« first day (977 days earlier)      last day (3952 days later) » 

12:01 AM
Just because we don't give physicists the right to do whatever they want on the site doesn't mean that it isn't for physicists
The purpose is more of to create a repository of useful Q&As, but that works too
I'm not particularly concerned whether or not you think that this site is "there for physicists", so please stop repeating that.
 
I mean all this talking about the needs of physicists are not catered, people governing the site light heartedly take a loss of experts for "civility reasons", at math too, etc ...
 
@Dilaton Yes. Many things will not be tolerated.
So?
We don't cater to all the needs of physicists
Doesn't mean we don't cater to the needs of physicists
 
@ManishEarth I'd just like to obtain some serious and honest answers now.
 
I am being serious.
And honest
We don't like losing experts. However, rules are rules.
And for that matter, many, many people would not stay on this site if they saw a lack of civility
I have seen many people leave Wikipedia for similar reasons. Many good people. Because they didn't like what was going on.
 
To me it seems the level of the physics the site can support is not at all important for the people in charge, as long as the traffic is high enough such that enough questions, answers, and new users are comming in (the site is growing).
 
12:07 AM
@Dilaton It's more of us not discriminating between levels of physics
I'm done here
I've had this conversation many times with you before
 
But in fact it is not important, if this is just a popular physics site or rather like MathOverflow which has a very high level I suspect.
 
You're going on the same old track
 
@Manishearth I'd like to obtain some answers what you really think once and for all and then I am done.
 
@Dilaton I already clarified that: While we prefer high level physics, we will not impinge on lower level users/physics to promote it
@Dilaton I have given those answers
Many times
I don't see what you're getting at here
I'm not going to respond anymore, bye. Rant all you want.
 
Why is wantig to know things ranting?
 
12:22 AM
@ManishEarth I want to know what really is the game now, since the content of the paritucla sites such as physics seems not to be that important as I once thought. And it is acceptable to drive very knowledgable users away because of rules. This is not a rant or anything. Just an observation I have made here and on math for example.
If this site is not here mostly to be userul for the physicists and students to help each other learn stuff, this is ok but should clearly be stated, for example in the about. I suspect many people get the wrong impression otherwise.
Ok, now I am scared...
 
ok guys can you help me understand why in some context quantisation should be unique
 
@ArnavTripathy perhaps, what exactly did you want to know?
 
here's the quotation I'd like to understand
dang it is too long
Once the appropriate phase space \mc{M}_{\Sigma} is identified, the association \Sigma \mapsto \mc{H}_{\Sigma} is made by quantizing \mc{M}_{\Sigma} to obtain a Hilbert space \mc{H}_{\Sigma}. Geometric quantization is not sufficiently well developed to make quantization straightforward in general (or perhaps this is actually impossible in general), but in the case at hand quantization can be carried out by choosing a complex polarization of \mc{M}.
This is accomplished by picking a complex structure J on \Sigma and using the Narasimhan-Seshadri theorem to identify \mc{M} with the moduli space of stable holomorphic G_{\mb{C}} bundles over \Sigma. This moduli space is then quantized by defining \mc{H}_{\Sigma} to be the space of holomorphic sections of a certain line bundle over \mc{M}.
This space is independent of J (up to a projective factor) because of its interpretation in terms of quantization of the underlying classical phase space \mc{M}.
 
oh, hm, it sounds like this might require some familiarity with the context which I don't have
Where is this from, anyway?
 
12:40 AM
hm
a witten address to the AMS
it was titled like ``geometry and quantum field theory'' or something
so here \Sigma is just some orientable surface
and M_{\Sigma} is some space we cook up out of \Sigma that's supposed to be the classical phase space of something or other
in fact it's Hom(\pi_1(\Sigma), G) // G-conjugacy, with some natural symplectic structure
and now the question is how to quantise this fellow
so I don't really know what quantise means first of all
but okay, sure
apparently you can quantise by doing this holomorphic quantisation thing
and then for some reason we expect all these various quantisations to be projectively the same
 
Hm, well, what I'm familiar with is that quantization is the process of replacing physical quantities in a classical theory with operators, to get the corresponding quantum theory
Normally you replace position and momentum with operators satisfying particular commutation relations, and then you can build up all the other operators corresponding to other physical quantities as combinations of those two
but if you start with different operators, you might get a different theory
So the question is probably related to whether all the different choices of initial operators will lead to the same results
 
I always thought that quantization was a process to make a continuous system discrete and operators inevitably turn up along the way (as operators are the easiest way to get discreteness due to eigenvalues yadda yadda), but I guess that means the same thing eventually
 
okay so what I know concretely is that quantization is supposed to associate to a symplectic space a Hilbert space
but I don't really know how these two things are supposed to be related
 
@ManishEarth More like the other way around - quantization is a process of replacing variables with operators, and discrete spectra turn up as a result
 
are you supposed to be able to recover the symplectic space from the Hilbert space, for example ?
 
