« first day (4598 days earlier)      last day (324 days later) » 

12:00 AM
hm i see
 
Yeah, so it would probably depend too on how clever the symbolic engine is, and if you fed him with the right identities before doing the computation, it may give you a neater result
 
also serendipitously i found a paper which solves the problem (hopefully) of differentiating a function over SU(2) 0: arxiv.org/abs/2303.11355
 
 
8 hours later…
7:51 AM
Please see this video on p-adics. It's great
 
One thing that annoys me is when theorems just talk about a "submanifold"
What the hell is a submanifold
Is it immersed or embedded???
apparently that article seems to imply that when they say "smooth submanifold" they really mean embedded
 
 
3 hours later…
10:30 AM
@ACuriousMind Regarding yesterday's discussion, I'm more aligned to say that Dirac adjoint spinors lie in the dual representation instead of the conjugate representation, which would explain the inverse matrix (for which the unitarity doesn't work out as the representation is not unitary). As per the transpose part, it's just a matter of row/column vectors
For unitary representations it's the same thing but this is not the case
 
10:44 AM
I just realised that we need to prove $[\phi(x), \phi(y) ]=0$ for spacelike x, y in the Heisenberg picture. But books prove it in the interaction picture by taking the commutator of free fields
But the interaction picture proof is not enough 2 prove locality because the state evolves with time in the interaction picture. So, a measurement at x could still affect y due to the time evolution of the state
But if we work in the full Heisenberg picture, then the state is time independent. So now $[\phi(x), \phi(y) ]=0$ wud ensure locality
The commutator of free fields being zero cannot ensure locality becuz the time evolution of the state vector may still spoil locality
 
@Mr.Feynman I think the problem is that the dual and the conjugate are isomorphic for $\mathfrak{su}(1,3)$, so it's impossible to say which one is the "correct" abstract starting point for what's going on here
but you're right that your particular issue is more easily resolved by looking at the dual rep
 
I also have a way to calculate the field evolution in the full Heisenberg picture
 
@Slereah When no further qualifications are given I'd always read "submanifold" as "smoothly embedded"
 
I have outlined the method here. It shud work becuz of the correspondence between Poisson bracket and commutator :physics.stackexchange.com/questions/730701/…
After using this method, we may prove that the commutator of spacelike Heisenberg picture fields is zero
But books only calculate the commutator of free fields which shud not ensure locality
 
11:12 AM
Apparently one of the old generation of physicist wondering about the geometry of the universe was Helmholtz
 
@ACuriousMind I see
 
He was one of the original "What does it mean physically to consider straight lines" guy
 
11:34 AM
How do people measure a perihelion anyway
I have no idea what that actually looks like experimentally
I know that planets trace weird paths in the sky, is it like a turning point
 
11:50 AM
@Slereah It is less a direct measure as a corroboration between prediction and observations. You must have already the orbital parameters to a certain accuracy for a good enough prediction, and then you take the data nearby the predicted event, and then you do a minimisation curve fit there.
 
Where's the perihelion dammit
 
That is retrograding. It has nothing to do with perihelion.
 
Nothing can be easy
I guess you're not typically gonna see the perihelion in yearly orbits anyway since most planets have a much longer orbit
 
The fact that Mars retrograding is almost yearly, directly points to it being nothing to do with perihelions or aphelions
Every 2 years
 
how does the sky even work
It is more closely related to the point of closest approach apparently idk
Also here is a vibes question : which planets are you allowed to disregard when you do solar system physics
I'm guessing Jupiter is likely to keep and Pluto not so much
what are the big gravitational influences
 
12:06 PM
@Slereah You make lots of observations, project them to the orbital plane, and try to interpolate where the minimum is. It's not too difficult to get a reasonable estimation because the body's speed is maximum at perihelion. And if the eccentricity is high it's pretty easy to see when the body is near the vertex.
 
is there like a standard astronomy textbook for all this
 
Yes, there are prominent texts about this stuff. But don't ask me to name names. :)
 
I have tried reading the Almagest but I'd like something a bit more modern
Let's go to the astronomy SE
 
Sound engineering fail @ Witten's lecture πŸ˜Άβ€πŸŒ«οΈ
 
Astronomical Algorithms by Jean Meeus
let's check it out
if you wish to steal it
 
12:10 PM
OTOH, we don't use analytical celestial mechanics much these days when we want precise solar system dymamics. We do heavy-duty number crunching: direct numerical integration on the n-body problem, for large n.
 
