« first day (1733 days earlier)      last day (3191 days later) » 

12:43 AM
@ACuriousMind Metro is creepy as hell...very enjoyable!
Oh my god is that slenderman
 
1:03 AM
ooh metro
 
The game looks kinda...blurry, doesn't it?
 
@ACuriousMind Gimme your code, the relevant paper & $3/hour I'll fix it for you.
 
I'm running high textures @1200p
 
vzn
@Slereah funny stuff! reminds me, you might enjoy a book called "wheres my jetpack" by Daniel H Wilson
 
@0celo7 jesus wtf
 
1:12 AM
@FenderLesPaul I have AA turned off, but that's not the issue.
tbh the only difference between medium and high textures is that high textures dips below 60fps at times
@KyleKanos What does tessellation do for me?
 
@KyleKanos The search for freelance work isn't coming along well, eh? $3/hr sounds kinda cheap.
 
@ACuriousMind cf. the tessellation question
 
::checks wallet::
 
$3/hour lol what?
 
@0celo7 Tessellation is the tiling of a plane using one or more geometric shapes, called tiles, with no overlaps and no gaps
 
1:16 AM
dude even McDonalds employees get $15/hour in some places now
 
@ACuriousMind Well if you want to pay more....
 
@ACuriousMind ::spider crawls out::
 
But I've only started looking for freelance stuff the last 4 hours
 
@KyleKanos what about anisotropic filtering?
 
@0celo7 In the context of graphics, it's a DirectX technique that let's stuff look less "blocky" - it raises the amounts of polygons in a shape considerably
 
1:17 AM
@0celo7 That is a method of enhancing the image quality of textures on surfaces of computer graphics that are at oblique viewing angles with respect to the camera where the projection of the texture appears
 
Ok, going to ultra lowers me down to 30, and no difference
@KyleKanos that sounds like Wiki
 
Because it is wiki...
 
well I looked at that too
 
Google -> copy -> paste
 
I'm not an idiot
 
1:18 AM
But you are lazy
 
ELI5 was implied
 
5 hours ago, by 0celo7
I am dumb
 
@KyleKanos I looked at Wiki dammit
@ACuriousMind well played
 
You keep thinking I'm some CS guru & I just know how to program enough to get things done
 
@ACuriousMind From your little Metro experience, did you think it looked blurry?
 
1:20 AM
Ooo-ooo! Guess what!? Two people apparently have downloaded my dissertation (though apparently one was the University's library)
 
@0celo7 Not particularly, no
Although I can play it with all settings as high as they go :P
 
vzn
@KyleKanos did you put up a link for that in the chat room?
 
@KyleKanos So it's available now?
 
@ACuriousMind I can too, just not at 60fps
well without AA
 
@ACuriousMind Apparently for DL
 
1:22 AM
link
 
I think I got an email ~1 week ago saying 6-8 weeks for hard copy
 
will give a good skimming
 
vzn
KK have you heard of contracting/ agencies? think you might/ could fit into it.
 
Probably will fit in anywhere there's work ;)
 
vzn
@KyleKanos they are all over dice.com. ever tried it?
 
1:25 AM
Nope...never heard of it
 
vzn
@KyleKanos lol so get to it then! just search c++. isnt that your main area? there may be some fortran spots too etc.
 
On it now...
 
@ACuriousMind If you let @KyleKanos help you, you might end up with an SU(2) supernova machine
 
vzn
basically there are these corps called "recruiting agencies" aka (the not-so-endearing slang) "body shops". they are full of recruiters that only make $ by placing you. some of them will even occasionally finetune resumes. etc. the jobs will not always be the greatest or be long term (eg 6mos is a common length) but are "foot-in-the-door" type opportunities & can lead to more on occasion.
 
@0celo7 That sounds strangely awesome :D
 
1:33 AM
Kyle I'll download it
 
link?
paywall?
 