12:48 AM
@DavidZaslavsky yeah
 
@ArnavTripathy yeah, that's the part I'm not sure about. Sorry.
 
I'm not sure whether quantizing a space here means something different from quantizing the coordinates of the space
You could probably ask that on the main site, it'd get a wider audience
 
@ArnavTripathy maybe you can ask this in one or more questions on our main page?
 
and it seems like a good question
 
12:49 AM
oh man I am pretty bad at asking well-formulated questions right off the bat
okay
 
 
1 hour later…
1:50 AM
@DavidZaslavsky @ManishEarth: I can see that dimension10 has been suspended. I don't wanna ask why...
But, is there any way I can find a clue on "Why" like some chat transcript or comment? ;-)
It's really a very short time...
 
@CrazyBuddy We don't publicly discuss details of suspensions.
 
You can look for stuff, but I won't recommend it
 
I understand... I'd better leave for my colg. ;-)
 
 
4 hours later…
user54412
6:19 AM
"From the scaling of the probes’ performance with star number, we conclude that a fleet of self-replicating probes can indeed explore the Galaxy in a sufficiently short time to warrant the existence of the Fermi Paradox" - from Slingshot Dynamics for Self Replicating Probes and the Effect on Exploration Timescales, on astro-ph
 
10:07 AM
@ChrisWhite Self-replicating probes? Sounds like The Messenger Theory from The Andromeda Strain
Also, where would these probes get the raw material? Landing on planets and mining metal is impractical, because you have to take off after that
 
 
2 hours later…
11:47 AM
@ChrisWhite Btw, I had a chat with my prof about fortran-vs-python-et-al. It seems that Fortran is still used because it's much faster than C/C++ (forget Python; python is quite slow being an interpreted language), and when you're doing band struture calculations of stuff like defects, you need speed.
 
12:08 PM
General statements like "fortran is faster than c++" are not really sustainable in the real world. You have to compare compilers not languages. You also have to keep in mind the time costs of producing the code, which is where fortran wins big in some fields: vast amounts of existing, debugged, highly optimized code.
None the less, the particle physics community switched to c++ in the last 15 years with great success. Our codes were suffering from the interconnectedness explosions in a bad way, and modern fortrans were not available, specified and stable in time.
Another thing that has happened is that we've gotten to a place where even supercomputing resources are cheap compared to the cost of hundreds of grad students and post docs (who are he ones who write most of the particle physics analysis code (thought the main tools are prepared by pros)).
That last point doesn't really apply to theory (those guys still work in small groups even in "Big Science"), but if I want a few thousand CPU hours and the use of Terabytes of RAM during that time, I just ask the grid and it doesn't even make a dent in my projects allocation.
 
@dmckee You particle physicists and our fancy grids
@dmckee Possibly. Though in this case we were comparing compilers
 
Fortran (at least up through watfor/mid-'80s DEC fortran) are easier to optimize than more expressive languages, which used to be a big deal and often may fortran compilers a lot faster than c compiler (for instance). However, both the computing resources available to compilers and the sophistication of the optimizer on the ground have improved a lot. A real lot.
 
And yeah, the existence of optimized code like lapack helps
 
Felix von Lichter (sp?) has some slides about compiler optimizations floating around, where he remarks at one point "gcc is smarter than the codex writer on all platforms". Damn.
 
 
1 hour later…
1:43 PM
@DavidZaslavsky: You owe me the permission from astrobites :P
 
1:53 PM
@CrazyBuddy knock knock...
Is it right to use formulae without knowing how they were derived ?
 
@ShuklaSannidhya Sometimes, depends
If you are at a level where you can understand the full derivation, then you should
The caveats become clearer
 
@ManishEarth Leibniz rule in 11th class?
 