There's a whole chapter on the jewish calendar
@PM2Ring I just need to write out some justifications in my book, I'm not looking to solve astronomy
 
Jet Propulsion Laboratory Development Ephemeris (abbreved JPL DE(number), or simply DE(number)) designates one of a series of mathematical models of the Solar System produced at the Jet Propulsion Laboratory in Pasadena, California, for use in spacecraft navigation and astronomy. The models consist of numeric representations of positions, velocities and accelerations of major Solar System bodies, tabulated at equally spaced intervals of time, covering a specified span of years. Barycentric rectangular coordinates of the Sun, eight major planets and Pluto, and geocentric coordinates of the Moon...
> As of DE421, perturbations from 343 asteroids, representing about 90% of the mass of the main asteroid belt, have been included in the dynamical model
 
@Slereah Hestenes was saying that GA makes classical mechanics so easy that he had no choice but to include some celestial mechanics to make the analysis not look trivial
 
@Slereah It does a decent job of moon alignment
 
lol
 
Lunar theory attempts to account for the motions of the Moon. There are many small variations (or perturbations) in the Moon's motion, and many attempts have been made to account for them. After centuries of being problematic, lunar motion can now be modeled to a very high degree of accuracy (see section Modern developments). Lunar theory includes: the background of general theory; including mathematical techniques used to analyze the Moon's motion and to generate formulae and algorithms for predicting its movements; and also quantitative formulae, algorithms, and geometrical diagrams that may...
> This work culminated into Brown's lunar theory (1897–1908) [...] The number of terms needed to express the Moon's position with the accuracy sought at the beginning of the twentieth century was over 1400
 
I can see why: quite a lot of students get really confused over, say, the various possible ways to make the projectile motion complicated. The one where you shoot up or down a hill, they struggle with the maths. In GA, you just specify the angle of elevation between initial and final point, and you can extract the time of flight and everything.
 
@Mr.Feynman How many theoretical physicists does it take to setup a microphone properly? 🀣
 
One time I had to read a book on lunar theory
 
@PM2Ring Even Newton got a headache long before that!
 
12:15 PM
Turns out the original description of the Mathieu function was in some 19th century book on lunar theory
 
@Slereah Meeus is pretty good. His algorithms for the Sun & Moon are usually within 5 minutes or so of the values from JPL.
 
@Amit none, that's an experimental business
Existence is a trivial application
 
Lol
 
@Slereah My condolences.
 
@PM2Ring this sounds like a classical mechanics test I had two years ago
 
12:17 PM
@PM2Ring Definitely enough to drive one lunatic!
 
The kinetic energy had 14 terms
Now, imagine studying equilibrium and first integrals with a 14 terms kinetic energy
 
The Moon's orbit is really hard because it's so large relative to Earth. It's probably the hardest major body in the solar system to compute. Sure, there are other double body systems (eg Pluto & Charon), but they aren't so close to the Sun.
 
Maybe it's the lagrangian for the city of kineticut
 
@Mr.Feynman Fun. I have C code I wrote a couple of decades ago, using algorithms from the USNO, to compute the 4 major phases of the Moon. It uses 16 sin & cos terms, and is usually within 30 seconds of JPL values. Intermediate Moon phases are harder to compute, but various errors (almost) cancel at the 4 major phases.
@naturallyInconsistent Imagine actually using that beast to prepare an ephemeris that astronomers & navigators can use. Sure, they had mechanical calculators, but it still involved a lot of drudgery.
 
@PM2Ring Throw in some cockroaches
 
12:29 PM
In a way, we're lucky that the Moon is hard to predict. Ptolemey's algorithm got the position roughly ok, but it was clearly not perfect. And his model implied that the range in the Moon's angular diameter over the monthly cycle would be huge. IIRC, the maximum size should be double the minimum size, which is obviously BS. So there was some pressure to devise a more accurate model of the cosmos.
The perihelion of the Earth is a bit messy because of the orbit of the Earth & Moon about their barycentre. I have some info on that here: astronomy.stackexchange.com/a/49605/16685
 
Some pressure from whom
Big Moon?
 
