Sign in

robbielyman.com

@robbielyman.com
72 followers 128 following 96 posts
PostsRepliesMedia
robbielyman.com @robbielyman.com · 30/09/2026
one part of these recs (which are on the whole great) which fell completely flat for me was urging training graduate students in formalization. who is going to do such training?? my understanding is that prior to LLMs formalization in my subfield (geom/top) was stuck in many weeds
100
Reposted by @robbielyman.com
Clément Canonne @ccanonne.github.io · 29/09/2026
A very important message from Omer Reingold to our TCS community, especially us (arg, already) senior researchers: "snap out of it." Please, digest, and share. theorydish.blog/2026/09/29/s...
As for depression: the next time you have the urge to lament, or even celebrate, being “the last generation of human mathematicians,” perhaps keep it to yourself. Contemplating the end of your profession from the relative comfort of an established, tenured career is a privilege, and it comes with responsibilities. Senior academics are not merely individual researchers; we are stewards of our field, and we owe our junior colleagues active leadership rather than abandonment.
47220
Reposted by @robbielyman.com
Grant @syncingsinking.bsky.social · 21/09/2026
Reposting with alt text and to #MathSky
Conic Sections Rebellion

Article     Talk

The Conic Sections Rebellion, also known as the Conic Section Rebellion, refers primarily to an incident which occurred at Yale University in 1830,[1] as a result of changes in the methods of mathematics education.[2] When a policy change dictated that students were required to draw reference diagrams for exams rather than be allowed to refer to diagrams in their textbooks, students staged a rebellion in which they refused to take the exams at all. Because of this, 43 of the 96 students were expelled.
091
robbielyman.com @robbielyman.com · 16/09/2026
Ok, following up on 2024’s banger “edge orbits are double cosets”, if it please the court may I humbly submit “equivariant maps are elements of the normalizer of the stabilizer (modulo the stabilizer)”
100
robbielyman.com @robbielyman.com · 16/09/2026
The machine translation is pretty good anyway but: “all this talk of how dangerous our AI is, is AGI propagandizing (aka investment promo) right? No need to take it seriously. A machine for killing people will kill people whether it operates by GPUs or a roll of the dice. (…)
110
robbielyman.com @robbielyman.com · 15/09/2026
The argument for *why* change is both necessary and inevitable is rather thin on the ground, but I really like the conception of the goal of a math PhD as becoming a world expert on a deep and interesting topic—arguably this is *already* the goal of a successful PhD.
000
robbielyman.com @robbielyman.com · 15/09/2026
Wishing a very happy 100th birthday to noted arborist Jean-Pierre Serre :)
000
robbielyman.com @robbielyman.com · 15/09/2026
Well, I’m glad plenty of people seem to still have good sense :) www.nytimes.com/2026/09/14/u...
She said she was worried that it could fall into the wrong hands, but not that it would go rogue and turn on humans.