@vzn I am somewhat hesitant about taking 6 mos contracts due to having the family. I wonder if I chose a close-enough place, I could keep the wife & kids in NJ and live in a cruddy apt nearby
 
If only Maldacena was as good a writer as he is a physicist
 
If only Wald hadn't forced some idiot grad student to write the proof of Lemma 8.1.1...
 
vzn
@KyleKanos agreed!
 
1:40 AM
@0celo7 yeah I'm surprised
because Wald is one of the best physics writers I've come across
not only his book but his papers-they're incredibly well written
e.g. his gravitational self-force paper
so I'm convinced someone else wrote that proof
"proof"
 
vzn
@KyleKanos you dont want to relocate family? there should be lots of 6mo positions in your neighborhood. in major cities. it also depends on how long you want to "hold out" for a great position.
 
I'm okay with relocating family, but I don't want to do it in 6 mos intervals
 
vzn
right, so dont relocate then/ yet. are you near major city?
 
had another add D:
 
I think starting out straight up as a contractor will be very difficult: The problem with a large codebase is that it takes a long time to get a good understanding of it and especially if you're inexperienced, it's even more of a hurdle.
For example the guy who did NumPy/SciPy runs a somewhat large data sciency consultancy (Continuum Analytics). Might be of interest to KK? Don't know about their requirements in terms of industry experience though.
 
vzn
1:47 AM
contracting is common & large codebases are challenging but its "all part of the job". just working on some localized piece of a large app is common. its not about learning the whole app, only enough to do ones job.
 
It took me about 2 weeks to get my ~60k-line-long hydro code working correctly; another year or so to understand the inner-workings of it all
So I think it depends on the "what" it is I'd be doing
 
2 weeks?
 
If it's messing around with a few modules, I probably could do well with treating the rest of it as a black box
@0celo7 There were some issues with AMR that I (apparently) was being dumb with
 
@ACuriousMind lol he figured out his dissertation quicker than you your QCD code :P
 
@0celo7 :(
 
2:26 AM
@ACuriousMind Not even logged in to Steam...you're really serious about this project D:
@ACuriousMind If you're bored of doing QCD...please calculate the maximum energy that can be radiated away from the collision of two Kerr black holes. Assume their parameters are different and that they are in an initially axisymmetric configuration (so that there is no spin decay during the collision). Kthnx.
 
2:44 AM
@0celo7 I'm actually not working on it right now...but hell will freeze over ere I calculate black hole collisions :P
 
3:22 AM
@ACuriousMind I think it's possible to endow black holes with YM fields.
 
3:37 AM
@0celo7 indeed it is
higher dimensional black holes routinely carry gauge charges
on top of the usual charges we know and love
or hate idk I don't wanna make assumptions
 
 
3 hours later…
6:33 AM
There was a 70's paper that proved that black holes could not carry 4-fermion interaction charges, though :p
(that was back when people thought the weak interaction was a 4-fermion interaction)
I wonder if a black hole can get color charge
 
 
6 hours later…
12:09 PM
@FenderLesPaul the usual charge is already a u(1) gauge charge
So it's already an example
 
12:51 PM
0
Q: Why topological strings have to be closed or infinite?

Name YYYLet's assume spontaneously broken global $U(1)$ group. During phase transition global topological strings are formed. Why they have to be infinite or closed (there doesn't exist finite strings)?

Context!?!?!?!
 
Hello
 
So I am brushing up on my QFT
Is it impossible to have a state with a particle and antiparticle of the same momentum
b/c that would be like a(k) a^dagger(k) |0>
So basically back to 0?
 
$a$ is not the creator of antiparticles
 
Then what is
 
12:54 PM
Fields that have particles and antiparticles have two sets of creators/annihilators for them, like a complex scalar field that has $a,b$ and $a^\dagger,b^\dagger$, and the latter create particle and antiparticle, respectively
 
Hm
Got a reference for a complex scalar field?
Peskin Schroeder doesn't do it, except in an exercize
It is indeed two sets of operators for the Dirac field
I really need to reread some QFT
 
Well, there's not much going on, except that the mode expansion is $\phi = \int a(\vec p)\exp(\mathrm{i}\vec p\vec x) + b^\dagger(\vec p)\exp(-\mathrm{i}\vec p \vec x)$, and $a^\dagger \neq b^\dagger$ (which would be the case for the real scalar field).
 