@Slereah Yeah, ok. Lambert devised a great algorithm for determining orbits from a few observations. His method was very useful when they started discovering asteroids.
 
Who is Lambert
is it Christopher Lambert
Is it the Lambert W function guy
 
Also, with analytic celestial mechanics it's annoying to find position as a function of time: you have to invert Kepler's equation. So it's a lot easier to find time as a function of position.
 
12:37 PM
@Slereah Give me a minute...
In celestial mechanics, Lambert's problem is concerned with the determination of an orbit from two position vectors and the time of flight, posed in the 18th century by Johann Heinrich Lambert and formally solved with mathematical proof by Joseph-Louis Lagrange. It has important applications in the areas of rendezvous, targeting, guidance, and preliminary orbit determination.Suppose a body under the influence of a central gravitational force is observed to travel from point P1 on its conic trajectory, to a point P2 in a time T. The time of flight is related to other variables by Lambert's theorem...
 
Thanks
It was indeed Lambert W guy
 
Johann Heinrich Lambert (German: [ˈlambɛɐ̯t], Jean-Henri Lambert in French; 26 or 28 August 1728 – 25 September 1777) was a polymath from the Republic of Mulhouse, generally referred to as either Swiss or French, who made important contributions to the subjects of mathematics, physics (particularly optics), philosophy, astronomy and map projections. == Biography == Lambert was born in 1728 into a Huguenot family in the city of Mulhouse (now in Alsace, France), at that time a city-state allied to Switzerland. Some sources give 26 August as his birth date and others 28 August. Leaving school at...
@Slereah A veritable genius
 
I mean just look at his skull
Giant brain
 
Even puts Witten to shame.
 
I dunno
It's pretty huge
 
12:45 PM
Ok. :)
 
Lol
 
Anyway, the Earth-Moon interaction means there's a slight wobble in the Earth's axis. You can see the equations for that here: neoprogrammics.com/nutations/nutations_1980_2000b from neoprogrammics.com There are also small irregular unpredictable wobbles due to crust motion and weather in the atmosphere, ocean, and mantle.
Newton realised that you could use the position of the Moon as a clock. So navigators could determine their longitude, if sufficiently accurate tables of the Moon could be prepared. However, it's hard to make those tables, and even with good Moon tables you need to be an excellent navigator (and have relatively calm seas) to make the necessary observations with sufficient precision. If you can fix your position with an error of <12 nautical miles, you're doing very well.
So it turned out to be more practical to spend a big chunk of cash developing good chronometers. :)
 
1:09 PM
Who needs a watch when I have a gigantic table of the moon
 
Exactly
 
A true "look up" table!!! 😁😁😁
 
Some stuff on the anomalous precession of Mercury's perihelion: astronomy.stackexchange.com/q/49253/16685 & astronomy.stackexchange.com/q/44654/16685
The latest analtical lunar theory is the ELP: Éphéméride Lunaire Parisienne "ELP has been fitted not directly to observations, but to the numerical integrations known as the Jet Propulsion Laboratory Development Ephemeris"
 
Is there a general relativistic lunar theory
 
1:33 PM
@Slereah Well, the JPL DE incorporates relativistic corrections, so ELP inherits that.
 
Does the moon eventually crash into the Earth from gravitational radiations
Adjust your calendar accordingly
 
There's some info about the relativistic stuff & how they calculate the Moon in The JPL Planetary and Lunar Ephemerides DE440 and DE441, Ryan S. Park et al 2021 AJ 161 105.
@Slereah Nope. The Moon's orbit's getting bigger due to tidal interactions.
 
But for how long
 
The loss of angular momentum due to gravitational waves is several orders of magnitude smaller than the tidal effects.
Eventually, the tidal loss will diminish as the orbit gets bigger. And eventually the ocean will be lost and the Earth will solidify, so tidal losses will be minimal.
The Earth would eventually get tidally locked to the Moon, so a day would be as long as a month. With the bigger orbit, that would be about 50 days. But the Sun will go red giant before that happens.
It's possible that the Earth-Moon system will survive the early red giant phase, but if it does we're likely to be swallowed by the Sun in the later phases.
Of course, the Earth will be uninhabitable long before that happens. It will be too hot even for extremophile bacteria in a couple of billion years, well before the red giant phase starts.
In ~half a billion years, photosynthesis will be impossible. See en.wikipedia.org/wiki/Timeline_of_the_far_future
Even if we establish colonies on Mars, or orbital platforms further out, the Solar System may not be a pleasant environment when the Sun goes red giant. Those guys spew out a lot of gas & dust. The Sun will have lost ~50% of its initial mass by the time it becomes a white dwarf.
 
