Alex Ji
@alexji.bsky.social
Near-field Cosmologist and Assistant Professor of Astronomy and Astrophysics at University of Chicago. he/him
I agree that's the issue; in principle ugrad QM is useful, in practice it depends on who teaches it. With enough effort available, pure astro classes (stars, radpro, galaxies, etc) would cover the relevant aspects more reliably, if they view teaching the physics as part of the class and not a prereq
November 19, 2025 at 4:17 PM
I agree that's the issue; in principle ugrad QM is useful, in practice it depends on who teaches it. With enough effort available, pure astro classes (stars, radpro, galaxies, etc) would cover the relevant aspects more reliably, if they view teaching the physics as part of the class and not a prereq
I was thinking about things like zeeman effect, hyperfine splitting, forbidden transitions, and the general concept of perturbation theory; which I think are usually not covered in a modern physics class.
November 19, 2025 at 3:13 PM
I was thinking about things like zeeman effect, hyperfine splitting, forbidden transitions, and the general concept of perturbation theory; which I think are usually not covered in a modern physics class.
I think it is pretty important to have at least some quantum for atomic physics and the hydrogen atom, which is a huge part of OIR spectroscopy (also statmech). The hard part is that depending on how quantum is organized, this can be anywhere from 1 quarter to 1 year.
November 19, 2025 at 1:13 AM
I think it is pretty important to have at least some quantum for atomic physics and the hydrogen atom, which is a huge part of OIR spectroscopy (also statmech). The hard part is that depending on how quantum is organized, this can be anywhere from 1 quarter to 1 year.
Reposted by Alex Ji
More coverage of OSU’s new restrictions on conference attendance:
“These conferences are open forums for us to present our work, network and collaborate. Telling us that they’re going to regulate whether or not we can participate feels very stifling, prohibitive,” said @ohdearz.bsky.social
“These conferences are open forums for us to present our work, network and collaborate. Telling us that they’re going to regulate whether or not we can participate feels very stifling, prohibitive,” said @ohdearz.bsky.social
Under Anti-DEI Pressure, Ohio State Limits Conference Funds
The Education Department recently criticized Ohio State University’s involvement with a nonprofit that encourages people from underrepresented backgrounds to pursue doctorates. Now, OSU is rethinking ...
www.insidehighered.com
October 15, 2025 at 12:51 PM
More coverage of OSU’s new restrictions on conference attendance:
“These conferences are open forums for us to present our work, network and collaborate. Telling us that they’re going to regulate whether or not we can participate feels very stifling, prohibitive,” said @ohdearz.bsky.social
“These conferences are open forums for us to present our work, network and collaborate. Telling us that they’re going to regulate whether or not we can participate feels very stifling, prohibitive,” said @ohdearz.bsky.social
Fair, but I think this is still less qualitatively different than you think; eg having tutored O(100) CS101-type students, something like 20-30% of people their attitude was “if i remove all the red squiggles from my IDE that means I did it right” until they learned otherwise
October 9, 2025 at 5:08 PM
Fair, but I think this is still less qualitatively different than you think; eg having tutored O(100) CS101-type students, something like 20-30% of people their attitude was “if i remove all the red squiggles from my IDE that means I did it right” until they learned otherwise
I think there's something where "did something that works" goes too far beyond "understanding what's actually happening" that is the issue here? And part of this is that abstractions that were initially leaky became more solid (e.g. no one worries about transistors failing now)
October 9, 2025 at 3:47 PM
I think there's something where "did something that works" goes too far beyond "understanding what's actually happening" that is the issue here? And part of this is that abstractions that were initially leaky became more solid (e.g. no one worries about transistors failing now)
Yes, but in this analogy this is like asking gcc to make a fancy figure for you; like technically it's able to do this, but you need to give it more guidance?
October 9, 2025 at 3:30 PM
Yes, but in this analogy this is like asking gcc to make a fancy figure for you; like technically it's able to do this, but you need to give it more guidance?
I agree (though I didn't do binary!) and think there's a similar gap going from high level languages like python to LLM-based coding; there's something missing for most users. But at the same time, it really broadens access in a way that I think is valuable.
October 9, 2025 at 3:28 PM
I agree (though I didn't do binary!) and think there's a similar gap going from high level languages like python to LLM-based coding; there's something missing for most users. But at the same time, it really broadens access in a way that I think is valuable.
Reposted by Alex Ji
We must now start calling this what it is: Trump’s Invasion. It started with federal agents, it will soon include deploying federalized members of the Illinois National Guard against our wishes, and it will now involve sending in another state’s military troops.
October 6, 2025 at 1:44 AM
We must now start calling this what it is: Trump’s Invasion. It started with federal agents, it will soon include deploying federalized members of the Illinois National Guard against our wishes, and it will now involve sending in another state’s military troops.