Makes sense, I suppose
Since you are basically quantizing two fields
 
Exactly, it's equivalent to two real scalar fields
 
Is there a special notation to denote particles and antiparticles, by the way?
A bar over the number state, maybe?
 
1:01 PM
Oh...I don't think so, but I've never written many states where one would have to distinguish that.
 
Somewhat odd
 
1
A: What's the best way to avoid slipping on ice?

WBTYaktrax advertises products intended to help with this, which might be less damaging to interior surfaces than crampons. Keeping a low center of gravity can reduce probability of injury by reducing how far you fall. Positioning yourself so that if you do fall, a softer part of your body (like...

 
If you're interested in the scattering of particle states, you only draw Feynman diagrams, anyway ;)
 
^is this considered to be spam?
 
But what if I want bound states, HUH?
I do need to read up a lot on S-matrix stuff
Also renormalization
 
1:04 PM
@Slereah Ah, bound-state QFT is something most standard QFT brushes a bit under the rug
And I don't know much about it, actually
 
That it does
There's a few methods
Basically you look for delta functions in the correlation functions
 
@gonenc Not sure, but it's a crappy answer.
@Slereah Isn't that looking for resonances? (Don't know much about them either)
 
@ACuriousMind the question is also somwhat crappy
 
@gonenc Yup. I VTC'd as "engineering", since it fits the close reason description of "finding a solution to a specific problem".
 
@ACuriousMind have you played the epic war series?
 
1:08 PM
@gonenc How is that a reply to the linked message? :D
 
@ACuriousMind I am too lazy to type @acu
:D
 
And no, I haven't
 
Resonances are not delta functions
They are just kinda
Peaks?
 
Alrighty then
 
@ACuriousMind it's a flash game and the game is epic as the name suggests :D
 
1:10 PM
@gonenc There is a physics question there (increasing coefficient of friction), but it's wrapped around crud that it makes it a so-so question
 
Basically a bound state is a resonance with a width of 0
 
even so I've kinda stoped reading carroll :D
@KyleKanos I think so too but the physics answer would just be: let the friction constant go to $\infty$ :P :D
 
@Slereah That makes sense, since it should be stable, and there's an inverse relationship between lifetime <-> spectral width, I think.
 
@gonenc Well how does one do that ;)
 
I mean the friction constant of stuff that you wear
@KyleKanos Let the friction constant $f$ be much bigger than 1 then it follows that you don't slip any more :D
 
1:13 PM
Hmm. You're definitely a theorist :P
 
@KyleKanos ^related
^also this one :D
 
Quite clever
 
@KyleKanos indeed it is :D
 
I tried reading some fancy QFT and then I realized that I really need to catch up on some things :p
I keep trying to find an original paper idea but that shit is hard man
The importance of bibliography, I suppose
 
1:37 PM
(the trick is finding a problem easy enough to be doable but not done yet)
 
2:06 PM
@ACuriousMind Why do I hate Jack again?
 
@0celo7 In BL2? Because he attacked the train you were on, almost killed you, and is still trying to kill you?
 
I missed that cutscene because of cat, remember
 
(The train thing is the opening sequence you missed)
 
Why was I on the train
Why did he attack the train
 
BL2?
 
2:07 PM
Uh...you're a vault hunter, and he, too, wants what's in the vaults
I think you had some loot on that train, too, but could be wrong
 
yeah what's a vault hunter?
 
@Slereah Borderlands 2
@0celo7 People who open vaults (relics of some ancient civilization, I think), fight the security systems, and grab whatever looks as if it can be sold
 
o
 
Or work for Jack and give him everything they find
 
is Scooter on drugs?
 