1:57 PM
Space engineering is rough
 
Does the model take into account that time they crashed a probe on the moon
 
Time, singular? Lol
When the probe hits the moon like a big a** kaboom that's a more (useful waste of money)
 
These are tables of space probes (typically orbiters or components thereof) which have been deliberately destroyed at their objects of study, typically by hard landings or crash landings at the end of their respective missions and/or functionality. This endeavor not only precludes the hazards of orbital space debris and planetary contamination, but also provides the opportunity in some cases for terminal science given that the transient light released by the kinetic energy may be available for spectroscopy; the physical ejecta remains in place for further study. Even after soft landings had been...
Guess there's a few
No probes into the sun yet though
 
I wonder if there are even more non deliberate ones
 
@Slereah The momentum of such impacts is negligible. But they do model the Moon as a ball of wobbly stuff.
 
2:06 PM
Time to send a BIG probe
There was that cold war project to nuke the moon to show the soviets who has the largest wiener
 
We've been making lunar distance measurements using lasers for >50 years. And the techniques of gathering and analysing that data have improved over the last decade or so. The lunar laser reflectors aren't evenly placed over the Moon's surface, and smart people have figured out how to properly compensate for that.
Recent Progress in Lunar Laser Ranging at Grasse Laser Ranging Station agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2019EA000785
 
Are you a moon scientist
 
No, just a loony. :D
 
A lunatic
Me I'm a moon hater
 
The Moon is interesting in its own right, but accurate Moon position info is important for several reasons. We'd like to be able to model the Moon's orbit into the far future (and distant past), so we need good current position & velocity values.
But mainly it's because the Moon wobbles the Earth, so it affects every precision Earth-based astronomical observation.
Astronomers have a love-hate relationship with the Moon. It's fascinating, but infuriating. You can see surface details without a telescope, but its light makes it hard to see other things in the night sky.
 
2:27 PM
I live in the city, I can barely see the moon itself
Clouds and lights
 
I also live in the city. But at least I can still see the Moon, planets, and a few bright stars, weather permitting. For a couple of years we had a lot of rain, and very few cloudless nights. But we've had a lot of dry weather for the last several months, thank goodness.
The last place I lived was rural. I could easily see the Magellanic Clouds.
 
What if, instead of parallel transport
Perpendicular transport
 
2:51 PM
Perpendicular transport is less useful when you're not using Euclidean geometry.
 
3:09 PM
I do talk about the moon a little in my book so far but mostly for the cosmic distance ladder
Since you can easily do parallax measurements for the moon
 
3:19 PM
Most GR books fail to mention the first step of the distance ladder tho
The stick
 
Also, the hand span.
Hand-based measurements of angular distance are convenient in casual astronomical observation. abc.net.au/science/articles/2009/07/27/3169109.htm
 
Apparently in addition to the usual homogeneous manifolds that are listed for FRW metrics, there exists other lesser known ones
Because the big theorems involve homogeneous symmetric manifolds
But there are homogeneous non-symmetric ones
I don't know if any of them are 3 dimensionals to apply to cosmology though
 
Egyptians nearly invented qubits a few millenia early, cubits... lol
 
3:40 PM
I'm certain that "cubit" influenced the creation of the term "qubit".
For some bizarre reason, Americans love those ancient measurements, like inches, feet, and cubits. Eg, vatican.va/archive/ENG0839/_P27.HTM ;)
 
The cubit is pretty cool in the sense that it was an early effort of standardized measurement
 
@PM2Ring Really? That's news to me
 
Pre-Sargon of Akkad, measurements were a big old mess
 
Well, he was a ruler
 
I've heard that joke before
 
3:50 PM
@Amit I'd be very surprised if Ben Schumacher had never heard of cubits. en.wikipedia.org/wiki/Benjamin_Schumacher
 
> The coining of the term qubit is attributed to Benjamin Schumacher. In the acknowledgments of his 1995 paper, Schumacher states that the term qubit was created in jest during a conversation with William Wootters.
 
