iiinova — where your digital self keeps evolving

Jon Ullman @thejonullman.bsky.social · 11/07/2026

OK, asking ChatGPT how to use a jackhammer without damaging the human flesh inside the concrete. I'll be right over.

010

Jon Ullman @thejonullman.bsky.social · 11/07/2026

Knowing you, I assume that this means the metaphor involves you somehow getting your hands stuck in cement?

120

Jon Ullman @thejonullman.bsky.social · 10/06/2026

Another lesson is don't pile braindead arguments on top of compelling ones. People, like me, will just pick at the silliest things you say.

010

Jon Ullman @thejonullman.bsky.social · 09/06/2026

I used to like this coffee shop that was open 4 hours a week, 4 days a week, and had a huge store with almost no seating. I didn't know I was supposed to feel guilty about accepting that I was buying coffee from a company with no viable business model.

130

Jon Ullman @thejonullman.bsky.social · 25/05/2026

I watched the same show and was similarly losing my mind at the fact that clearly no one in that writers' room knew what tenure is.

010

Jon Ullman @thejonullman.bsky.social · 19/05/2026

Yeah that's true, although there is some actual fun in traveling to an in person conference so I wouldn't say (b) dominates (a) in general but it does for the scope of this comparison.

110

Jon Ullman @thejonullman.bsky.social · 19/05/2026

At least if it's an in-person home conf, then its probably an away conf for most attendees so its more attractive to treat it that way for myself. For a virtual conf there is the collective action problem that I need a critical mass of other people to make it worth it.

110

Jon Ullman @thejonullman.bsky.social · 19/05/2026

It's great if everyone does that but it's genuinely hard. An "away conf" gives me two choices "stay home and parent/teach" or "travel and have my wife/colleages do it." A "home conf" adds "go but take a break to take kids to the dentist." That third option is hard to say no to when available.

210

Jon Ullman @thejonullman.bsky.social · 13/05/2026

Is LLM slop defined precisely enough for there to be incontrovertible evidence of it?

250

Jon Ullman @thejonullman.bsky.social · 19/02/2026

Discussions about a new research center sounds a lot more like 2 than 1 to me.

010

Jon Ullman @thejonullman.bsky.social · 12/02/2026

I'm sorry, I'm not familiar with this "nuance" thing you speak of.

010

Reposted by Jon Ullman

Clément Canonne @ccanonne.github.io · 21/12/2025

The list of accepted papers at the Algorithmic Learning Theory Conference, a.k.a. #ALT2026, is out! h/t @thejonullman.bsky.social (PC chair). algorithmiclearningtheory.org/alt2026/acce... "ALT: topics so hot, it has to be held in Canada in February"

algorithmiclearningtheory.org

Accepted Papers | ALT 2026

0215

Reposted by Jon Ullman

Clément Canonne @ccanonne.github.io · 20/12/2025

"It's the first drug in quite a while that has really promising signs to help actually reverse the signs of MS, whereas a lot of the drugs out there at the moment just help reduce your symptoms, not necessarily reverse it." auscricket.com.au/news/jemma-b...

061

Jon Ullman @thejonullman.bsky.social · 16/12/2025

Why would you ever do this? The goal of having a conference should be to bring as much of the community as possible together. Why put it in a city known for prohibitive costs, narcissism, and having an overwhelming number of other things to do?

010

Jon Ullman @thejonullman.bsky.social · 18/11/2025

jk, I have the most imposter syndrome.

110

Jon Ullman @thejonullman.bsky.social · 18/11/2025

Yeah, I'm way beyond imposter syndrome though. Imposter synrome was like soooo 1992.

110

Jon Ullman @thejonullman.bsky.social · 18/11/2025

My wife has a "most cake" t-shirt.

110

Jon Ullman @thejonullman.bsky.social · 13/11/2025

Lol, I have been frustrated by this so many times. Also no love for monotonic.

010

Jon Ullman @thejonullman.bsky.social · 06/11/2025

Huge congratulations to my student Konstantina Bairaktari for defending her thesis on Monday!🍾🥳🎊 I'm kinda jealous that Konstantina is headed to Aarhus to do a postdoc with the amazing @kasperglarsen.bsky.social. I can't wait to see what she does next!

Konstantina, her committee, some flowers, and a small decorative gourd.

090

Jon Ullman @thejonullman.bsky.social · 05/11/2025

It's an amazing contribution of theory for sure, but I'm jealous of the systems guys who invented "turn it off and turn it on again."

091

Reposted by Jon Ullman

Gautam Kamath @gautamkamath.com · 20/10/2025

Announcing (w @adamsmith.xyz @thejonullman.bsky.social) the 2025 edition of the Foundations of Responsible Computing Job Market Profiles! Check out 40 job market candidates in mathematical research in computation and society writ large! Link: drive.google.com/file/d/1zvsr...

02112

Jon Ullman @thejonullman.bsky.social · 09/10/2025

I once read an actual crypto paper that said "we design a probably secure scheme."

040

Jon Ullman @thejonullman.bsky.social · 15/09/2025

📢Adam Smith, @gautamkamath.com, and I are putting together a list of job market candidates in Foundations of Responsible Computing! Last year's list was a great success so we're keeping it going! If you want to be included, or nominate someone, see link in the replies!

