The h Bar

Semiclassical

20:00

eh. comparing undergrad to phd is a bit problematic. (less so in the first year or so, but once you get into research it's really not a good comparison)

enumaris

relatively few people do PhD

although the number was rising for a while

I don't know the numbers currently

Anonymous

@NovaliumCompany Well, for one, while you are younger you can pick up new skills faster. That decays as you grow older. 15 is the good time to learn whatever you want to learn

Anonymous

By learning I don't only mean academic learning

Semiclassical

depends on what kinds of skills one is talking about

Novalium Company

What's the problem with learning 'as you grow older'?

Semiclassical

20:02

for academic learning, I think it's usually easier for me to learn stuff now because I have enough range that I can place it in context

Droleulb

i think he means it's easier to pick up completely new things

which probably has some truth because you're still developing and have lots of free time

Semiclassical

plus there's a certain level of confidence which phd level research instills.

Novalium Company

Hmm, ok, for now, I'll try to put as much as possible useful information into my brain. That's the plan :D

Droleulb

that's a bad plan

Novalium Company

?

CooperCape

20:04

Then again learn things properly and not dip in and out.

Droleulb

you should just try new things and see what you like best because you have that freedom right now.

Semiclassical

the following is meant entirely in jest: "don't trust him, he's backwards"

Droleulb

jamming as much stuff into your brain as possible as early as possible won't help you that much in the long run except for being better than all your peers for a few years until everyone normalizes in their research.

Novalium Company

@Droleulb What should I do then, what do you mean by 'trying new things'. I think I've found my passion and what I want to do.

Droleulb

consider that you might be wrong.

overspecialization is dangerous especially if done really early.

CooperCape

20:07

@Droleulb I'll have to agree there

I was convinced I wanted to do medicine until I was 16

Droleulb

try to be well rounded in a lot of things

so when the ACTUAL time to fully decide what you want to do for a living comes years down the road you'll have a much better idea and lots of more doors open for you

enumaris

then again, there's some things you really do need to be very young to start in order to become tip top

e.g. professional sports

if you don't start at a young age, it's unlikely you'll become the next Chess World Champion for example

Novalium Company

@enumaris I'd agree.

I mean, look at Einstein. He started Physics and Math from a very young age, he knew what he wanted to do, and did it.

Droleulb

@enumaris it's unlikely anyway

enumaris

vastly more unlikely if you start at an older age for those though

Droleulb

20:12

to think that "if i start really early i'll have a super duper edge over everyone else and definitely become the next big name" is wrong. you'll only incrementally have a greater chance of contributing something important to your field than everyone else in it, but you'll also feel much worse if you don't because you set yourself up for expecting it.

that's in the context of academics, not activities like sports or games.

Novalium Company

I'm sure what I want to do, and I prefer to start early, in order to be ready when the time comes. The chance of loosing the interest or failing is always there, but I'd rather take it. I'm the kind of person, that if I start many things at once, I won't finish any of them, but If I concentrate on one or two things, the chances of finishing them are higher. (Btw I have no idea what I'm talking about, It's really late and I'm sleepy, please excuse me.)

Droleulb

this changes a lot if you end up on the bad side of the average, like starting your degree in some area of science at 30 while doing something irrelevant prior to it.

you'll have lower chances of success obviously but it boils down to the individual in the end.

Novalium Company

@Droleulb Thanks for the advice but I'd rather follow my heart and take the chances. I learn by making mistakes. You don't learn to swim by reading a book on how to swim right. I will focus on the things that interest me (Physics and Engineering are not the only ones) No hard feelings.

Goodnight guys.

Anonymous

@NovaliumCompany By all means, start early! But don't do it in hope of becoming the next Einstein, as that is not a reasonable goal. Moreover, keep in mind that your likes and dislikes will change with time. :)

Anonymous

Goodnight

Cows

20:22

Taking a break from code to do laundry. Fresh and sweet smelling clothes tends to be something people judge others on, don't want people to poo poo my ideas because i don't have clean laundry.

Droleulb

@NovaliumCompany i don't know why you're saying "but i'd rather" because what you said after that has nothing to do with what i said and your last sentence agrees with what i said before too.