@PM2Ring Ah interesting, he does have a strong philosophy & art background apparently
 
And works at an Episcopalian college.
 
Yeah, I wonder what was the jest :)
I mean context of it
 
4:05 PM
@Amit No idea. I guess you could email him...
There's an amusing anecdote connected with this term:
In quantum field theory, penguin diagrams are a class of Feynman diagrams which are important for understanding CP violating processes in the standard model. They refer to one-loop processes in which a quark temporarily changes flavor (via a W or Z loop), and the flavor-changed quark engages in some tree interaction, typically a strong one. For the interactions where some quark flavors (e.g., very heavy ones) have much higher interaction amplitudes than others, such as CP-violating or Higgs interactions, these penguin processes may have amplitudes comparable to or even greater than those of the...
 
Lol, the illegal substance is a funny twist
 
4:22 PM
Probably hash. Weed wasn't very common in Europe in those days.
 
 
2 hours later…
6:10 PM
@Slereah is that a thing?
 
it's not
Although parallel transport with the LC connection does preserve angles
 
7:09 PM
@PM2Ring @SillyGoose
 
7:21 PM
@Slereah thanks god
 
7:42 PM
does the general discussion of quantization refer to "Canonical quantization"? i asked one of my professors to distinguish what is meant when people refer to quantizing something, and they said it simply means to make quantum, but i think there should be something more specific? another professor said that in any quantum class i should have learned to quantize a particle, but i am not really sure that is the case or what it means
 
@Relativisticcucumber "Quantization", in general, means "start from a classical description of a system and obtain a quantum description of it"
"canonical quantization" refers to the most common way to do that
 
If he gives such a vague answer he probably means canonical quantization
 
ok i think i see. so i do not feel i have done this process of canonical quantization XD ok thanks
 
@Relativisticcucumber have you argued in some way that QM should have $[x,p] = \mathrm{i}$ and then gone from there, in particular just re-using the classical Hamiltonian with $x$ and $p$ being operators instead of numbers? then you've essentially done canonical quantization
 
ok i see. i was also actually just thinking the harmonic oscillator formalism seems in line with what wikipedia considers canonical quantization. i think it seems to require creation and annihilation operators?
 
7:49 PM
not necessarily?
you get the c/a operators for the QHO just by $x\pm\mathrm{i}p$, they're not that special, just a neat switch of variables
 
it's a polarization of the phase space πŸ˜”
 
ok so this process only requires establishing commutation relation and then establishing operators and doing mech in that formalism? so then what is "quantized" about this
i thought some aspect of discreteness should enter the picture
 
@Relativisticcucumber that you have non-commuting operators
the discreteness just enters because operators like the angular momentum or the Hamiltonian will usually have discrete (parts of their) spectra
 
Also note that not all quantizations imply discreteness
The quantization of the free particle has no discreteness to it
It only occurs when you have some compact operator
 
i see. ok that makes sense
 
8:04 PM
the old school quantization had a much more discrete result oriented method called the Bohr-Sommerfeld condition
Where the trajectories in phase space had a condition $$\int p dq = n h$$
 
0
Q: Should new tags only be created when **no** other tags apply, or just generally when there's not one that applies to a given relevant topic?

Logan J. FisherOn the privileges page for creating tags it says: When should I create new tags? Most common tags already exist on a mature site. You should always favor existing tags; only create new tags when you feel you can make a strong case that your question does cover a new topic that nobody else has as...

 
This condition is still sort of true today but you need to qualify it
 
8:54 PM
@naturallyInconsistent Liked your answer but I don't quite understand, isn't that $T$ that gets increased is of the box rather than the object within the box, in as per the OP's imaginary experiment?
 
@Amit The stuff inside has to be in thermal equilibrium with the radiation inside the box, so there is no difference.
 
@naturallyInconsistent I see, I suppose it could be nice to actually see that equation applied to the scenario he is describing, because there he appears to suggest already starting with a system in thermal equilibrium, so idk how that plays out exactly...
 
@Amit But the equation I was talking about can already be directly applied. I mean, you use it to derive the temperature of the Earth, and that should be more than good enough to start.
 
