All Things Physics

Disappointed I won’t be at the winter meeting to see this. Congrats Rhett!

YouTube video by All Things Physics Taylor Series | Mathematical methods

I like Boas for a first course. Arfken is more intimidating, but fine for more advanced students. I made a video on Taylor series for such a course; i have my students watch it in advance of my presentation, so we can have a discussion. I hope to make many more in the future. youtu.be/GINFaUB9zrc

Don’t ever send an email when angry! Sleep on it!

Not the “good” ones! But yes, we have a not-so-good one and it’s still pretty cool!

YouTube video by All Things Physics Taylor Series | Mathematical methods

New video just released on the beauty of Taylor series! Made with manim_community and designed for #physics students. #iteachphysics

youtu.be/GINFaUB9zrc?...

Welcome to the club. It is a little bittersweet, but then you start to love it!!

Wow! That sounds like the course I took as an undergraduate. Thankfully, it’s not the course I teach! (There’s a lot of flexibility at a small, liberal arts college.)

No one has mentioned thermodynamics. I think a minimal coverage is important.

Thanks! And I enjoyed yours. You and I share some programming history.

Heh…I usually continue revising my talks until right before I give it!

And yes, I do realize it’s nearly impossible to foresee benefits before the fact.

Don’t get me wrong. I’m a theorist at heart, and agree that theories are cheap. But I do wonder whether pushing experiments to the quark/string/whatever level would have any benefits beyond ruling out a few theories, and the expense is likely to be staggering.

You’re right, it could be asked of any science. I guess I’m interested in fundamental physics because the experiments that are undoubtedly necessary to rule out some of the theories get more and more expensive as we get more fundamental. Is there a point when the rewards are not worth the effort?

I guess that’s true. But this will likely lead to competing theories that will eventually need to be decided by larger and more expensive experiments. At some point this becomes unsustainable, and I’m wondering where this point lies.

I’m asking Sean, of course, not you.

Okay, so I appreciate the clarification. Now I’d like an answer to my original question. When does searching for a final theory no longer make practical sense (assuming it relies on public funding)?

I don’t even think mathematical logic is necessary. Okay, we “know” we don’t have a final theory at this point. But what would it mean to “know” we have a final theory, even if “know” simply means “have in hand”? Does final theory here just mean something that connects QM and gravity?

Because science works by finding flaws in theories to improve them. There is no mechanism by which a theory is ever proven correct (or final).

I’m curious about your statement that “We don’t know the final theory.” The nature of science is such that we will never know the final theory. So my question to you is, when does searching for a final theory no longer make practical sense (assuming that public funding is necessary)?

Heh…love the video! We should seriously consider collaborating on a project in the future!

Agreed. I just get frustrated when people are opposed to using it at all rather than finding out how it’s useful.

It does take a little practice, but you learn quickly to be very detailed.

It’s also been good at things like making tv show recommendations, based on some preferences, or providing directions for how to get somewhere in a strange city, including a combination of walking, bus, and trains.

Here’s a “google search” example. We were at a vacation home and saw a strange appliance. We sent a picture to ChatGPT and asked: what is this device and how do we use it? And it told us exactly what we wanted to know.

The latest thing I tried was to: write a manim program to create a ten by ten by ten grid of cubes that are arranged in the form of a larger cube. I know that’s not a difficult program, but working out the spacing would have taken me a while and it gave me a perfect program in about 5 seconds.

No. I use it for info gathering and programming tasks, among other things. See my response to Dan Schroeder.

easier to find information that I would have normally “searched the web” for. It’s also incredibly useful for helping me write code. If you have a block of code that will require nested loops, for example, a good prompt will have the code written in 6 seconds, saving me hours of frustration.

AI can be incredibly useful (if you don’t think so, then you either haven’t tried it, or you’re in denial). But so far there’s little sign of actual intelligence.

You’re right, it does sound thunderous!

Where’s the video?