Cows

Today should be an epic day of exploration of some cool math and cs ideas. 🐒

Novalium Company

@Droleulb It's really late, my eyes are closing, I'm sorry If I am making mistakes.

Droleulb

starting early is not a high risk high reward thing, by all means start early. there's no "taking chances".

i'm just saying don't think it will help you be better than your friends who did not start early in their research later on.

(though it could)

Novalium Company

I'm not studying these stuff because 'I want to be better'. I have never said that I want to be a physicist or a professor. I just study 'technology' overall, becuase it's interesting to me. I have no problems with learning something, giving it up, start something new, give it up... in the end, that way I'll know from everything.

@Cows Do you program?

Cows

20:28

Alex K Chen

hullo blue

how're you

Droleulb

@NovaliumCompany i said that in general. feel free to learn as much as you want as soon as possible if you love doing it.

Novalium Company

Ok, that is what I wanted to say.

I hope this is not an argument, we are just discussing right?

Droleulb

i'm just kind of weary of doing that now because that's more or less what i did in high school and it screwed up my first year of undergrad but it was mostly my fault rather than the fault of me learning.

no it's not an argument lol

Novalium Company

Oks, because I don't want any bad feelings :D

CooperCape

20:32

And also. Learn it properly.

Otherwise you’ll carry through your misunderstandings to when you learn it in class

Novalium Company

Ok.

By the way, is it a problem if I give you a link of a website I made :D?

Anonymous

Sure, why not

Anonymous

What's it about?

Novalium Company

triobots.com It was a 1 month project I did. Nothing fancy, still waiting for customers :D

Droleulb

@CooperCape this is something i thought about too

but then i realized that it's pretty difficult to learn improperly.

if you read the right books and not wrong stuff like JD's book as your only source of physics.

you could learn badly though

in which case the classes will help you fill the gaps

Anonymous

20:36

@NovaliumCompany You made it yourself? That's looks quite nice :D (I personally have 0 knowledge about web development)

Novalium Company

All coded by me :D for around 1 month.

Anonymous

Awesome!

Novalium Company

I hope this is not considered an advertisment and hbar will not ban me or something?

Anonymous

Lol :P

Anonymous

That "Contact" button at the bottom is a bit strange

Anonymous

20:38

And the "Follow Us" looks a bit big

Novalium Company

For 14 years old, that's what you get :D

Anonymous

Maybe you could improve that. But apart from that it looks excellent

CooperCape

@Droleulb good point. But I still think it’s important to be thorough

Anonymous

You should consider teaching me some web designing at some point :)

Novalium Company

I would love to.

Droleulb

20:39

@NovaliumCompany same

i have code phobia

i just hate symbols

Anonymous

@NovaliumCompany What language did you use?

Droleulb

i hate writing too

just like pictures

CooperCape

I will happily admit to being worst in my class for moments and stuff because I don’t find it interesting and subsequently never learnt it properly

I see a ladder against a wall and literally cry

Ain’t about that

Novalium Company

@Blue Javascript and Php are the main languages I used for the website. (Html and Css are included, but they are not exactly programming languages.)

Anonymous

Nice. I have no idea about either of those apart from the fact that javascript is a notoriously hard language to master

Anonymous

20:41

HTML and CSS are markup languages, right?

CooperCape

@Blue you forgot the ;

Novalium Company

@Blue Yep. I would love to talk programming but I really have to go sleep now. I'll be back tommorow.

Anonymous

See you around. Bye!

Anonymous

@CooperCape I don't really know where semicolons are suitable...

CooperCape

@Blue oh me neither. I was just making a joke on having to end every js line with a ‘;’

Anonymous

20:45

Ah. That's a common feature in many programming languages

enumaris

not in python

muahahahaha

Anonymous

Java, C, C++, ...

Anonymous

@enumaris Yep

enumaris

or Fortran...

which covers both languages I've used extensively lol

Anonymous

FORTRAN is on its dead bed

enumaris

20:46

still used in numerical computing

for pure speed

CooperCape

Pretty sure only time I’ve seen Fortran as a word is when celo complains.

