Oh this turned out well! Here's an illusion I made. Stare directly at the face for several seconds and a green cloud will appear between the trains. But then when you look directly at the gap between the trains, you see gray. The more fixed your gaze is and the longer you wait, the better it works.

I think many people assume that sunlight contains mostly visible light. But there's a huge amount of energy also present in the non-visible frequencies. This is true whether you plot as a function of frequency or as a function of wavelength. So I made these plots (black-body spectra for simplicity).

The squares may appear to shift vertically or horizontally, but they are actually stationary and perfectly aligned in either vertical or horizontal positions. References Flynn, O. F. & Shapiro, A. G. (2018). i-Perception, 9(6), 1–6.

I was looking at a family scrapbook with my 6 YO daughter when she saw a photo of her sister and asked "is that AI?" I was surprised. How does she even know what AI is? (She's never used it.) Also, why would anyone use AI to generate a fake photo of a family member to paste in a scrapbook?

Impressive. But I've seen more: www.youtube.com/watch?v=hVr1...

Last July Physics Magazine asked readers to weigh in on some of the biggest controversies in cosmology, particle physics, and general relativity. The survey results are now available. Only one topic achieved near consensus: how to think about the big bang.

The goblin in this illusion that I made is always gray in reality, but it looks slightly green at first because of color constancy effects, and it looks bright green if you stare at the black smiley face for a while, because of visual afterimage effects. The longer you stare, the greener it gets.

The goblin in this illusion that I made is always gray in reality, but it looks slightly green at first because of color constancy effects, and it looks bright green if you stare at the black smiley face for a while, because of visual afterimage effects. The longer you stare, the greener it gets.

Oh this turned out well! Here's an illusion I made. Stare directly at the face for several seconds and a green cloud will appear between the trains. But then when you look directly at the gap between the trains, you see gray. The more fixed your gaze is and the longer you wait, the better it works.

Here's a fun illusion I made. If you stare directly at the center for a while, a red cloud seems to appear in front of the snake's head.

Here's a stylized version of White's illusion that I just made for a class. The short bars are all the exact same color in the image, but the group of short bars on the left looks lighter than the group of short bars on the right.

I took the original image and blurred it uniformly and this was the result, indicating that the smudged dots suggest to the mind that each square is rotated by a certain amount. I had made it so that the effective square rotations suggested by the smudged dots follow hyperbolic grid angles.

I took the original image and blurred it uniformly and this was the result, indicating that the smudged dots suggest to the mind that each square is rotated by a certain amount. I had made it so that the effective square rotations suggested by the smudged dots follow hyperbolic grid angles.

This illusion I just made is a checkerboard version of the café wall illusion. This is literally just a normal checkerboard (i.e. perfect squares distributed on a perfectly square grid) with some smudged dots. However, it appears to be a hyperbolic grid. (You may need to squint your eyes to see it.)

"Fluorescence" It's such a cool physical effect. Why does it have to have such a weird spelling?

Here's Lego art that I made that shows an image of my wife. The inset shows what it looks like from far away, from which vantage point the face seems more real because your brain blends together the tiles and adds in details.

More ribbon art that I made. Can you tell who it is?

I'm playing around with an algorithm that I made to create ribbon art.

I made a simultaneous contrast illusion for the first time and I think it turned out pretty good. The stripe on the cab of the truck looks red in the first image, purple in the next, blue in the third, and yellow in the last. But within the image they are actually all the same color: gray. 🧵1/3

I may be a purist, but I never lie to my children. You don't have to. From birth, I have told my kids that Santa Claus is not a real person but is a real symbol of giving, and the tooth fairy is not a real fairy but is a real symbol of growing up, and the Easter bunny... My kids handled it fine.

Technically, the Artemis astronauts were in "free fall" during 99% of their 9-day journey to the moon and back, but during the seconds when they were re-entering Earth's atmosphere and fireballing, they were not in free fall.

When I write, using the em dash is my passion. I refuse to stop using it—even if AI abuses it. I have to limit myself to using the em dash to threes times per page—at most—so that I don't overuse it.

Cool, but to be clear, the moon doesn't have to be exactly 400x closer to the earth than the sun for total eclipse, it just has to be *less* than 400x closer. In other words, total eclipses of a host star by a moon as seen from the moon's planet are more common than this makes it sound.

Yes! I already changed my computer's background image be one of the photos snapped by #ArtemisII yesterday, and it looks gorgeous, like it belongs there and will always be there.

... and building machines that can make observations that human eyes cannot make directly, and developing models that can successfully predict the observations, and making hypotheses and testing them with observations.

science is really just reporting observations

In my view, histories of science books by scientists tend to get the history wrong/oversimplified, while those written by historians tend to get the finer details of the science wrong, which often goes unnoticed by general readers but is infuriating to scientist readers. Joint-written books are best

This turned out better than I expected. Happy Pi Day!

On the moon, g is 1.625, so according to your logic, pi = 1.27 on the moon. You can't simultaneously choose L = 1 m and T/2 = 1 s. By doing so, you are treating pi as an adjustable parameter. Rather, to get T/2 = 1 s on earth (which is a grandfather clock) and with true pi, you need L = 0.994 m.

Seen on this morning’s walk

My TED-Ed video on lasers is finally published, and it already has 121,000 views on YouTube. www.youtube.com/watch?v=cmO2...

My kids got out a game they haven't played in a while and forgot a rule. My daughter: "I'll just do a google search on my phone. This is what AI says-" Me: "No, don't do that, it will just give you nonsense." My daughter: "It should be close enough to jog my memory... Oh, it's just nonsense."

My typical response to such people is: I absolutely do teach on Fridays, just not in a classroom. I teach my student research assistants while we do research together and build presentations and papers together, which is way more effective teaching at the advanced level than classroom lectures.

To prepare my children for adult life, I impose a candy tax on their Halloween candy. It's a win-win situation! I get to eat some of their candy and they get to prepare to be tax-paying adults. (I seriously do this.)

This is a fun practical if you haven’t seen it before: youtu.be/gZK9A_waYP4 #SciTeachUK #ITeachPhysics

Well that was a first. I regularly receive science questions and comments from my readers that I enjoy answering and responding to. For the first time, an intelligent-sounding person emailed me telling me that my article was wrong because he asked Gemini and Gemini gave a different answer. 🤦

Yes! And I'll add that professors also exist to motivate and inspire students. Articles like this assume that all college students are 100% self-motivated to do all the hard work required to learn. A professor's genuine passion for a subject is contagious in a way that a computer never can be.

Wenn ChatGPT das Mathe-Paper schreibt 😅 (eigentlich gar kein lustiges Paper: www.ams.org/journals/not...)

Hours of daylight throughout the year in the Northern Hemisphere.

The entirety of Roy Kerr's paper solving the Einstein equations for a spinning black hole in case anyone wondered what it looks like. In terms of impact per line this paper has to be one of the all-time top 10.

I just got word that the faculty senate of my university is now dissolved and no longer exists, in compliance with the new Texas law. Yikes! I thought my university would somehow find a loophole, or an exemption, but it's official. It's a sad day for shared governance and checks and balances.

Along these lines, I googled my own book to find reviews and the AI Overview caught my eye. Literally half of what it said about my book is not in my book at all. If it gets it so wrong for a subject I know about, it's probably getting it so wrong on everything!

Oh cool! A video that I helped make is now up to 10 million views.

I made this chart for a class I teach, showing various electric charge setups. I think that it shows the pattern well. An arbitrary charge configuration can be represented as a combination of these, and often the higher-order multipoles can be approximated to be zero. #ITeachPhysics