@naturallyInconsistent Right. I also see your latest comment... and I agree it's more important to answer the heart of the matter than auxiliary stuff -- just note, you begin your answer with "Actually it does" which can be interpreted as answering yes to his question about the described experiment, I don't know if you meant that literally
 
No, I meant for the crucial thing.
Note that his question is worded ambiguously, and that
is precisely why I chose to interpret it as asking the question we want to answer, rather than pedantically what he wrote.
 
9:08 PM
Yeah, yet there's only one literal question mark in his text lol and that's in: "Will the object get warmer from re-absorbing its own radiation?"
 
And the answer to that is yes.
Since he did not state that there would not be outside radiation influence, I can add that for him.
miahahahaha
 
lol, as long as he understands that, it's fine... I mean, probably what you wrote will be inscrutable to the OP anyway
at least the mathy part
 
If there is no incoming radiation, then the equilibrium temperature of all objects will be absolute zero. Or maybe we can take CMBR as the minimum.
@Amit That would be sad. That is a simple algebraic relation. And if I were to make it any more simpler, I would have to draw arrows for $\varepsilon$ decreasing, so $T$ increases
That would be rather condescending. Not that I cannot be condescending to climate change deniers
 
lol
 
for even I have a limit to patience
I am not John Rennie
 
9:14 PM
the best layman explanation I came across was in another post where someone pointed out that it's the wavelength shift that's really causing the trouble. that is, energy comes in the form of visible light mainly, but emission rate is less really because of the conversion to IR wavelengths, that's why the GHGs really work only "one way"
 
That is a very nice way to put it indeed. But I was only given one emissivity in the exam, probably engineered precisely to give the correct mean temperature of Earth.
 
Sounds fun, I never took a proper thermodynamics/stat mech (they're the same right? <_<) course
 
9:34 PM
hahhahah, no
But very rarely do we entertain classical thermodynamics as a module in and of itself, which is sad; students tend to be confused if they were just thrown into the depths of introductory statistical thermodynamics
 
Are you saying that the name of the course makes it less frightening? lol, maybe I'm not following
 
I can see an argument that you should have seen the intuitive mess that is thermodynamics before trying to make sense of it with actual statistical mechanics
unfortunately, I never really made sense of statistical mechanics either :P
 
9:52 PM
Found out something interesting today, if a question gets migrated, it allows for double voting apparently ^_^ , also had a ponderment about that... what happens if the OP or the answerer never actively joined the target SE site... is it legit to 'force' a new account creation on them, or does it block the migration? Or maybe it gets created with a ghost account that's only kinda half active?
 
@Amit The user name just isn't clickable on the target site - there's no profile associated with the question until the user themself creates their account on the target site manually
 
ah, cool, that's very reasonable
 
In meta news: This is the data SE claims justifies the current debacle.
 
@ACuriousMind I would have thought that stat therm is much easier to understand than all the symplectic shit that you have done...
 
@naturallyInconsistent the course I had on it was really terrible :P
 
9:58 PM
@Amit No, I meant it as ACM read it, which is that stat therm is too difficult to jump straight into, and a simplistic classical therm module should be given as an intro to the intro to stat therm
@ACuriousMind bad teaching is something I feel stabby stabby over
 
@naturallyInconsistent That makes sense, I get the impression many of that stuff look highly unmotivated otherwise
 
To be fair it was the same semester I took QFT and I found that vastly more interesting from the get-go even without the difference in pedagogy
 
WTF
I mean, how are people supposed to do the preliminaries to QFT without some stat therm background?
I always thought that stat therm is quite a necessary thing to understand before you can understand why we have to do QM, and current pedagogy wants QM before QFT
 
you just apply the ideas of canonical quantization to classical field theory, what do you need statistical mechanics for?
 
And then you need QM to understand Atomic and Molecular.
A&M and stat therm are prerequisites to solid state
@ACuriousMind very rarely do people even get to see classical field theory, outside of basic intro to electrodynamics, worded in ways that does not even begin to look like QFT
 
10:04 PM
@ACuriousMind Interesting :P What do you think about this
 
Also, to me, stat mech is even worse than stat therm. To me, stat therm is the thermodynamics leaning side of things, and while we have quite a lot of dealing with atoms flying about, we try out bestest to avoid a microsopic derivation of things. Stat mech is then the next level upgrade, where we use the tools of stat therm and derive the flow of nearly ideal gases, etc. Those make me sick.
 