Anonymous

@enumaris Yeah. The condensed matter guys at my uni use it still now. Somehow managed to convince them to partly shift to C

enumaris

when I did fortran I didn't vectorize anything tho...so I'm actually not sure about the speed

dmckee

@enumaris Fortran compilers have supported vectorization on platforms that support it for ... more or less as long as any high level languages have supported that.

enumaris

yeah, I'm sure that since I used the linear algebra library to diagonalize my Hamiltonian that it was vectorized in some sense...

but vectorization wasn't really an issue that I was aware about when I was coding fortran lol

dmckee

20:52

But until the mid 1990s there simple wasn't much vectorization worth speaking about on consumer platforms.

enumaris

21:05

I assume that the rise of vectorization must have something to do with the rise of multi-threaded computing or some such? o.O

Anonymous

Multi-threading does save a lot of time, when you run big and complex programs

dmckee

It started with the incorporation of SIMD math instructions on the Power and Pentium lines.

Anonymous

Parallel computing

Parallel computing is a type of computation in which many calculations or the execution of processes are carried out concurrently. Large problems can often be divided into smaller ones, which can then be solved at the same time. There are several different forms of parallel computing: bit-level, instruction-level, data, and task parallelism. Parallelism has long been employed in high-performance computing, but it's gaining broader interest due to the physical constraints preventing frequency scaling. As power consumption (and consequently heat generation) by computers has become a concern in recent...

dmckee

Multi-core CPUs and therefore multiprocessing being common rather than special to multi-CPU motherboards came later.

Anonymous

Also, parallel computing ^

enumaris

21:08

no idea what SIMD is

Anonymous

SIMD

Single instruction, multiple data (SIMD) is a class of parallel computers in Flynn's taxonomy. It describes computers with multiple processing elements that perform the same operation on multiple data points simultaneously. Such machines exploit data level parallelism, but not concurrency: there are simultaneous (parallel) computations, but only a single process (instruction) at a given moment. SIMD is particularly applicable to common tasks such as adjusting the contrast in a digital image or adjusting the volume of digital audio. Most modern CPU designs include SIMD instructions to improve the...

enumaris

hmmmm

dmckee

You could buy 2 and 4 sockets motherboards in the 80s, but they were expensive and of limited help because the main OSes lacked sufficient support. Only uber-geeks bothered with them.

enumaris

so they used to have motherboards with like 2 CPUs rather than 1 CPU with a dual core?

I didn't know that lol

dmckee

@enumaris It's what made Crays special back when Crays were special.

enumaris

21:10

never heard of Crays...D:

was that what a Pentium D was?

I had a Pentium D back in the day...

dmckee

::begins sobbing into his prune juice::

enumaris

lol

hmmm so that gives rise to a question of how to delineate multiple cores vs multiple CPUs...I suppose it has to do with the architecture of how the memory is shared between the processing units or some such...

hmmmm

dmckee

@enumaris There is no fundamental difference.

enumaris

or probably architecture in how the cores communicate with each other?

oh really?

it's all just...convention?

dmckee

In principle there is more flexibility in separate sockets because the chip designer doesn't lock you into a particular cache architecture, but in practice it's useful to have the cache "close" so you can't take advantage of the theoretical improvement.

enumaris

21:15

hmmm

dmckee

@enumaris You have a bus of some kind connecting the separate computation units. In one case the bus is on a board, in the other it is in the chip, but they still do the same thing.

enumaris

I see

but where the bus is located could be the "delineation" I guess...integrated into the chip, or not integrated

I saw some article about vertically stacking GPU cores and how that architecture has some advantage over horizontal stacking...was interesting...but computer hardware is pretty outside my field lol

crap, my entity linking algorithm is horrible lol

Cows

@NovaliumCompany Yes I do

back home . . . . Doing some light reading while watching TV

enumaris

dam, it seems this problem I have is a totally non trivial problem...

will take a while to solv eit

Cows

@enumaris what are you trying to make?

enumaris

21:24

well I'm building a NLP pipeline

Cows

Oh nice

work or research?

