Associate Professor, @IULuddy · Core Developer, @racketlang · Member, @TC39 · Handler, @gravemaker_ulti · @samth@types.pl
Bloomington, IN
Joined October 2008
- Tweets44.4K
- Following1.3K
- Followers4.6K
- Likes19.4K
People who don't vote I can understand; people who don't vote but answer polls and tell pollsters they voted I have no theory for.
Sam Tobin-Hochstadt retweeted
This keynote by Pat Hanrahan at #RacketCon is phenomenal! Einstein summation, Penrose diagrams, APL, J, K, Cajori's A History of Mathematical Notation, Bringhurst's The Elements of Typographic Style and more! 🤯 youtube.com/watch?v=pIoogidK…
Sam Tobin-Hochstadt retweeted
ok new week new topic. what’s everyone’s favorite joke ever made in a PL publication? I think mine is this one (from “The Myths of Object-Orientation” by James Noble)
I really want a @pangram browser plugin for the hotcrp paper bidding page.
I reached out to Michigan Senate candidates Abdul El-Sayed and Mike Rogers for their thoughts on the Detroit Red Wings.
Here's what they had to say:
octopusthrower.com/what-each…
See what you've done, @sliminality ?
"Rationalists are a bunch of toxic internet racists and it would have been better if effective altruism had not gotten involved with them" is also my view.
Ok, now help choose this week's topic:
- are papers too incremental nowadays?
- what's the meaning of memory safety?
- are full first class continuations necessary?
- how should monads be introduced to beginners?
@lexi_lambda @ShriramKMurthi @samth @krismicinski @disconcision attn: I just spoke on the phone with Alexis and we've decided try reviving PL Twitter, starting with this thread. How's everyone doing
Sam Tobin-Hochstadt retweeted
@lexi_lambda @ShriramKMurthi @samth @krismicinski @disconcision attn: I just spoke on the phone with Alexis and we've decided try reviving PL Twitter, starting with this thread. How's everyone doing
This is the most disturbing thing I will see on the Internet this week.
what i love about daniel kolitz's work about the extremes of digital content is that his pieces are so luxuriously written, they're an argument for the continued viability and necessity of writing in an increasingly post literate world nytimes.com/2026/10/06/magaz…
It's funny that the author of this piece has subsequently spent a lot is time persuading with fiction. lesswrong.com/s/pmHZDpak4NeR…
Sam Tobin-Hochstadt retweeted
Apropos of nothing... here is a quote from a blog post by @JulesJacobs5 in December 2023:
"It seems reasonable to assume that AI will be able to do mathematics better than humans, and that this will happen in the near future."
julesjacobs.com/2023/12/15/h…
That the labs haven't done this, even though it would obviously be incredibly valuable for them, but continue to steamroll math, TCS, programming, cyber security, etc has to be a bearish sign for progress outside these areas.
Ok, finished going through the OAI math list. It is bonkers and should be front page news around the world if folks understood what this meant. But if I am running strategy at a lab, #1 priority is "do this for medicine, oncology, battery efficiency, etc as fast as possible".
Sam Tobin-Hochstadt retweeted
As a Dad, I endorse working hard. – at Somerville, MA
This quoted post is unavailable.
I wrote this thread three weeks ago, when the arXiv preprint came out. But then someone told the POPL PC chair that I was violating the double-blind review rules, and I was asked to delete it.
Our paper has now been rejected, so I can talk about it again.
These rules are bad.
New paper! With my student Fred Fu and my colleague @carloangiuli, we fix the terrible problems in the proofs in my paper on the Typed Racket type system. arxiv.org/abs/2609.16299
New paper! With my student Fred Fu and my colleague @carloangiuli, we fix the terrible problems in the proofs in my paper on the Typed Racket type system. arxiv.org/abs/2609.16299
So, as additional lessons: semantic soundness proofs are not only nice to do, but they mechanize easily, and modern coding models can autoformalize an entire (straightforward) PL paper autonomously.