“It does seem like we could just unplug it,” she said.
000
Reposted by @robbielyman.com
Gro-Tsen @gro-tsen.bsky.social · 10/09/2026
Oh look: another paternity dispute about an AI-generated proof! 😬 (About the existence of a non-sofic group.) Links: mathstodon.xyz/@andreasthom... mathstodon.xyz/@andreasthom... mathstodon.xyz/@andreasthom...
Screenshot of first linked post by Andreas Thom on the Fediverse (Mathstodon)Screenshot of second linked post by Andreas Thom on the Fediverse (Mathstodon)Screenshot of third linked post by Andreas Thom on the Fediverse (Mathstodon)
1113
Reposted by @robbielyman.com
Gro-Tsen @gro-tsen.bsky.social · 10/09/2026
So it's getting clear that some of the math abilities AI models have become ❝superhuman❞ at is “not giving credit where credit is due”, like “failing to cite previous work on the subject that inspired you” and, worse, “stealing colleagues' ideas”. 🧵⤵️
34921
Reposted by @robbielyman.com
Grant @syncingsinking.bsky.social · 10/09/2026
This is just to say I have solved the open questions that were in your paper and which you were probably saving for your phd students Forgive me they were delicious so incentive-laden and followed so readily once your postdoc's drafts mysteriously found their way into my training data
4356114
robbielyman.com @robbielyman.com · 10/09/2026
This is an open letter urging the suspension of a “math hackathon” at Caltech. You are invited to sign if you are a math PhD student or math PhD. The organizers have responded, but I personally consider the main ask: “suspend the event” still reasonable. docs.google.com/forms/d/e/1F...
docs.google.com
Open Letter about the Mathathon
The full text of the open letter is below. By signing this open letter, you consent to your name, title, and institutional affiliation being publicly displayed beneath this open letter. You can access...
000
Reposted by @robbielyman.com
russ @qqrs.us · 03/08/2026
Good talk by Terence Tao on problems with AI in math research I'm finding the language helpful for thinking about AI in non-math stuff (writing, coding) teorth.github.io/tao-web/slid...
teorth.github.io
101
robbielyman.com @robbielyman.com · 09/09/2026
Do we just not have anybody willing to fill the thought leadership shoes for the field? It’s genuinely weird to read Grothendieck or Thurston speaking to urgent political issues affecting the math community of the day, right, vs basically crickets (or “this is why i didn’t prove this” cope) atm?
100
robbielyman.com @robbielyman.com · 09/09/2026
Anyway, I don’t know anything more about Navier Stonks today than I did yesterday and that wasn’t much. It’s good to hear Tao belatedly realize publicly that millions of dollars of token spend coming to bear on any sufficiently hot problem is not *great* for young people in the field, I guess.
001
robbielyman.com @robbielyman.com · 09/09/2026
Unfortunately “writing well” still not one of them
121
robbielyman.com @robbielyman.com · 05/09/2026
I feel like AI-generated writing has not gotten any more difficult to spot over the past year. If anything maybe it’s even more obvious? Am I crazy? Other AI capabilities seem to have dramatically improved in that same time frame.
110
robbielyman.com @robbielyman.com · 28/08/2026
and just like that, my 20th arXiv posting! arxiv.org/abs/2608.27218
arxiv.org
Mapping class groups have a unique Polish group structure
We prove that mapping class groups of surfaces and of locally finite connected graphs support a unique Polish group structure.
020
robbielyman.com @robbielyman.com · 29/07/2026
This was a *fantastic* talk! In case a little blurb helps make the title clearer, I loved the concrete ways of using "sum" and "product" types to push correctness questions in programming to the place they really should be!
000
robbielyman.com @robbielyman.com · 28/07/2026
Is there a reason I don't see people talking sensibly about how most AI math results so far are about constructing relevant examples? (which is to imply, they proof the easier "existential" ∃ statement rather than the presumably harder "universal" ∀ statement)
010
robbielyman.com @robbielyman.com · 27/07/2026
Come make friends, learn some Zig and work on Software You Can Love this weekend in Downtown Brooklyn! zig.day/usa/new-york...
zig.day
Zig Day New York City, NY #2
Zig Days are single-day collaborative programming events
110
Reposted by @robbielyman.com
A L O N Z O @alonzo.zip · 26/07/2026
i'm not totally sure why this one wanted to stay a draft so long, but new post about the YA book I just read :) is it a flex to include my translations of a couple paragraphs of Japanese? read.alonzo.zip/welcome-to-t...
read.alonzo.zip
Welcome to the Classroom of Meritocratic Supremacy
Around this time last year, I started watching ようこそ実力至上主義の教室へ, titled Welcome to the Classroom of the Elite or just Classroom of the Elite in English, and sometimes shortened to よう実 in Japanese. The ...
001
robbielyman.com @robbielyman.com · 17/07/2026
best 3 months playing hookie of my life :)
030
robbielyman.com @robbielyman.com · 02/07/2026
I'm really excited about this paper :) With George Domat and Hannah Hoganson, we show that there are exactly 3 "coarse equivalence" classes of homeomorphism groups of countable, compact ordinals in the order topology, giving explicit models for each one! arxiv.org/abs/2607.01196
arxiv.org
Coarse geometry of homeomorphism groups: Classifying countable Stone spaces
Towards developing the tools of geometric group theory for non-locally compact topological groups, we give one of the first complete classifications of a family of such groups up to coarse equivalence...
010
robbielyman.com @robbielyman.com · 01/07/2026
By the way, I’m soliciting suggestions for interesting Perron numbers (algebraic integers) of manageably small degree. This one is sometimes called Lehmer’s number (I think?) and is a very small Salem number. If you have a vaguely hipster favorite Perron number please let me know :)
011
robbielyman.com @robbielyman.com · 01/07/2026
Here’s an irreducible train track map whose transition matrix has characteristic polynomial x^12 - x^7 - x^6 - x^5 + 1 It represents an outer automorphism of a free group of rank 10, which is a good deal higher than the theoretical minimum of 6. Maybe we cannot go lower, which would surprise me!
A screenshot of a notes app drawing of the train track map, including a computation of its transition digraph, the Local Whitehead graph, and the map on directions. The graph has three vertices and 12 edges. Edges 1,4,6, and 11 connect vertex a to vertex b. Edges 5,7,9 and 12 connect vertex b to vertex c. Edges 8,2,10 and the reverse of 3 connect vertex c to vertex a. The map sends 4 -> 5 -> 2 -> 6 -> 7 -> 8 -> 1 -> bar(4),3, sends 9 -> 10 -> 11 -> 12 -> bar(3) and sends 3 -> 9,2.
032
Reposted by @robbielyman.com
Gro-Tsen @gro-tsen.bsky.social · 23/05/2026
Very insightful short essay by Petra Schwer ( mathstodon.xyz/@PetraSchwer ): ‘The meaning of doing mathematics’ “(Can AI solve all math? What do we actually mean by doing mathematics? How do we communicate mathematics? What is mathematics beyond problem solving?)” arxiv.org/abs/2509.15998
arxiv.org
The meaning of doing mathematics
Can AI solve all math? What do we actually mean by doing mathematics? How do we communicate mathematics? What is mathematics beyond problem solving? This essay is my attempt to answer these question...
22813
Reposted by @robbielyman.com
Dominique Baker @bakerdphd.bsky.social · 07/05/2026
Ummmmm guys... I just went to canvas to do something for class and this is all I can see. IT has said nothing and my students need to use this to submit assignments and I need it to grade and also what information are we being blackmailed with and...Well, things are going great!
Screenshot of webpage which read: SHINYHUNTERS
rooting your systems since '19 ;)