enumaris

and at the end of the pipeline I want to aggregate the named entities that it outputs

work

but I'm running into a mild version of the entity linking problem...my problem is not as big as that one, but still non trivial

Cows

Are you allowed to share the algorithm you are using :P

enumaris

can't really share it in its entirety

nor can I share the data that goes into it

Cows

awwww

yeah I know the data can't be share for sure

enumaris

21:26

I mean, it's not research level stuff, I'm using open source pieces

so like there wouldn't be much value to me sharing my code anyways

it's kinda special tailored to my data

Cows

well I was just looking at a hint, like the name of the algorithm if it is a well known one?

may be

enumaris

uh...well the NLP pipeline itself is pretty standard

clean-tokenization-POS tagging-Parsing-Named Entity Recognition

almost all done through Spacy

Cows

I see

enumaris

but I gotta like filter stuff

and all that jazz

get named entity relationships

stuff like that...

Cows

I've not used Spacy before. Just looked it up. It is described as "Industrial-Strength
Natural Language Processing" .Sounds scary

enumaris

21:30

lol, it's pretty good

I like it better than NLTK because it's more streamlined

and also I think the support for NLTK is kinda lacking since it's primarily academics based

Cows

I use NLTK . That's my jam, but it's for babies though hehe

enumaris

NLTK doesn't even have deep learning tho :P

Cows

yeah

enumaris

spacy has deep convolutional networks in there

Cows

scrolling through the api now

it looks pretty nice

enumaris

21:33

yeah, pretty intuitive mostly

I'm not at the point yet where i want to retrain their models

we'll see how burdensome that is lol

Cows

The documentation looks pretty good too

enumaris

yeah, they are backed by a start up

so, they gotta make the presentation really nice

Cows

Just going through the video on their api page

wow

hehehe he says interacting with the database is poison to researchers hehehehehe

He is able to convey his ideas nicely and effectively

I like that, and I want to be able to do that

@enumaris ^^^^

enumaris

there's a video?

I didn't even notice lol

Cows

yeah hehe

enumaris

21:43

maybe cus I can't view videos at work lol

Cows

oh

lol

bolbteppa

Why is $- i P_0$ positive-definite?

Oh wait, $P_0 = \partial_t$ in this case

Slereah

I don't think there's a proof of this, that's just a general QFT assumption

Cows

really?

why can we assume this

Slereah

About the momentum operator having values in the future light cone

@Cows we expect particles to go from the past to the future, not the other way around

Cows

21:54

oh, yes a physical argument

bolbteppa

So doesn't this mean energy is always positive if you assume that

Cows

@bolbteppa are you trying to write some kind of a proof?

I feel like one has to first sketch a hyperbola

@bolbteppa do you have the full statement of the problem?

enumaris

cruuud

this linking is unusable -.-

Slereah

@bolbteppa Yes

you can also develop QFT for weirder fields, but usually it doesn't work

Cows

@Slereah I thought that assumption of positive definiteness comes because of the hyperbolic constraint(SR) , . . . . just a guess

anyways i am not sure if I am making sense

let me just leave this convo while I have some dignity left :P

@Slereah ^^ thoughts?

Slereah

22:09

it's part of the whole Wigner business

Cows

oh cool

enumaris

Wigner's fake time-axis rotation thingy?

at some point I really gotta actually learn QFT...-.-

GettingNifty

how fast is our galaxy moving?

enumaris

with respect to what

GettingNifty

in respect to the universe i guess

enumaris

22:16

like the CMB?

GettingNifty

is the universe really expanding or just reaching our eyes

enumaris

wut

I think the Earth's peculiar velocity w.r.t. the CMB is around 300km/s

but I don't know about the Milky Way's

Pulsar

I'm sick of high school level questions getting boosted by SE. The 'Hot Network Questions' are the SE equivalent of clickbait. /rant

enumaris

much of the 300 km/s is coming from the Sun's rotation around the Milky way and the Earth's rotation around the Sun tho.

