Dave Bacon
@dabacon.tachyon.institute
Theoretical ski bum. Former pseudo professor. Quantum bridge builder. Will math for food.
mastodon: @dabacon@tachyon.channel
https://dabacon.org
mastodon: @dabacon@tachyon.channel
https://dabacon.org
Pinned
Dave Bacon
@dabacon.tachyon.institute
· Apr 25
Much Madness is divinest Sense -
To a discerning Eye -
Much Sense - the starkest Madness -
’Tis the Majority
In this, as all, prevail -
Assent - and you are sane -
Demur - you’re straightway dangerous -
And handled with a Chain -
—Emily Dickinson
To a discerning Eye -
Much Sense - the starkest Madness -
’Tis the Majority
In this, as all, prevail -
Assent - and you are sane -
Demur - you’re straightway dangerous -
And handled with a Chain -
—Emily Dickinson
Reposted by Dave Bacon
Do you want to help develop new theories for systems that adapt, evolve, and interact?
Applications are open for SFI’s 2026 UCR program, a fully funded, 10-week experience for undergraduates, offering mentorship, seminars, and project development.
Apply by Jan 14, 2026
www.santafe.edu/ucr
Applications are open for SFI’s 2026 UCR program, a fully funded, 10-week experience for undergraduates, offering mentorship, seminars, and project development.
Apply by Jan 14, 2026
www.santafe.edu/ucr
December 25, 2025 at 6:02 PM
Do you want to help develop new theories for systems that adapt, evolve, and interact?
Applications are open for SFI’s 2026 UCR program, a fully funded, 10-week experience for undergraduates, offering mentorship, seminars, and project development.
Apply by Jan 14, 2026
www.santafe.edu/ucr
Applications are open for SFI’s 2026 UCR program, a fully funded, 10-week experience for undergraduates, offering mentorship, seminars, and project development.
Apply by Jan 14, 2026
www.santafe.edu/ucr
A pretty good haul.
December 25, 2025 at 11:34 PM
A pretty good haul.
Will recite this one before my next fencing bout.
I know my heart, I have gained power over my heart, I have gained power over my two hands and arms, I have gained power over my feet, I have gained the power to do what pleaseth my Ka. (BDB)
December 25, 2025 at 3:08 PM
Will recite this one before my next fencing bout.
This is a cool result, but to me represents why the Turing model is actually the wrong physics model of computing. This is because it is serial, and so cannot deal with errors. I’ve written about this here dabacon.org/pontiff/2010...
December 23, 2025 at 3:57 PM
This is a cool result, but to me represents why the Turing model is actually the wrong physics model of computing. This is because it is serial, and so cannot deal with errors. I’ve written about this here dabacon.org/pontiff/2010...
Reposted by Dave Bacon
My colleague Sanjay Ghemawat & I have done a fair bit of performance tuning of various pieces of code. We wrote an internal Performance Hints document ~2 years ago as a way of identifying some general principles & we've recently published a version of it externally.
Doc: abseil.io/fast/hints.h...
Doc: abseil.io/fast/hints.h...
December 19, 2025 at 10:25 PM
My colleague Sanjay Ghemawat & I have done a fair bit of performance tuning of various pieces of code. We wrote an internal Performance Hints document ~2 years ago as a way of identifying some general principles & we've recently published a version of it externally.
Doc: abseil.io/fast/hints.h...
Doc: abseil.io/fast/hints.h...
Reposted by Dave Bacon
Scientists should read this.
artificialbureaucracy.substack.com/p/context-wi...
artificialbureaucracy.substack.com/p/context-wi...
Context Widows
or, of GPUs, LPUs, and Goal Displacement
artificialbureaucracy.substack.com
December 16, 2025 at 6:09 AM
Scientists should read this.
artificialbureaucracy.substack.com/p/context-wi...
artificialbureaucracy.substack.com/p/context-wi...
And it’s actually even crazier than that (the epidemic stuff is the beginning). Wheelers four no’s include: self reference instead of hierarchy of laws, no laws, no continuum, no spacetime!
Seems to be a losing battle to make people understand that Wheeler's "It from Bit" does not mean "reality is made of information" (it was about defending an epistemic view of quantum mechanics, where the wave function isn't real, only measurement outcomes are), and likewise...
December 16, 2025 at 12:10 AM
And it’s actually even crazier than that (the epidemic stuff is the beginning). Wheelers four no’s include: self reference instead of hierarchy of laws, no laws, no continuum, no spacetime!
Reposted by Dave Bacon
Applications open for SFI’s 2026 UCR program.
Are you fascinated by transdisciplinary questions that transcend or combine disciplines? UCR is a fully funded, 10-week summer research experience for undergraduates curious about complexity science.
Apply by Jan 14, 2026
santafe.edu/ucr
Are you fascinated by transdisciplinary questions that transcend or combine disciplines? UCR is a fully funded, 10-week summer research experience for undergraduates curious about complexity science.
Apply by Jan 14, 2026
santafe.edu/ucr
December 11, 2025 at 10:04 PM
Applications open for SFI’s 2026 UCR program.
Are you fascinated by transdisciplinary questions that transcend or combine disciplines? UCR is a fully funded, 10-week summer research experience for undergraduates curious about complexity science.
