Jeremy Bassis
banner
bassisjeremy.bsky.social
Jeremy Bassis
@bassisjeremy.bsky.social
University of Michigan Glaciologist interested in climate change, ice sheets, sea level rise and equitable adaptation and mitigation | he/him/his |
I don't want AI. I don't want AI doing my writing. I don't even want AI to help me code. But if I'm going to have AI shoved down my throat, at the very least can we get a copy-paste function that doesn't immediately and always reformat a Word document to look like a ransom note?
The Math on AI Agents Doesn’t Add Up
A research paper suggests AI agents are mathematically doomed to fail. The industry doesn’t agree.
www.wired.com
January 27, 2026 at 12:44 PM
Historical quirks mean that the quixotic location of our state capitals primarily serve as trick questions on geography tests.

Australians are like hold my beer, we built our capital city, Canberra, from scratch so that it is maximally inconvenient for everyone AND is vulnerable to wildfires.
January 21, 2026 at 2:15 PM
Reposted by Jeremy Bassis
The level of racism and transphobia and shear callousness involved in people outside the US looking at what the US government is doing to people of color and trans folks here, some of whom are US citizens, and saying “these people are privileged” is intense
January 21, 2026 at 1:18 PM
I would like to propose a small corner of academia be kept AI free. A handful of journals could vow to forbid AI in any form by authors, editors and reviewers. Let's go back to old fashioned mail correspondences.
January 21, 2026 at 1:00 PM
I still think that Roger’s theory must be right under some circumstances, but we haven’t been able to identify those yet. So, in the tradition of Nye’s correspondence, we decided to write a short comment so that the community can tell us where we misunderstood Roger’s theory. 11/
January 20, 2026 at 4:48 PM
As you may have gleaned from the title of our paper and to our surprise, we found that our numerical simulations of closely tracked Nye’s theory for crevasse depths and not Roger’s. 10/
January 20, 2026 at 4:48 PM
My experience is that both Roger and Nye are usually right so we decided to do some numerical experiments to see if we could identify conditions when Nye is right and when Roger was right. 9/
January 20, 2026 at 4:48 PM
This is intriguing because Nye’s estimate tends to predict crevasses that are too shallow in many cases. Roger’s calculation seems initially intuitive, but relies on some approximations that aren't entirely easy to follow. In other words, it is possible that Nye and Roger are both right. 8/
January 20, 2026 at 4:48 PM
Recently, Roger Buck, entered the fray and challenged Nye’s theory. Roger argued that Nye’s theory for closely spaced crevasses doesn’t account for the finite thickness of ice. When you include the finite thickness of ice, crevasses can penetrate much deeper in some circumstances. 7/
January 20, 2026 at 4:48 PM
Nonetheless, Nye's theory has been widely applied, including by myself, because it is simple and easy to use. 6/
January 20, 2026 at 4:48 PM
Nye recognized the limits of his theory and was clear that it should apply best when crevasses are closely spaced. There is, however, little evidence that Nye’s theory actually agree well with observations of crevasse depths. 5/
January 20, 2026 at 4:48 PM
Almost immediately after publishing his estimate, Nye proposed a new estimate based on the emerging laboratory data that ice undergoes power-law creep deformation. I find this published dialogue somewhat quaint in today's research environment. 4/
January 20, 2026 at 4:48 PM
Building on earlier work by Hopkins, Nye used a new theory called “plasticity” that had emerged to describe the deformation of metals. Using this theory Nye computed the patterns that crevasse would take and estimated how deep crevasses would penetrate. 3/
January 20, 2026 at 4:48 PM
Understanding crevasse patterns had been a key very early question in glaciology at a time when there was vigorous debate about whether ice flows like a fluid or breaks like a solid. 2/
January 20, 2026 at 4:48 PM
Very technical, but fun preprint up for comments. The idea behind this dates back to a paper and series of correspondences written by John Nye, one of the founders of quantitative glaciology in the 1950s that has recently been revisited by the geophysicist Roger Buck. 1/
Brief communication: Nye was right!
Abstract. Despite decades of study, predicting crevasses penetration depths remains controversial. Nye provided one of the earliest estimates of crevasse penetration depths. Recently, a new theory, ca...
egusphere.copernicus.org
January 20, 2026 at 4:48 PM
Europe’s temptation is to seek a diplomatic solution to Trump’s threat to invade Greenland. This is the wrong approach because it treats a lack of impulse control as a diplomatic problem. Y’all need to assemble a team of preschool and kindergarten teachers and let them have at it.
January 19, 2026 at 6:43 PM
Frederick Douglas gave a speech 150 years ago today in Ypsilanti about civil rights, masked kidnapping and violent repression of free speech.
January 16, 2026 at 11:41 PM
This is an example of when this app can be truly wonderful with an expert chiming in with a deep and delightful dive about the history of Mark Twain quotes in response to my inability to track down a source for a pithy statement attributed to Mark Twain about climate.
in conclusion, this is the kind of quote we'd be suspicious of at the museum simply because the language is really straightforward. it sounds like the simplified version of him that we're familiar with, but not necessarily like his actual writing. and yet sometimes he wrote pithy things so who knows
I'm teaching a 200 person intro climate course and the textbook has a nice quote attributed to Mark Twain "Climate is what you expect; weather is what you get".

The problem is that, as far as I can tell, Mark Twain never said those words.
January 15, 2026 at 12:41 PM
I'm using it in class as an example of (a) everybody makes mistakes, but also (b) information without a specific citation should raise your spidey sense that things might not be kosher.
January 15, 2026 at 1:21 AM
The thing is nobody who attributes this quote to Twain gives a source so I'm pretty convinced that this is one of those things where it was misattributed to Twain because it sounds like something Mark Twain would say and then everyone went with it because is founds like something Twain would say.
January 15, 2026 at 1:21 AM
At this point, I have gone through the common sources like books, essays, letters and can't find any evidence but . . . 👇I can't really rule out some obscure offhand comment to a newspaper.
Yeah everything gets attributed to Mark Twain and it’s impossible to check because he did interviews with every podunk newspaper as he traveled across the country. Essentially unverifiable if there’s no cited source. I’m having the same problem with a George Carlin line right now.
January 15, 2026 at 1:21 AM
I'm teaching a 200 person intro climate course and the textbook has a nice quote attributed to Mark Twain "Climate is what you expect; weather is what you get".

The problem is that, as far as I can tell, Mark Twain never said those words.
January 14, 2026 at 10:44 PM
GIS and project management are high on my list. I do a lot of community engaged design in my capstone courses so I include evidence based best practices for community engagement.
January 12, 2026 at 8:58 PM
Anyone asserting that professors are indoctrinating students needs to teach a 200+ person intro class. How am I going to find time to indoctrinate students when I spend all day responding to emails with variations like "As stated in class and in the syllabus . . ."
January 12, 2026 at 8:28 PM
The United States already sends and maintains a military force to Antarctica. Look, I don't think Trump even knows Antarctica exists. But returning to disputed territorial claims is exactly aligned with the dipshits in this administration and we are one stray Mar-a-Lago conversation away from that.
January 10, 2026 at 12:37 AM