mkprc

dupe

https://news.ycombinator.com/item?id=50207248#50208018

It's going to be tough to assess the efficacy of these schools, mostly because we can only wait-and-see how these kids turn out. But also, how do you correct for the fact that many (most?) of these students have savvy, affluent, involve parents and would probably be doing comparatively as well in any other school?

Selecting trig, "A right triangle has tanθ= ¾. If the side adjacent to θ is 30, what is the length of the opposite side?" The language/grammar is imprecise. What's it mean for a triangle to "have" tanθ= ¾? Should be that the triangle has an angle θ such that has tanθ= ¾. Also, "If the side adjacent to θ is 30 …" should refer to the "leg" adjacent θ; the hypotenuse is also a side adjacent to θ.

Taking the diagnostic, I just got the question "What is the opposite of -2?" Is it +2? Is it -½?

Edit: I restarted, selecting just Abstract Algebra as my diagnostic topic, and was shocked to get this question again! In Abstract Algebra! So far none of the questions in Abstract Algebra have anything to do with Abstract Algebra …

Link to Terry Tao's announcement[1] of this and the corresponding HN comment thread[2].

[1]: https://terrytao.wordpress.com/2026/10/16/sair-open-math-os-...

[2]: https://news.ycombinator.com/item?id=50121980

Radio Garden is the OG. I've been seeing quite a few vibe-born clones of Radio Garden recently. E.g. https://airdrift.stream too

I wonder who would install this onto their calculator in earnest. Like, who’s the segment of folks with the savvy to learn about this and install this on their TI calculator, but who aren’t savvy enough to just learn to do the computations in curricular calculus class?

Snap judgment, but it appears to be a vibe coded interface added to an existing CAS for a TI-84 Plus CE calculator.

> Systems like OpenAI’s GPT-5 are accelerating research in …

Lesson plan is already behind the times.

This article features a few students, kids, who seem cognizant of what AI is doing to their thinking. They're aware of the temptation of cognitive surrender and the risk of cognitive. I worry about the kids who have no idea what their reliance on AI as a crutch has done to their ability to think, like I worry about the kids who have no idea how "scrolling" has nerfed their attention. (There's a big overlap between these two groups too). Like, some kids have no frame of reference for "normal" or "average" cognitive performance since so many of their peers are affected too, and pretty often too so are their parents.

> True to its name, OpenAI’s new group will mostly serve in an advisory capacity, assessing the significance of new results and coordinating their release. While members won’t be paid, they can offer unsolicited advice, go public with their views, and control their own membership, giving them a measure of independence from the company. However, the group won’t be given leeway to slow down or redirect OpenAI’s ongoing mathematical research.

Incredible

Here' a link to the article this one is responding to: https://archive.ph/lPtBr

The original article is excellent, and this one appears to be urging folks to stop having such basic reactions to the original. I think the appeal of the original is that the daughter is now 16(?) and has opinions on her parents' decisions in her schooling. She makes good points, but she's also a teenager and the grass is always greener … nuance.

This is an interesting post.

Teachers are often asked "why do I have to learn [this] when my calculator can do it?" or "why should I memorize [this] when I can just look it up online?" and now "why should I learn to reason through [this] if an LLM can just do it?". To respond to these questions some math teachers think quite hard about the purpose of learning mathematics (some just have a canned response). The consensus answer is something like: the topics themselves don't matter too much, but they serve as an instrument of becoming better at thinking logically, and becoming quantitatively savvy. Most students are not savvy. The author of this post appears to be an exception, probably becoming savvy as a child from all the unstructured activities they did while being home-schooled.

> What did I do with my time instead? … I took apart old computers and pieced them together to try and build something strong enough to emulate Super Smash Bros on the PC, and voraciously read vintage young adult fiction like Tom Swift or G.A. Henty novels.

> … if I ask my calculator the square root of 231, I also have no easy way to evaluate the output. I can sort of guess (10×10 is 100, 20×20 is 400, 15×15 is somewhere in the middle, so if the answer is 15-ish it's probably right), but I can also sort of guess at LLM outputs.

The standard mathematics education machine has no idea what to do with a student who is already savvy, capable of critically interpreting what the calculator/internet/LLM reports to them.

For the author, and for any other savvy folks, LLMs are a great new tool. I worry about everyone else who because they are not savvy, when they ask "why should I learn to reason through [this] if an LLM can just do it?" are quite happy to cognitively surrender to the tool, deferring their critical thought to the LLM, making it a crutch they now happily rely on.

Baader–Meinhof phenomenon:

Baudelaire popped up in this article two days ago.

https://www.noemamag.com/a-new-kind-of-creative-poverty/

This looks really cool at first, a great "use of the internet" for anyone with one of these old receivers. But I want to vibe check with HN.

Ambient notes: The site design is clearly AI-assisted. I couldn't find a human name associated with the site anywhere, just references to a "hobbyist", which is maybe fine if the creator values their anonymity. The Donate page requests crypto, no card payment. It's listed as free, but "After 90 days your receiver may show a 6-character code. Enter it here to check status" … why? What's the point of this? And for folks with a receiver that doesn't support a custom DNS, the landing page suggests they change their home network's DNS server to this one, which seems like a terrible idea.

Terry Tao is on the board of scientific advisors and posted about this a couple days ago:

https://terrytao.wordpress.com/2026/08/22/mathematical-disco...

I'm not optimistic at the staying-power of this initiative. There've been a few things like it before, like a journal dedicated to asking authors to submit more informal "ELI5" versions of their popular papers for broader appeal. But the incentive structures in academia continue to favor only "original research". Just like expository papers aren't valued by hiring committees, grant reviewers, etc, submissions to this journal won't be valued.

Also posted here, with little discussion

https://news.ycombinator.com/item?id=49366811

Calculus by James Stewart, Daniel Clegg, and Saleem Watson

https://www.stewartcalculus.com

Previous discussion, still active

https://news.ycombinator.com/item?id=49260250

Here's the radio station website: https://www.sealionradio.org

Under the site's FAQ for “Is this a livestream?”

> Not yet, maybe someday. You’re listening to a set of field recordings I made in the spring of 2026. There are close to four hours of recordings, taken from various locations at Shilshole Marina over a few weeks. You could listen for quite a long time and never hear the same thing twice. I’ll keep adding recordings as conditions allow.

Previous discussion:

https://news.ycombinator.com/item?id=49101040

"Works" means "results in better average scores on exams". A fine thing, but it seems like every take I here on explicit instruction versus inquiry-based learning is so basic, these days from teachers who are rediscovering the effectiveness of just telling their students things explicitly after years of trying to shoehorn every lesson into an inquiry-based structure

For readers not working in academia, this essay and it's prequel provide a window into one aspect of the bureaucractic nightmare currently underlying US higher education.

I'm surprised company reputation, as a metric, doesn't come up more often in discussions of enshittification. This post is spot on, but the author only focuses on the negative externalities that profit-maximization imposes on the public. Every act of enshittification also impacts an organization's reputation. What if we were to make up some measurement for reputation and convince organizations to optimize for some combination of their reputation and profits?