@BalarkaSen It doesn't make sense to me: Their "gold standard" is a weird metric easily gamed, they dismiss the potential effect of the 30 min answer rate limit to explain the trends with a single sentence, and compared to what we mods saw, they remove the explicit mention of the number of actually questionable suspension appeals because they knew it its smallness would make them look bad.
most of the objections and questions you're seeing in the comments and answers have been previously raised (and not been responded to) in private
 
Huh
 
I think this data would've been a good starting point for a discussion about what's wrong, but it's not so clear-cut that it would explain SE making a complete reversal on their policy towards GPT answers out of nowhere
 
@ACuriousMind What did you get to cover in classical field theory? I always found treatments to be so different between CFT and QFT, as to be incomparable.
 
10:15 PM
@naturallyInconsistent just "basics": What a functional derivative is, Hamiltonian/Lagrangian formulations, maybe solve for the waves on a classical string, it was like 3 weeks at the end of the 2-semester classical mechanics course
maybe just 2 weeks, I don't remember (it's been a decade D:)
 
@ACuriousMind That is just horrendous. I do not understand how people are supposed to understand how things come together by having this little. Hmm.
 
eh, worked for me :P
 
lol
I suppose I did spend way too much time fighting over every experimental building block to the acceptance of quantum theory.
 
Sometimes it is less confusing and more streamlined to know a little as opposed to a lot
 
But then the little would have to be fully understood when covered, and that is incredibly difficult to do. Most people require some examples to see how the phenomena are described by the mathematics.
 
10:23 PM
Perhaps, I don't really know physics enough to comment. I have seen people trying to learn math by doing completionist runs and that is almost always a recipe for disaster.
I am a big fan of taking (quasi-)geodesic paths to the thing I want to understand at the end
 
I'm a big believer in living with uncertainty (not the quantum notion)
don't disregard things you don't fully understand but don't let it stop you - see where the path leads and then go back and fill in the potholes
partly it's the teacher's/writer's job to present stuff in a way so that this works, but I feel often people get stuck too easily on details
 
@ACuriousMind That I am perfectly fine too. I often organise my teaching so that students can get different levels of understanding at one go, and then slowly come to understand the same system in greater depth later.
@ACuriousMind I feel that it is often the case that the teachers are at fault rather than students. Often in physics it is difficult to get professors to understand the simple idea that, FFS, the professors themselves obviously would not be needing the rigour, but the poor students who are listening, have no idea what the dayum symbols even mean!
 
FWIW, I know too many physics grad students who think symbolic fiddling, without knowing what the intrinsic meaning of the symbols are, will magically lead to understanding. This seems like a occupational hazard for physics undergrads going into their PhDs
 
One thing that is fun, however, is that because I write things down in a way that is too verbose, in anticipation of student misunderstandings, it is often the case that the students would omit some parts of the notation I am using, and then I get to scold them for misunderstandings when their simplifications catch up with them. Happens rather regularly.
@BalarkaSen That is rarely the case on the physics side. It is more the norm on the maths side. The entire French school, for example. But they do produce wonderful results doing things that way.
 
I don't think that was the problem with the French school.
By symbolic fiddling, I mean for example endless manipulations with differential equations in local coordinates.
I know people who are actively averse to thinking coordinate-free, which makes life simpler conceptually very often, from my experience conversing with physics grad students.
 
10:33 PM
@BalarkaSen They do well; ergo they must be doing something correct.
 
@BalarkaSen I agree
 
@BalarkaSen But this is necessary. One can only deal with complicated scenarios by having done enough work on them.
@BalarkaSen standard uni physics education does not emphasise this as being important...
 
I am not saying coordinate computations are unimportant, often they're the right thing to do. But often they're also not the right thing to do. The pragmatism is lacking; "complicated scenarios" may not be that complicated from a different point of view.
@naturallyInconsistent The French school actually understood their category theory extremely well. The problem was not one of understanding, but their disassociation with a rich source of concrete examples in other branches of math.
They had strong intuition for when to apply a certain kind of abstract, categorical argument to prove certain things.
This is my impression, at least.
Anyhow, that's old story now. These hyper-abstract trend of French mathematics, happily, no longer exists :) We have French differential geometers and probabilists now.
 

« first day (4598 days earlier)      last day (324 days later) »