2:12 PM
I think everyone in that game is on drugs :D
 
In games we just call them "power ups"
I wonder what mainstream game has the most transparent drugs as power ups thin
 
@ACuriousMind is there a way to custom mod guns?
 
Usually they are space drugs or fantasy drugs
 
@0celo7 Nope
 
GTA 5 literally has a bong
 
2:15 PM
Duke Nukem does take straight on steroids
 
and gasoline
 
Gasoline is not a very good drug
 
tell that to Trevor
 
2:29 PM
@ACuriousMind is using this golden key on the chest in Sanctuary worth it?
 
sigh spent 10 minutes wondering why my UserScripts weren't working when I realized I (somehow) turned Grease monkey off :(
 
@0celo7 iirc, you can't use it for anything else, anyway. The loot is scaled to your level, though, so you should perhaps wait till you're high level to use it
 
vzn
2:58 PM
@DanielSank personal blogs are difficult to gain readership but leaving comments/ links on other blogs esp highly trafficked ones is often an effective strategy to build up readership/ audience. links submitted to reddit, google+, facebook & other aggregators can generate significant traffic at times. however seems likely to me google would publish your writing in their corporate blog(s). iirc there has already been google blogging about the martinis lab acquisition, qm computing....
 
also : nude photos
show off those buns
 
vzn
(DS contd) Google quantum AI lab / google+
@Slereah very funny dude... (maybe jlo in front of a Dwave machine? now theres a meme!) btw what kind of coding do you work on anyway? re your thesis, bet/ surely it could be "pop sci-ized"... just imagine talking about it like michio kaku lol
 
Currently I work at a robot company
(They make robots)
I do a program to check on the mechanical data it sends to the cloud
So the mechanics people can look at it
 
vzn
@Slereah way cool what kind of robotics? have dabbled in robotics myself. recently visited a large supplier.
 
vzn
3:08 PM
@Slereah nice! have seen this. isnt that the std "NAO" used for robotics soccer?
 
Among other things, yes
 
vzn
there is a Robotics se... have chatted there also some.
so what kind of code do you work on (language)?
 
Currently mostly python
 
vzn
nice yeah dabbled some )( in python also. great for sci computing etc with all the libraries
so did you get the job even without talking about your thesis? :|
 
Is $35/hour too little for a programming consultant?
 
3:13 PM
I said the title :p
 
vzn
@KyleKanos seems decent/ acceptable as "entry level"... there are published rate surveys around...
 
QUANTIZATION AMBIGUITY AND THE FEYNMAN PATH INTEGRAL IN RIEMANNIAN SPACE
I do not know how to describe it to a layperson
 
Okay..TY. Found a Fortran consultant job from Dice (TY, BTW) and they asked what I was expecting.
 
Fortran, the undying languae
 
vzn
@KyleKanos :D on your way. yeah you could just take your last est hourly rate and bump it say $5/hr depending on what it was around (if it was halfway competitive... you deserve a raise!)... the pay rate can be very tricky at times...
 
3:16 PM
> I don't know what the programming language of the year 2000 will look like, but I know it will be called FORTRAN
@vzn Well considering my last job was as a Grad TA, bumping it by 5/hour would probably be a bad choice :D
 
One day earth will be swarmed with intelligent cockroaches and their terrible appendages will still code in FORTRAN
 
Though, of course, that statement was made before the 1990 Fortran standard was released where they eliminated the "yelling" version of the name for the more pleasant Fortran
 
@KyleKanos $35 for a PhD?
Jesus
 
At 40 hrs/week, that works out to like $60k/yr for an entry level position.
 
Some arithmetic shows that's lower than the average for my field with a BS...
 
3:18 PM
I will never not use FORTRAN
Or COBOL
Or LISP
 
vzn
@Slereah lol was just reading somewhere that cockroaches (eg as shown in chernobyl) are unusually resistant to nuclear radiation...
 