@ShuklaSannidhya Not completely necessary, but it's not hard to prove at your level
i think
 
In calculus, the product rule is a formula used to find the derivatives of products of two or more functions. It may be stated thus: :(f\cdot g)'=f'\cdot g+f\cdot g' \,\! or in the Leibniz notation thus: :\dfrac{d}{dx}(u\cdot v)=u\cdot \dfrac{dv}{dx}+v\cdot \dfrac{du}{dx}. In the notation of differentials this can be written as follows: : d(uv)=u\,dv+v\,du. The derivative of the product of three functions is: :\dfrac{d}{dx}(u\cdot v \cdot w)=\dfrac{du}{dx} \cdot v \cdot w + u \cdot \dfrac{dv}{dx} \cdot w + u\cdot v\cdot \dfrac{dw}{dx}. Discovery Discovery of this rule is credit...
I don't understand this part-
@ManishEarth ^ ?
 
@ShuklaSannidhya Delta means a small change in
$f(x_0 + \Delta x)-f(x_0)=\Delta f$
 
2:11 PM
@ManishEarth how did we get this part - $f(x_0 + \Delta x).g(x_0 + \Delta x)=f(x_0)g(x_0) + \Delta fg(x_0) + f(x_0) \Delta g + \Delta f \Delta g $
 
@ShuklaSannidhya Use the formula I just gave you to substitute
 
@ShuklaSannidhya: Hey ;-)
@ShuklaSannidhya IMO, It's actually wrong :/
If you try to understand the derivation of something and its application, it's you who will be more happier ;-)
I'm not trying to bring philosophy here...
But, If I learn something new and I know the usefulness of it, (if I complete it), I'll be happier the whole day :D
But, that's just me :)
@ManishEarth: I've started to retag the ;-)
Only 17 questions or so...
 
2:29 PM
Great. I was planning to do it partially, but no time
 
I make time ya know :)
Atleast half an hour or so - doing this reviewing :P
@ManishEarth: Can you do me a favor? How many questions do you see in ?
 
Me --> 6... A retagged question comes under "untagged"...
And, it's gone now ;-)
Okay... Either I'm tired or I'm limited in my knowledge (to do more retagging) or I have a vague idea about the remaining questions :)
Thanks to @QMechanic (He retagged a few things along with me) ;-)
Sadly, he won't get my thanks :D
 
3:06 PM
@CrazyBuddy : Team work!
 
@Qmechanic You actually saw it... Thanks QMech :)
 
user54412
3:30 PM
@ManishEarth try writing the same program in fortran and C some time - if you compile with gfortran and gcc, they'll be essentially the same speed, since gfortran is gcc (the only catch is you may have to use -fargument-noalias-global with the C program to turn off alias checking, and of course re-order your loops over arrays to match how the languages use memory)
 
user54412
;)
 
@ChrisWhite Compilation speed or runtime speed?
 
user54412
runtime
 
ah. hmm
 
user54412
C++ may be a little slower, if you bury lots of numerics underneath many layers of library-defined objects
 
3:32 PM
Either way, I hate c.
:P
Pointy things are annoying
 
user54412
lol
 
I hate programmers ;-)
 
user54412
fortran lost my respect the day I saw a program that subtlely overwrote the constant 3 in memory, so that all hard-coded 3s were executed with a different value starting halfway through the code...
 
Wha? Lol
 
user54412
I think they claimed to fix that in a more recent version, but yeah, originally if you assigned a literal to a variable, that variable would be aliased to the same place in memory the code was storing all its literals, to save time on copying the value at runtime, so you could overwrite that value
 
user54412
3:38 PM
@CrazyBuddy agreed :P
 
@CrazyBuddy ... and we hate you...
>:|
 
What..? You guys hate me (Crazy..?) or us (physics people?) :D
 
@CrazyBuddy you are not one of those (physics people)...
:P
 
Hmmm... I think I should figure that out..! ;-)
(gimme some time...)
 
BTW, I was finally able to derive that Leibniz Product rule...
@ManishEarth This world was created with C.
 
3:43 PM
@ShuklaSannidhya I know, but I don't like programming in it :)
 
@ShuklaSannidhya What..? I think I've finally figured out that the cosmologists have been fooling me..! They say that the worlds were created by the big bang ;-)
 
@CrazyBuddy Hmm... You really are not aware what "world" exactly means... Well here is something that cosmologists didn't tell you - Big bang didn't create SE, nor did it create the OS you are running on... the OS which powers the servers which store your FB profile... the browser you are using... your all-time best-friend Google... all these things make the world livable and would not have been possible without C.
 