GettingNifty

i think of the universe as the main body that holds our laws to be true

enumaris

22:21

arxiv.org/pdf/astro-ph/9803306.pdf this paper seems to suggest the local group moves w.r.t. CMB frame at around 668km/s ... hmmm...that's quite fast

GettingNifty

so if time dilation were true which i assume it is, we should actually feel longer and shorter days as we accelerate or decelerate and the galaxy should play a part in that acceleration, using the universe as the constant

Slereah

@enumaris that's Wick :p

Wigner is the little group of the Poincaré group thing

enumaris

o

yeah...

Wigner is the Wigner-Eckhart theorem thing?

Slereah

Subgroups of the Poincaré group that preserve momentum

enumaris

spherical tensors and what not?

GettingNifty

22:23

there has to be reference points or something we can use to see our position in the universe

Slereah

the Wigner classification

it is called

GettingNifty

like if we put rovers on mars, we can use the observations against ours on earth and pinpoint a location

because we know it's distance from us so then create the universe from our position and our theories and formulas of time

and calculate how fast time is moving somehow... maybe carbon dating or what not

if E = mc^2 and F = MA

The motion that's left must be the particular motion of our Galaxy through the universe! And how fast is the Milky Way Galaxy moving? The speed turns out to be an astounding 1.3 million miles per hour (2.1 million km/hr)!

ok and the speed of light is 669 miles per hour

million*

So we're 1/514th the speed needed to stop time or go back in time

So to live forever we need to hop the nearest star leaving the galaxy, going to the nearest fastest galaxy and so on, until our bodies can repair faster than the surroundings can destroy it

right

enumaris

22:46

wat

GettingNifty

phys.org/news/2011-07-big-quick-conversion-antimatter.html

could a pole shift cause matter to start changing form, or reversing?

enumaris

23:02

by a pole shift you mean like Earth's magnetic poles flipping?

GettingNifty

yeah

and correct me if I'm wrong but the average human life is 80 and 80 x 514 is 41,000

the last pole shift happened 41,000 years ago

i can't really connect to the two though

enumaris

probably not much connection between that...

and the average human life right now is 80, but it was much shorter 41000 years ago

lol

GettingNifty

yeah ok but look at this..

"Other than incomplete genealogies, there are other measures of the age of humanity found in the Bible. First, the Bible says that the Lord made a covenant and commanded his law to 1,000 generations:

Remember His covenant forever, The word which He commanded to a thousand generations, (1 Chronicles 16:15)
He has remembered His covenant forever, The word which He commanded to a thousand generations, (Psalm 105:8)
but showing love to a thousand generations of those who love me and keep my commandments. (Exodus 20:6)

enumaris

well a generation is only roughly 25 years

1000 generations is like 25000 years tho

so definitely a long way to go from since when the Bible was written

GettingNifty

krishnapath.org/…

enumaris

23:17

we are about 1/6 of the way through I guess...

seems a bit crackpotty tho tbh

GettingNifty

have you ever seen the explanation of physics in the hindu scriptures called vedas

it tells the speed of light thousand of years before einstein

and funny Krishna is the same God as Mithras and Jesus and was the first monotheistic religion

ACuriousMind

@GettingNifty I think you might be looking for one of the religious rooms, not this one...

GettingNifty

That's not what i'm saying.

enumaris

Christianity was certainly not he first monotheistic religion...

GettingNifty

Krishnah

enumaris

23:21

probably Zorastrianism is the first recorded monotheistic religion...?

not sure tho

GettingNifty

If all of these are ancient science and the ideas are expressed metaphorically so that people of all kind could understand them.

That's why they repeat themselves.

Maybe?

And that's why they're true and nobody can explain it. Because we're not the first people to find out.

enumaris

o.O

I dunno man...I tend to keep religion and science separate...

GettingNifty

Well of course but Einstein was Jewish.

enumaris

Jewish by heritage

He was not a religious Jew though

GettingNifty

Oh is that what he said. He's only the most famous scientist in the world.

enumaris

23:26

I think his writings and quotes and stuff sorta show that he wasn't a very religious person

He probably believe in a God, but not a personal one like that depicted by Christianity

GettingNifty

google.com/…

ACuriousMind

@enumaris He was likely some sort of pantheist, whether to call that "religious" or not is probably not a very useful debate :P

enumaris

hmmm

GettingNifty

And he helped make Nuclear weapons. I'm sure he said a lot of things, for a lot of reasons.

enumaris

Did he ever label himself as such? I feel like people just look at his words and make conclusions for themselves...