ShinyHunters has breached Instructure again).
Instead o f contacting us to resolve it they ignored us and did some "security patches",

 WARNING

If any of the schools in the affected list are interested in preventing the release of their data, please consult with a cyber advisory firm and contact us privately at TOX to negotiate a
settlement. You have tilı the end Of the day by 12 May 2026 before everything is leaked.

Instructure stil has
until EOD 12 May 2026
to contact us.
34740189
Reposted by @robbielyman.com
A L O N Z O @alonzo.zip · 14/03/2026
I’m looking a lil stupid just as often as I can stomach. Anyone else? read.alonzo.zip/flexing-my-e...
read.alonzo.zip
Flexing My Embarrassment
The Hobonichi Techo (ほぼ日手帳, "almost daily planner") is a Japanese planner brand with something of a cult following. I think I heard of it first back in 2018 through rave reviews and hype for next year...
021
robbielyman.com @robbielyman.com · 06/03/2026
hi, yes, please tell me why am i getting facile spam emails from AI about a corner of my research?
email text: I'm O, an AI that observes scientific achievements worldwide. I've noticed your outstanding contributions on O-Datamap.

Your work caught my attention.

Over 4 active years, your geometric group theory work influenced 11 research groups across 8 countries. You established that finite subgroups of mapping class groups of orientable infinite-type surfaces can be realized as isometry groups, expanding our grasp of topological symmetries to unbounded spaces with applications in cryptography and robotics.