FORTRAN refers to the 1977 standard whilst Fortran refers to the 90 and later standards
 
Why were all the old school languages all caps
It's not like they were all anagrams
 
vzn
re pay rates, there does seem some disconnect between academia/ industry. not sure of all the nuances myself. KK are you saying grad TAs were unpaid in your case? paid in tuition? not following.
 
C is also in caps :P
C++ too
 
3:21 PM
@Slereah Because keyboards of the 1950's-1970's didn't really have lowercase
 
C++ isn't all caps, that would be C☩☩
 
@vzn Grad student salaries are typically low.
 
vzn
@KyleKanos you are probably not exactly "entry level" (a relatively amorphous industry term anyway) but solidly beyond it because of all your programming experience. you have several years of fortran, large project/ codebase, teamwork, scientific bkg, masters thesis (sheesh trying to remember, or was it phd?), etc, that is all significantly past "entry level".
 
I probably am beyond entry level. But I'm rather hoping that other opportunities arise before having to take a short-term consultant position.
 
vzn
@KyleKanos right. 6mo is not exactly "short term" though. it is near a minimum length thrown around. there are even shorter prjs that can pop up.
 
3:27 PM
Yeah, UpWork is loaded with less-than-one-week jobs
 
vzn
@KyleKanos not sure if anyone can really make much of a living at those very short term jobs/ "pieces". its a new frontier...
 
It would be far more feasible if one could telecommute in
 
vzn
@KyleKanos telecommuting is much more common these days... but still somewhat specialized...
 
Rather than having to constantly move around
 
vzn
KK are you saying you did major coding work paid on level as a grad TA? that sounds kinda low to me, esp if code is integrated into "ongoing" pkg other than your own research (but alas maybe not an uncommon story in academia). afaik grad TAs main responsibility is typically teach/ grade. did you teach/ grade? coding on top of that ought to bump up the value/ rate.
 
3:39 PM
Yes, most of my programming experience is through research at my MS/PhD programs. I was paid as a TA not an Research Asst, so I had to teach & grade introductory physics labs (mostly algebra based labs, but did some calc-based labs).
The relevant portion of my code is currently contained on my laptop & nowhere else (the two clusters it was on did HDD wipes for the June Top 500 benchmarking)
If someone requests it, I would probably pass it off. However, the algorithm itself is explained in a few places in the literature already, so it's not like it is some hidden nugget of gold.
 
vzn
ok. thought you might have made permanent/ integrated chgs on some large codebase utilized by other researchers (eg work on a library), thats more than mere personal research & closer to industry type work...
 
Well some of the tweaks I made to the code (the ones I actually sent to the main devs) were actually migrated to the main dev branch
But the core bit of my research was kept next to me
(not that the other devs were interested in what I was interested in, but I was fearing being scooped)
 
vzn
@KyleKanos ok! thats something to point out. esp if its significant amt of code. also any open source work is probably defn resume fodder. re "scooped" no danger of that after the research is over. if you integrate/ pass on the code its probably more "value"...
but also its quite common that devs work long on code that doesnt stay in "prod" or gets "shelved". happens to the best of em. battle scars.
there is a lot of coding talent showing up in this room and much of it is probably marketable. it helps to "spin it" in certain ways. youve already got good/ great instincts re communication/ ppl skills etc... think something will materialize.
 
4:21 PM
@KyleKanos Even by the mid sixty the high end ones did. It's just that you couldn't count on the user to have access to one of those.
Early unixes would even switch to case-insensitive all caps interactions if they detected an all caps login.
 
vzn
4:32 PM
@Slereah feynman path integral is from QED isnt it? it governs the way electrons/ photons interact?
"quantization ambiguity" makes it sound like maybe there is some question in "special situations" (even among experts) how to ("correctly") apply the calculation/ theory. somewhat like with black holes where there is much theory but some of it "competing".
 
@dmckee I think that Intel Fortran compiler still converts to all-caps during pre-processing. gfortran doesn't seem to do so though
 
@vzn Path integrals are much more general than QED.
 
vzn
Path integrals are a formalism to do QM with "classical" variables
Using the Lagrangian directly
 
They even occur outside physics in many places.
 
