Linkage

Gray on gray cats (my cats but not my photo of them; \(\mathbb{M}\)).

A brief roundup of Chegg “video killed the radio star” news (\(\mathbb{M}\)):

• The Wall Street Journal, “How ChatGPT Brought Down an Online Education Giant: Chegg’s stock is down 99%, and students looking for homework help are defecting to ChatGPT” (archived; via; see also)

• Gizmodo, “Chegg Is On Its Last Legs After ChatGPT Sent Its Stock Down 99%” (via)

• SF Chronicle, “Bay Area tech company Chegg lays off hundreds more, CEO blames Google and AI”

There’s much more of the same out there. An interesting comment from the first YC link, that may partially explain Chegg’s failure to respond to this threat to its business: “No one internally at Chegg admits that it’s cheating software”.

Foams made of felt (\(\mathbb{M}\), via). Tanya Khovanova experiments with a physical model to check the following claim: if you have a foam (a complex of cells meeting three at an edge and four at a vertex as soap foams do) whose cell walls can be assigned three colors (all different along each edge), then removing the walls of any one color produces a topological manifold, a surface in which every point has a neighborhood that looks like a disk.

Kathryn Cooper uses a 19th-century photographic technique to capture the complex movements of a starling murmurations (\(\mathbb{M}\)).

Terry Tao on target audiences: aim your research papers at yourself from when you were first learning about the topic, and take care to explain the basic points that you found confusing then.

Grid of overlapping circles on the exterior of the Library of Birmingham, England. For more shots of the same subject, see its Wikimedia category.

Dan Piponi made a heatmap of the evolution of a sparse random field under the Game of Life, producing an interestingly organic-looking texture.

For-profit academic publishers love LLM garbage (\(\mathbb{M}\)), Kevin Munger on Crooked Timber. Munger argues that under current open-access pay-to-publish standards, they make a profit on each LLM-produced paper, are motivated primarily by profit, and have little short-term incentive to make their journal content meaningful to subscribers. Therefore, if we want to continue to be able to find meaningful scholarly content, we need to look elsewhere than these publishers. Munger advocates diamond open-access instead, cutting out the profit motive.

St Bernard on collaboration (\(\mathbb{M}\)). Peter Cameron finds a theological “answer to the annoying bureaucrats who ask what percentage of the work on a publication was done by each of its authors”. Not sure this will go over better than my standard boilerplate “this is a collaborative field in which the authors contributed equally and are listed alphabetically” (repeated separately after each publication listed), but it’s tempting to try.

Fake-journal publishers start trying to hijack the big commercial publishers (\(\mathbb{M}\)). These include look-alike web sites for journals from Elsevier, Springer, and Nature, with the same journal title and plausible but fake domain names (e.g. springer.uk.com instead of springer.com, sciencedirects.com instead of sciencedirect.com), and their own series of dois. Obviously legal action is impending but I guess the hope is they can take the money and run first. Anyway, as the link concludes: Be aware!

Why average class size means different things when averaged by class or by student, and why schools advertise them by class when what you probably want is by student.

Richard Elwes debuts on Numberphile (\(\mathbb{M}\)) with a video on Goodstein numbers and the link between quickly-growing sequences and unprovability.

Embedding any knot in a Menger sponge (\(\mathbb{M}\), arXiv). New research from three Canadian high-schoolers and their faculty mentor.

Can you see through a cannonball packing (\(\mathbb{M}\))? Answer: yes, along lines parallel to the grid lines of its embedded square grids.

The deterioration of Google (\(\mathbb{M}\), via). We’ve all experienced the proliferation of bad search results from Google and seen stories pointing it out and blaming the internal conflict between search and ads at Google. This piece has a different theory: they switched to a machine learning system for ranking search results and now don’t have any idea how to control what it does. However, Jeremy Kun finds this theory unlikely.