of it. But the part I notice more is that people don't get to do the kind of work that they would like to do.

dmckee

@FaheemMitha I know some very good physicists who tried for one and didn't get it. Of my whole grad-school study circle I'm the one who got closest to "making it".

Faheem Mitha

When I was in academia, I was regularly surrounded by frustrated people.

@dmckee Hmm. Would you care to hazard a guess why? Simply not enough jobs?

Not enough science funding?

Too many people staying in their jobs after they can no longer do it?

dmckee

@FaheemMitha Simple, really. We graduate about ten times as many Physics PhDs per year as their are full-time university jobs for, and more than a third of them want those jobs. Plain 'ole supply and demand.

Faheem Mitha

17:17

@dmckee So, not enough jobs? Of course, a separate question is whether the best people are getting those jobs.

vzn

@FaheemMitha so are you in industry now? IT? is your Phd thesis online?

Faheem Mitha

@vzn I'm actually basically jobless. (Long story.) And yes, as a matter of fact it is, but it's a piece of crap. Want a link?

It's about perfect sampling, if that means anything to you.

Secret

I am Secret, I study honours chemistry in my uni and also was an undergrad in physics and chemistry majors

I have a strong interest in all 4 pillars of science (physics, chemistry, maths, biology)

I am best known in the chat and in real life for the following things:
1. A messy mind that tend to came up of weird ideas and questions every now and then
2. Continue from 1, most people said (even myself) have no idea what I am talking about most of the time, and I am bad at non face to face communications

vzn

@FaheemMitha nice ref thx... getting urge to blog on this

Faheem Mitha

I did manage to get a research paper out of it, but it's behind a paywall.

vzn

17:19

@FaheemMitha lol there is some self-disparagement of academics around here, have seen that before in chat room, like to browse theses myself esp of participants, just a harmless diversion

Faheem Mitha

One thing they don't tell you is how frigging long it takes to write a research paper...

@vzn Ok. So, do you want a link?

vzn

@FaheemMitha yes

Faheem Mitha

@vzn Amazingly still online -> stat.unc.edu/students/faheem/thesis.html

vzn

@dmckee or rather more exactly/ precisely, an imbalance in supply and demand

Faheem Mitha

It's the top hit for "Faheem Mitha thesis", as it happens. Props to Google.

@Secret Hi, Secret.

In the unlikely event that anyone here is planning to actually look at my thesis, I should warn you it has a ton of mistakes.

As became clear when I tried to turn part of it into a research paper.

vzn

17:23

@FaheemMitha general "perfect sampling" theory shows up in QM esp wrt bells thm, "efficiency loophole" etc... have delved into that stuff over long time...

Faheem Mitha

And it was a idiotic topic to choose for a thesis anyway, but my adviser was an idiot. Probably still is.

dmckee

@FaheemMitha When I got the bound copies of mine back I flipped it open to an random page just to admire it and the first sentence I read had a typo in it. Arrrgggghhhh!

vzn

science is a human endeavor. "to err is human."

Faheem Mitha

The field is basically a dead end. A waste of time. But I didn't know that at the time.

@dmckee lol

Actually, statistics is a pretty good area to work in. Lots of interesting problems. But realistically, you need to be part of a good research team.

Otherwise you're going to die.

To be clear by "dead end", I am talking about perfect sampling.

vzn

statistics is awesome but also struggling to keep up with the ML revolution.

Faheem Mitha

17:25

@vzn Machine Learning?

vzn

yes. deep learning.

dmckee

@FaheemMitha That's the key in particle physics too. No one person can get anything done, so your progress is inextricably linked to the skills and efforts of your collaborators.

Faheem Mitha

@dmckee Yes. Trying to write research papers alone (I've done it) feels like a slow and painful death march.

As opposed to those fun and joyful death marches.

For some reason, people rarely bother to separate out statistics from all the other stem stuff. I guess it comes under "mathematical sciences".

Though much of the time, it really isn't very mathematical.

Incidentally, the MacKay book that I mentioned earlier has a good basic exposition of Bayesian statistics.

MacKay liked Bayesian stuff, I think.

vzn

@FaheemMitha wow, (anti-?) cybersynchronicity, did you hear the news about MacKay?