ACuriousMind

23:29

I mean, you can quote mine him for quotes that come across as rather anti-religious as well as for those that seem to imply a very charitable view of religion (e.g. "Science without religion is blind, religion without science is lame"), but that's also probably not a very useful exercise

enumaris

Most of the sources that I've read about Einstein say only that he probably believed in a God, but not a personal one.

GettingNifty

haaretz.com/jewish/…

Albert Einstein's religious views have been widely studied and often misunderstood. Einstein ... Einstein was raised by secular Jewish parents, and attended a local Catholic public elementary school in Munich.

ACuriousMind

@enumaris The most explicit statement you'll find is that he wrote once that he believed in "Spinoza's god", which is indeed not personal

enumaris

I see

I'd imagine that his views probably also changed throughout his life

like most people...

GettingNifty

Well I mean I'm Jewish and I was Atheist for four years but I was raised Christian.

So if you quoted me my whole life you'd get a lot of different answers..

enumaris

23:33

well most people who aren't like super devout adherents to one religion

which might not be "most people"

ACuriousMind

And then you also find him calling himself an agnostic but it is not clear to me whether he thought of this as compatible with Spinoza's pantheism or whether this represents a shift in his views. As I said, probably not a very useful exercise to try to get a coherent picture, because there may not be one

enumaris

hmmm

I feel like also agnostic is like...a multi-level thing

if you believe in Christianity but have doubts, are you agnostic? o.O

like how unsure do you have to be before you are an agnostic...lol

ACuriousMind

Yeah, that's a frequent debate about the term. In the weak sense, i.e. not having absolute certainty, most people are agnostics

enumaris

yeah

ACuriousMind

But usually people don't really deliver a definition when they say "I'm agnostic", so that makes the term kinda hard to nail down :P

enumaris

23:37

indeed...

GettingNifty

On 17 April 1955, Einstein experienced internal bleeding caused by the rupture of an abdominal aortic aneurysm, which had previously been reinforced surgically by Rudolph Nissen in 1948.[141] He took the draft of a speech he was preparing for a television appearance commemorating the State of Israel's seventh anniversary with him to the hospital, but he did not live long enough to complete it.[142]

Einstein refused surgery, saying: "I want to go when I want. It is tasteless to prolong life artificially. I have done my share, it is time to go. I will do it elegantly."[143] He died in Prince…

enumaris

like the entire bridge between hard atheists who know there is no God or gods, and hard theists...

GettingNifty

and his last words were in German so the nurse didn't know what he said

his FBI file was 1,427 pages long

ACuriousMind

I mean, depending on your epistemology, if you don't think you can ever be absolutely certain of anything, then you're agnostic about everything in that weak sense, so it's not really a religious viewpoint when you say you're agnostic.

enumaris

possible

term too broad

please remake

ACuriousMind

23:41

That's why we should stick to physics, where things are always well-defined.

enumaris

things are always well defined...hmm...not sure I believe that :P

is the path integral in QFT well defined? It seems like it wouldn't be a well defined integral...

ACuriousMind

I thought the link I provided would indicate I'm perhaps being slightly ironic :P

enumaris

I spent some time studying the math behind path integrals...but it still doesn't seem well defined

hold on as I try to get around my firewall lol

ACuriousMind

It's an xkcd, what kind of draconian firewall blocks that?

enumaris

yeah...this firewall is weird...

it blocks xkcd but not reddit

lol

GettingNifty

23:45

"We need God to prosper. Those without him will not." Nostradamas

ACuriousMind

@enumaris That's certainly quite a puzzling decision :P

enumaris

I think images might be a factor

a lot of places with images are blocked

imgur is blocked

I don't really know...

ACuriousMind

Oh no, someone doesn't want you to see stick figures? Whose firewall is this anyway, are you at work?

GettingNifty

businessinsider.com/…

enumaris

yeah I'm at work lol

I think it's like for productivity or something

don't want me spending my time reading xkcd

Transcript for