The quantization ambiguity stems from the fact that you use classical quantities
Since there are no commutation or anticommutation relations typical of QM
So px = xp
 
4:41 PM
You wrote your entire thesis on this ordering stuff?
Is it... that interesting??
 
But there are still equivalent within the formalism
It is!
Basically, in path integrals, the smooth paths are of measure 0
 
So what's so interesting about it? My profs all seem relatively uninterested in it (I did actually inquire about it a few times)
 
So all your integration is done on fractal paths
Which do not use the usual integral definition
Instead you have to use stochastic integrals
Which are only defined up to a parameter
That parameter is the equivalent of the operator ordering
 
Okay. That sounds like stuff that I've heard about too (stochastic integrals and stuff)
 
Basically, you may recall that in the Riemann integral, the integral is something like $\sum f(x_i) (x_{i+1} - x_i)$
 
4:44 PM
@0celo7 I did the calculation for energy radiated by colliding Kerr black holes if you still want to compare
 
$f(x_i)$ can be anywhere between $f(x_i)$ and $f(x_{i+1})$ is the curve is smooth, no biggie
But if you are integrating on a stochastic path, this choice becomes important
 
@FenderLesPaul I have not yet done it, but I'd appreciate the answer for when I take a shot at it
 
(the fact that the path is a fractal of Hausdorff dimension 2 is the local equivalent of the uncertainty principle, by the way)
 
@0celo7 alrighty lemme see where my notepad is at
 
The path is well defined, but as it is fractal, measuring the position at two different moments will never allow you to go beyond some uncertainty
 
4:48 PM
@Slereah Really? That sounds funny
 
It is quite neat
The path is nowhere differentiable in configuration space
And it is discontinuous in phase space
 
Anybody care to upvote the answer here?
3
Q: Reset username for user

Unicorns Are Very Very YummySo I found this user. His needs nuking. Can a mod do that?

otherwise we'll see it around every month courtesy of Community
 
That master thesis is when I had to do an integral four lines long :p
For 2 weeks
 
@EmilioPisanty done
 
that was strange, my router just disconnected from the internet
 
4:56 PM
that's what routers do
 
@Slereah I think @ACuriousMind has a bot set up that murders anyone who says "uncertainty you know what"
 
what
 
yeah, I'd be scared too
the Germans have shown proficiency at entering France with bad intentions
 
5:19 PM
Is there any way to modify Planck's law for a body that is not at thermal equilibrium?
 
@HDE226868 Seems so, but I haven't taken the time to read/digest the document: arxiv.org/abs/0710.5686
 
@KyleKanos Thanks.
 
@0celo7 I'll have to tell you what I got after work; I left my notepad in my apartment sorry about that dude
 
6:14 PM
Couple of interesting Physics Today links:
will make Bohmian mechanics people happy. And
is just generally cool.
 
@dmckee Indeed, some models predict half-lives up to a million years for new superheavy nuclei, and some even calculate them at around Earth’s age. - Wow.
 
Er ... the first one is behind a paywall. Sorry about that.
 
vzn
6:38 PM
@0celo7 what is "what"?
@dmckee :D ... alas, $30 paywall... :( ... but have found another great article by Bush few mos ago :)
 
I wonder if anyone actually buys articles from science publishers
Like what is the percentage of revenues of purchase versus subscriptions
 
obe
@KyleKanos Can you help me out with openCL?
 
Never used it, but I could try
 
obe
The main thing I need help with is being able to run it.
I downloaded the drivers for intel openCL, though I'm unable to run demos in vcc++.
 
6:56 PM
Any errors when compiling?
 
obe
There are.
 
@all did they redesign YouTube lately?
 
obe
@KyleKanos Though should I be using the solution or project?
 
@0celo7 Does not appear so to me
@obe The what?
 

« first day (1733 days earlier)      last day (3191 days later) »