2147

Jon Ullman @thejonullman.bsky.social · 04/09/2025

🚨 I am co-chairing ALT 2026 this year with Matus Telgarsky. The submission server is open so please submit your best work! Deadline: Oct 2, 2025 AoE Confernece: Feb 23-26, 2026 in Toronto! Website: algorithmiclearningtheory.org/alt2026/

algorithmiclearningtheory.org

ALT 2026 | ALT 2026 Homepage

The 37th International Conference on Algorithmic Learning Theory

0123

Jon Ullman @thejonullman.bsky.social · 26/03/2025

In their defense, I don't remember what my contribution was either 🤷

0181

Jon Ullman @thejonullman.bsky.social · 14/03/2025

I see that you also respond to tweets either immediately or 2mo later.

100

Reposted by Jon Ullman

Clément Canonne @ccanonne.github.io · 09/01/2025

Another thought-provoking* talk by @thejonullman.bsky.social at the #EnCOREPrivacy25 workshop, on auditing differentially private algorithms by designing membership inference attacks(and why we should, and how). encoredp.github.io *The thoughts! They were provoked!

A slide titled "auditing differentially private algorithms"

0122

Reposted by Jon Ullman

Gautam Kamath @gautamkamath.com · 24/12/2024

With @adamsmith.xyz and @thejonullman.bsky.social, we have compiled a set of profiles of 29 people in the "foundations of responsible computing" community ("mathematical research in computation and society writ large") who are on the faculty job market. Link: drive.google.com/file/d/1Hyvg... 1/3

23916

Jon Ullman @thejonullman.bsky.social · 24/12/2024

This effort was inspired by a similar list for the EconCS community maintained by @jasonhartline.bsky.social and @gkatzelis.bsky.social

010

Jon Ullman @thejonullman.bsky.social · 24/12/2024

Adam Smith, @gautamkamath.com, and I compiled this list of FORC Job Market Profiles---faculty candidates who work in the foundations of computing and society, broadly defined. There are 29 awesome candidates for you to check out! drive.google.com/file/d/1Hyvg...

drive.google.com

FORC Job Market Profiles 2024.pdf

1166

Reposted by Jon Ullman

Gergely Neu @neu-rips.bsky.social · 17/12/2024

i recently gave a talk about questions like this at a workshop organized by the amazing "learning theory alliance" (let-all.com), and several students reached out to tell me how much my talk resonated with them. link to the slides here: docs.google.com/presentation...

docs.google.com

let-all-gergo

Research styles: “fast” or “slow”? Some thoughts by Gergely Neu

1519

Jon Ullman @thejonullman.bsky.social · 17/12/2024

I received a splashy email announcing our new dept website www.khoury.northeastern.edu Perfect microcosm of NEU: very slick, but doesn't accurately list members of each research area, even though faculty said this was a top priority. This is what an administrator-driven university looks like.

khoury.northeastern.edu

Home - Khoury College of Computer Sciences

Explore cutting-edge computer science programs at Khoury College. Empower your future with innovative education and research opportunities.

030

Reposted by Jon Ullman

Ola Svensson @algosvensson.bsky.social · 17/12/2024

Postdoc positions @ EPFL (soft deadline Jan 20) available in vibrant theory group that covers complexity, quantum, algorithms, theory of ML ... and includes faculty E. Abbe, A. Chiesa, F. Eisenbrand, M. Göös, M. Kapralov, O. Svensson, and last but not least T. Vidick. go.epfl.ch/theory-postdoc

go.epfl.ch

Postdoctoral position in the Theory Group

1193

Reposted by Jon Ullman

Rasmus Pagh @rasmuspagh.net · 13/12/2024

Interested in a post-doc position in theoretical computer science? Want to find out why they call it "Wonderful Copenhagen"? Look here! Application deadline January 10. Hiring in multiple areas including geometry, complexity, learning, and privacy. jobportal.ku.dk/videnskabeli...

jobportal.ku.dk

Postdoctoral positions in Theoretical Computer Science

1166

Jon Ullman @thejonullman.bsky.social · 13/12/2024

I didn't know they call it that, but it really is great!

110

Reposted by Jon Ullman

Mahdi Haghifam @mahdihaghifam.bsky.social · 11/12/2024

I’ll be at #NeurIPS2024 this week! Looking forward to presenting my joint work with Thomas Steinke(@stein.ke) and Jon Ullman(@thejonullman.bsky.social) NeurIPS page with video: neurips.cc/virtual/2024... Link to arxiv: arxiv.org/abs/2406.07407

0103

Jon Ullman @thejonullman.bsky.social · 10/12/2024

Is there a more pointless criticism than "this paper doesn't have a conclusion section" to put in a review?

230

Jon Ullman @thejonullman.bsky.social · 06/12/2024

I spent a really long time trying to figure out how this was a pun before I read on.

010

Jon Ullman @thejonullman.bsky.social · 05/12/2024

Show me the rule that says dogs can't drive cars.

120

Jon Ullman @thejonullman.bsky.social · 05/12/2024

My personal experience is that I spent a year outside of academia before my PhD precisely so that I could experience the boundaries of a 40hr week and not be preoccupied with work I could be doing. I was just as preoccupied, so I figured I should at least be preoccupied with something of value.

040

Jon Ullman @thejonullman.bsky.social · 05/12/2024

Sadly it turns out none of my nonacademic friends have 40 hour work weeks either.

040

Jon Ullman @thejonullman.bsky.social · 03/12/2024

Totally agree. If you hate pageflipping, Google scholar pdf reader is life changing

530

Jon Ullman @thejonullman.bsky.social · 03/12/2024

I'm not dogmatic about it and I agree that it depends a lot of I am reading a paper as a relative expert vs reading it as an outsider.

000

Jon Ullman @thejonullman.bsky.social · 03/12/2024

Personally I prefer [xyz00] style, but I totally agree that I hate when camera ready formats just ruin papers. This is why the arxiv version is always the "real" paper imo.

270