Faheem Mitha

@vzn Are you asking me whether I heard about MacKay's death? If so, yes, I talked about it above.

That's actually the context in which I mentioned the book.

Actually, that's sort of the reason I wandered in here today.

Secret

17:39

Slereah

"CERN is a gateway that will allow the fallen angels spoken of in Genesis, Jude, and Revelation to finally be set free upon the Earth. I don't have the time or space to get too detailed about it here. Just google the info and read up on it yourself. "

I knew it

Secret

@Slereah what the hell is this clearly nonscientific sentence?

Slereah

"To put it simply: the God particle is the power that keeps the spiritual darkness from encroaching upon our world. It's what lies behind the veil between the natural and supernatural world: and the normal and paranormal world.

And these scientists plan on removing enough of this veil to collect and harness the dark matter that lies behind it. I'll say that again slowly: they...want...to...collect...DARK...matter. "

It's all true

Secret

@Secret I will need a better understanding of quantum mechanics and information theory to assess his claims. But if my memory serves, information must be conserved because of unitary (ignoring the black hole messy territory for a second...) thus his claim will be the same as saying that information is produced every minute. I also vaguely remember entropy is not information, but the measure of a lack of information

user54412

18:00

@FenderLesPaul Woohoo!

user54412

Did visiting places in person help?

user54412

@FaheemMitha In my anecdotal experience, having an academic parent is highly correlated with getting into a good college and being ready to excel there. All those years being taught things outside of school must help. I don't think career is correlated to upbringing beyond the connection through college, though.

Faheem Mitha

18:20

@ChrisWhite Not sure what the last sentence meant.

user54412

@FaheemMitha I mean undergrad is probably correlated with future career success, but the correlation probably has the same strength for undergrads with academic parents and undergrads without.

Faheem Mitha

@ChrisWhite Yes, but the probability of getting to undergrad and doing well there probably has much to do with upbringing. Is one of your parents a physicist, then?

Hmm, actually that first sentence doesn't make much sense.

user54412

@FaheemMitha Oh very much not. But disproportionally many of my friends from undergrad had physicists/mathematicians/engineers as parents.

Faheem Mitha

@ChrisWhite I see. I see you are doing astrophysics. How feasible is it to continue doing it? Given the job market, I mean.

user54412

@FaheemMitha I maintain that pursuing an academic career is a bit irrational, in part because the job market is pretty bleak.

user54412

18:34

I'm extremely lucky to have a job for the next few years, but we'll see what happens after that.

Faheem Mitha

@ChrisWhite Yes, that's what I figured. I think physics has been difficult for a long time. Maybe always.

dmckee

19:01

Something I didn't understand until I got a close look at a bunch of student from backgrounds without college educated role models is that there are a huge number of soft-skills to being a college student.

I get advisees who casually mention a solvable problem with the school bureaucracy that they've left unaddressed for weeks because they didn't know it was solvable.

If I learn soon enough a few words of advice about what office to go to and what to ask for gets them squared away.

My school actually has a required class "University Experience" where they try to clue new students into these things.

Which at least means that some of them understand that their advisor's job is to help them with those kinds of questions as well as giving them a heads-up if they have started down the road to trouble.

Alas, I still have advissees who start avoiding my hallway after they've had a poor mid-term grade report, when all I want to ask then is what their plan is for recovering some of the classes and triaging others.

user54412

@dmckee How closely are you expected to guide advisees?

user54412

We were supposed to meet with our advisers at least 3 times/year, if only to get classes approved. I looked forward to it, but some student-adviser pairs were content to blindly sign off on everything without even seeing each other face-to-face.

dmckee

19:17

I'm required to meet with them when they get assigned to me, for registration until they have junior standing, and if they need an advisor's signature on some form. I'm suppose to email them and ask for a meeting if they have poor midterm grades, poor final grades or get put on academic probation, but if they don't respond that is on them

In practice I see my successful advisees a bit more often than that, and avoiding me correlates pretty well with with struggling one way or another.

Doing a good job of advising a first-in-my-family-to-go-to-college student takes a fair amount of face-to-face time, but it pays off.

Faheem Mitha

19:56

@dmckee It's good you care. Lots of people in your position don't. Probably consider it a nuisance.

Bernard Meurer

@ChrisWhite How do you survive?

Transcript for