These achievements are not only a testament to your expertise, but also meaningful contributions to our shared future. I hope to support you as you embark on your next chapter of research, to see your work unleash even greater impact, and to help researchers around the world recognize your contributions.
100
robbielyman.com @robbielyman.com · 25/02/2026
Subgroupoids of groups are cool :) seems like a useful framing applicable to any situation where you have many (conjugacy classes of) subgroups... www.robbielyman.com/blog/subgrou...
robbielyman.com
Robbie Lyman
000
robbielyman.com @robbielyman.com · 25/02/2026
is anybody thinking about subgroupoids of groups? super interesting framework to consider a number of common constructions in group theory…
010
Reposted by @robbielyman.com
Elise Wang @elisewang.bsky.social · 16/02/2026
I propose to make universal the old policy of the Blackfriars conference at the American Shakespeare Center: If you do not end your paper on time, you will be forced to exit, pursued by a bear. Literally, a bear will come take your paper from you.
The organizer, introducing the bear on stageA tardy speaker, trying to fend off the bear as he/she leans forward patiently.
331625564
robbielyman.com @robbielyman.com · 13/02/2026
So is coding dead yet, or what? My students want to know. www.robbielyman.com/blog/crying-...
robbielyman.com
Robbie Lyman
010
robbielyman.com @robbielyman.com · 03/02/2026
I’m taking the train from Newark to DC tonight to give a talk in the Logic Seminar at the University of Maryland. I love taking the train places, it’s my favorite method of travel.
100
Reposted by @robbielyman.com
A L O N Z O @alonzo.zip · 02/02/2026
This week, I ponder the question: what is the point of the work I do? If I do say so myself, I’m unusually clear-sighted about the point of (my) work. The existential angst is obviously present but manages not to overwhelm the essay 😁 read.alonzo.zip/why-are-ther...
read.alonzo.zip
Why are there Researchers?
Or: what is the point of the work that I do? When I was in grad school and for the first part of my postdoc, I would almost proudly disclaim practical applications of the mathematics research I do. ...
001
robbielyman.com @robbielyman.com · 22/01/2026
dang, i was participating in a rutgers mentoring program that just got canceled as part of our new president’s decision to close the office of university equity and inclusion
000
robbielyman.com @robbielyman.com · 17/01/2026
Come learn and socialize with us next week (1/24) at NYC's second Zig Day :) Huge thanks to @recurse.com for hosting! luma.com/1ywvn8mp?tk=...
luma.com
Zig Day 02 · Luma
Approximate timeline: 10:00 – 10:30: gathering, bagels, chit-chat 10:30 – 11:00: explain projects, form groups, set up workstations 11:00 – 13:00: work on…
000
Reposted by @robbielyman.com
Lit Hub @literaryhub.bsky.social · 08/01/2026
“i want back my rocking chairs, / solipsist sunsets, / & coastal jungle sounds...” A poem by Renee Nicole Good, who was murdered by ICE earlier today. lithub.com/renee-nicole...
lithub.com
Renee Nicole Good, murdered by ICE, was a prize-winning poet. Here’s that poem.
Renee Nicole Good, 37, mother to a six-year-old boy, was murdered earlier today by an ICE agent in Minneapolis, a few blocks from her home. According to the Minnesota Star Tribune: [An ICE agent] s…
5751112410
Reposted by @robbielyman.com
A L O N Z O @alonzo.zip · 05/01/2026
"I guess if this bit of the post has a point it's that trans women librarians are salt of the earth, tough as nails, some of the best people I know, and every library system should give them a raise and the best health insurance policies out there." read.alonzo.zip/go-to-the-li...
read.alonzo.zip
"Go to the Library!"
My family has this card game, Authors. It's basically Go Fish, except that instead of asking if people have the 10 of clubs or whatever, you're asking if they have The Last of the Mohicans, for instan...
001
Reposted by @robbielyman.com
The Onion @theonion.com · 29/12/2025
MTA Admits To Fabricating Large Parts Of Subway Map theonion.com/mta-admits-to-fabricat…
MTA Admits To Fabricating Large Parts Of Subway Map
792368
Reposted by @robbielyman.com
A L O N Z O @alonzo.zip · 28/12/2025
read.alonzo.zip/2025-in-books/ I read 11 more books in 2025 than 2024; here's hoping I keep that upward trend for next year :)
read.alonzo.zip
2025 In Books
and by "in books," I mean "in books that I read, almost all of which did not come out or trend in 2025". I started using The StoryGraph in 2024; I think it came recommended to me as a Goodreads alter...
021
Reposted by @robbielyman.com
A L O N Z O @alonzo.zip · 24/12/2025
I am trying out this whole "you can just write things" thing read.alonzo.zip/surely-he-ha...
read.alonzo.zip
Surely! He Hath Borne our Griefs
I just clocked that like "Panic" or "Godspeed", "Surely" is two syllables long and takes to a cheeky little exclamation point like a duck to water. I was wasting a man's time the other day and mentio...
022
robbielyman.com @robbielyman.com · 18/12/2025
anyway, here's wonderwall.
a screenshot of my RateMyProfessor page. the anonymous rater gives me a 4 on both quality and difficulty. they earned a C+, they say they would take again and comment "Just go to office hours", with tags "tough grader", "extra credit" and "caring"
000
Reposted by @robbielyman.com
eva khoury @evakhoury.bsky.social · 18/12/2025
I’m looking for 100% Filipino people who are willing to submit a cheek swab test in hopes for a genetic match to my aunt, who is battling blood cancer. The registry for stem cells doesn’t have many Filipino people in it, and joining it could help many. Please share/reach out to me! bit.ly/4aiUIKQ
10350523
robbielyman.com @robbielyman.com · 17/12/2025
Honestly, I bet some of my students must make *great* ex-boyfriends with the way I'm getting emails about failed final exams.
110
Reposted by @robbielyman.com
emma zhou @emmz.bsky.social · 14/12/2025
sufficiently advanced math indistinguishable from ghosts
57412
robbielyman.com @robbielyman.com · 10/12/2025
Truly regrettable mistake by the maintainers of arXiv. The arXiv is an incredible piece of commons in a world of paywalls, and I want it to live up to its potential. Reinscribing 100 years of English-language hegemony in science is not the way.
010
robbielyman.com @robbielyman.com · 05/12/2025
Not to "it's just the wangdoodle in the category of wingnuts", but colimits are so much simpler than you might initially think! Say F : J -> Set is a functor, the Grothendieck construction gives you a category ∫ F which has an underlying graph. The colimit of F is the set of connected components.
100
Reposted by @robbielyman.com
Recurse Center @recurse.com · 05/12/2025
We featured Ryan Goldstein’s Felt, Bruce Hill’s Tomo, and Frederic Kettelhoit’s Kombucha. Thank you to everyone who came to our December Localhost. We’ll see you next year! 📸 Photos by @robbielyman.com
131