@ShuklaSannidhya ...and finally, you techies throw away all these as electronic waste and dump this world :D
 
@CrazyBuddy "you guys"?
 
22 mins ago, by Crazy Buddy
I hate programmers ;-)
 
3:58 PM
@CrazyBuddy ... and create a new one...
 
user54412
I thought Dennis Ritchie programmed the Big Bang...
 
@ManishEarth That's old....
 
I know
Obligatory xkcd is obligatory
 
@ManishEarth If you want to impress me, you'll have to try harder... :P
 
4:07 PM
I don't want to impress you :P
 
@ManishEarth this ^
839
Q: What's your favorite "programmer" cartoon?

Dan WilliamsPersonally I like this one: P.S. Do not hotlink the cartoon without the site's permission please.

 
Already seen that
 
4:23 PM
Oy, my professor mentioned a software that could plot output files of most programs. Its name started with x, but it's not xmgrace. Any idea what it is?
I could ask him, but I'd have to wait for tomorrow to get the answer most probably
 
All this talk of programming warms my heart. I think, however, @Manishearth, that Python is being unfairly maligned. In a lot of cases, someone doing a numerical computation is mainly concerned with assembling well-known trusted components (e.g., routines from Lapack or CernLibs) in an intelligent fashion. Python is a good tool for this purpose.
 
@user1504 Hey, don't worry, I love python :)
 
good
 
For my own purposes either would work
And I would prefer python because it doesn't yell at me for using big numbers
However, if I'm writing something which is to be useful to the condensed matter/nano/solid state community, it needs to be fast.
 
@ManishEarth Python is implemented in C.
 
4:29 PM
I know
 
The default Python is, anyways
There are other implementations floating around.
 
user54412
::religious war brewing::
 
lol
@ShuklaSannidhya The key aspect here is that it needs to be as fast as possible at runtime. When you're calculating, say, the charge density for a crystal with defects, you're taking a system that has a bazillion atoms per unit cell. Speed becomes critical.
 
user54412
I actually do most of my coding in python - it's nice, except for its $&@*$& lazy array copying
 
Have you played around with Cython?
 
4:32 PM
Cython O_0
@ChrisWhite Me too. Data analysis I do in JS though; old habit
 
user54412
i haven't used cython alas
 
You might like it. Do your array copying in C.
 
@ManishEarth yeah, you're right... Had Python been implemented in say, Java then it would have been really slow... But still Python is too slow....
 
lol
 
user54412
I will say, python makes me feel productive, but I have much more fun implementing some awesome numerical technique in C for a large simulation on a cluster
 
How many of you here use python as your primary calculator?
 
user54412
memememe
 
@ManishEarth are you talking about IDLE?
 
The Python's word for IDE...
 
4:37 PM
@ShuklaSannidhya no, fire up terminal, type python, and do calculations
 
@ManishEarth yeah sometimes... but I keep my scientific Casio always under my pillow...
In other news, IDLE sucks.
 
user54412
I always have a terminal tab dedicated to python-as-a-calculator - one for every screen
 
@ChrisWhite Try this shell script on for size: github.com/Manishearth/Manish-Codes/blob/master/Bash/hawking.sh
It's python-as-a-calculator with audio
 
@ManishEarth hmm... so you do shell-scripting too... boy!
 
I made it as a joke one day (because espeak sounds a lot like Hawking) and I ended up using it a lot. Hilarity ensues when you use it in front of people who aren't that familiar with Linux
 
user54412
4:41 PM
lol
 
@ShuklaSannidhya I do all sorts of programming
 
@ManishEarth Brainfuck?
The brainfuck programming language is an esoteric programming language noted for its extreme minimalism. It is a Turing tarpit, designed to challenge and amuse programmers, and was not made to be suitable for practical use. It was created in 1993 by Urban Müller. The name of the language is generally not capitalized except at the start of a sentence, although it is a proper noun. Language design Urban Müller created brainfuck in 1993 with the intention of designing a language which could be implemented with the smallest possible compiler, inspired by the 1024-byte compiler for the FA...
 
Tried it, but I'm not proficient
The CTF this weekend required a BF program though. But I wasn't around while writing it
 
@ManishEarth good.. atleast you tried... I never thought of even trying...
@ManishEarth "CTF"?
 
Capture The Flag
8
Q: SIGINT CTF 2013 starts Friday 5 July 16:00 GMT (ended, including write-ups!)

GillesLet's participate in the SIGINT CTF 2013 as team Sec.SE. During this years SIGINT in Cologne the CCCAC will hold it's second CTF. The competition is open to everyone and can be played online. Challenges will be online from July 5, 18:00 CEST until July 7, 18:00 CEST (48h). As always...

 
4:48 PM
@ManishEarth hmm... so that sigint's website has some ssl certificate issues...
 
yeah, I know
 
"I understand the risk"
 
I do understand the risk :P
 
@ManishEarth hmm... so some sort of hacking competition...
 
 
1 hour later…
6:10 PM
@ManishEarth Self-evaluation. All sites or beta only? Consequences?
 
Hm?
Sorry, on mobile
 
Beta only, but it was for all sites for a month or so
Consequences? Makes your problrms clear to you
 
@ManishEarth What's the output? What's done with it?
 
Dunno
The comm team may use it
But its mainly for the community to look at
(the results are posted on meta)
 
6:14 PM
@ManishEarth Ah, OK.
@ManishEarth Thanks!
 
Search chem.se's meta, we have around 4 with half of them having useful conclusions
 
@ManishEarth Yup, seeing it. As an aside: Are you a chemist?
 
6:33 PM
@Gugg Huh? Nope :P
I am a physics (engineering physics, but it's really just plain physics) student
No plans of becoming a chemist
I just like chemistry
far from it
 
OK with me! I just like (intelligible parts of) physics.
 
@Gugg The same goes for me and chem :P
 
 
2 hours later…
8:17 PM
checking size of char... 1
checking size of short... 2
checking size of int... 4
checking size of long... 8
checking size of long long... 8
checking size of unsigned int... 4
checking size of unsigned long... 8
checking size of unsigned long long... 8
checking size of float... 4
checking size of double... 8
checking size of long double... 16
checking size of size_t... 8
checking size of ptrdiff_t... 8
^ Grr, C
 
8:44 PM
@ManishEarth That's why c99 offers stdint.h. And float and double are supposed to correspond to IEEE 754 single and double precision respectively.
 
@dmckee yep. But this software (abinit) is a couple of years old
I wouldn't be compiling it in the first place, except that I'm on an amd processor now and my old abinit directory doesn't work
It works on my Windows dual boot, but behaves a bit strangely. Piping data to it works, but the interactive prompt doesn't -- and that's important for cut3d which has like a billion submenus
 
If piping works and interactive doesn't I suspect a terminal configuration mismatch.
 
@dmckee Cygwin
 
First thing try printenv TERM.
 
Interactive works in a fashion
The prompt comes out after I enter the data
But it's cygwin, so I sort of expected that. Could be the Unix-vs-Windows line endings thing
 
8:52 PM
I can't help you with cygwin. I only ever use windows under protest. All my boxen are unix native in one form or another.
 
@dmckee Yeah. I'm only using Windows because it wasn't working on Ubuntu
I keep a Windows dual boot for software like SolidWorks, adobe stuff, and gaming
still compiling... O_0
> configure: error: C++ compiler cannot create executables
I know how to fix that, but grrr. I waited 15 minutes for that?
 
9:20 PM
oh yeah I remember another question I had
I can find the exact quotation maybe
but it was something like
when you quantise a system
you have one quantum state for each unit volume in the classical phase space
what's that all about ??
 
@ArnavTripathy That's something I came across recently which confused me. I had reconciled it... I think.
thinks for a bit
Sorry, I forgot. I'd need that book in front of me.
Though if you can give a concrete example (like states in a finite lattice), that would make a good question on the main site
 
0
Q: Removing thanks/signature lines

tpg2114I got in the habit on SO of removing thanks and signature lines (and greetings) during edits because they don't add to the question and are relatively unneeded. I think we should adopt the same policy here and remove them from questions where they appear (and ideally not include them when writing...

 
9:43 PM
42 minutes and still not compiled? O_0
 
 
1 hour later…
10:49 PM
damn
yeah I don't have any concrete examples
this was again just a remark in a witten paper
 

« first day (977 days earlier)      last day (3952 days later) »