Apply by Jan 14, 2026
santafe.edu/ucr
Are you fascinated by transdisciplinary questions that transcend or combine disciplines? UCR is a fully funded, 10-week summer research experience for undergraduates curious about complexity science.
Apply by Jan 14, 2026
santafe.edu/ucr
Reposted by Dave Bacon
OpenAI leadership are promoting a paper in Physics Letters B where GPT-5 proposed the main idea — possibly the first peer-reviewed paper where an LLM generated the core contribution. One small problem: GPT-5's idea tests the wrong thing. My technical comment: scirate.com/arxiv/2512.0... 1/
![As part of a broader effort to demonstrate AI’s po-
tential in scientific research [1], OpenAI executives have
pointed to a recently published paper by Hsu [2] as evi-
dence that AI can contribute original ideas to physics [3,
4]. Hsu credits GPT-5 with proposing the core idea of the
paper de novo, possibly the first published physics article
where the main idea came from an LLM, and discusses
the methodology in a companion piece [5]. We examine
whether GPT-5’s criterion is correct.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:uknpj5mh4cfutreceecdbv7x/bafkreia3jerjfcqdui45n4upaqci5u5r6st7o7pqmdzahhjtidtostpooq@jpeg)
December 9, 2025 at 5:17 PM
OpenAI leadership are promoting a paper in Physics Letters B where GPT-5 proposed the main idea — possibly the first peer-reviewed paper where an LLM generated the core contribution. One small problem: GPT-5's idea tests the wrong thing. My technical comment: scirate.com/arxiv/2512.0... 1/
Reposted by Dave Bacon
Great visit to Tufts a few weeks ago. They have an open-rank job in Quantum Information / Computing theory. Open Positions | Department of Physics and Astronomy share.google/NosuSUs77OZU...
Open Positions | Department of Physics and Astronomy
Below are job opportunities within the Department of Physics and Astronomy. For additional employment opportunities at Tufts University, please visit Careers at Tufts.Full-time Positions:
share.google
December 9, 2025 at 1:37 AM
Great visit to Tufts a few weeks ago. They have an open-rank job in Quantum Information / Computing theory. Open Positions | Department of Physics and Astronomy share.google/NosuSUs77OZU...
Reposted by Dave Bacon
A claim of indirect detection of dark matter via gamma rays from the galactic center. I would not get too excited until independent confirmation comes along; many people have looked for exactly this signal. It’s too easy to make gamma rays from more conventional sources.
This just in: a NASA telescope may have just "seen" #darkmatter for the first time! By exploring the Milky Way's galactic halo and using the latest data, a team of researchers detected high-energy gamma rays extending towards the center of the Milky Way. Find out more here: https://ow.ly/OJe550XE40B
December 7, 2025 at 12:35 PM
A claim of indirect detection of dark matter via gamma rays from the galactic center. I would not get too excited until independent confirmation comes along; many people have looked for exactly this signal. It’s too easy to make gamma rays from more conventional sources.
Don’t tempt me langjamgamejam.com
Langjam Gamejam
A 7-day challenge to create a programming language and then use it to build a game.
langjamgamejam.com
December 1, 2025 at 8:35 PM
Don’t tempt me langjamgamejam.com
Preprocessing direct in QEC microcode
May I guide the hearts of the gods, and may they protect me, may I be mighty among those who suspend themselves on high. (BDF)
November 26, 2025 at 3:24 PM
Preprocessing direct in QEC microcode
Bullshit is sort of the opposite of a problem in coNP. Instead of hard to find but efficient to verify it is false, it is easy to find and hard to verify it is false.
Who was it who said it takes 10x more effort to refute bullshit than to create it?
November 26, 2025 at 3:22 PM
Bullshit is sort of the opposite of a problem in coNP. Instead of hard to find but efficient to verify it is false, it is easy to find and hard to verify it is false.
Thoth is short for The Toric Code in Ancient Egyptian.
Thoth is the protection of all my flesh. (BDQ)
November 26, 2025 at 3:17 PM
Thoth is short for The Toric Code in Ancient Egyptian.
Ancient Egyptian resource theory with Clifford operations
The sky quivers, the earth quakes before me, for I am a magician, I possess magic. I have come that I may glorify Orion, that I may set Osiris at the head, that I may set the gods upon their thrones. (PTF)
November 25, 2025 at 5:49 PM
Ancient Egyptian resource theory with Clifford operations
Reposted by Dave Bacon
A few quick notes:
1) Magic - not my favorite word choice - comes from magic angle directions and into quantum computing by arxiv.org/abs/quant-ph... . It is not about entanglement.
1) Magic - not my favorite word choice - comes from magic angle directions and into quantum computing by arxiv.org/abs/quant-ph... . It is not about entanglement.
The Large Hadron Collider has a new job. It’s being used to study how information flows through quantum systems — ideas that lie at the core of quantum computing. @shalmawegs.bsky.social reports: www.quantamagazine.org/particle-phy...
Particle Physicists Detect ‘Magic’ at the Large Hadron Collider | Quanta Magazine
The supercollider is now being used to explore quantum phenomena, including a “magic” form of quantum entanglement.
www.quantamagazine.org
November 25, 2025 at 4:26 PM
A few quick notes:
1) Magic - not my favorite word choice - comes from magic angle directions and into quantum computing by arxiv.org/abs/quant-ph... . It is not about entanglement.
1) Magic - not my favorite word choice - comes from magic angle directions and into quantum computing by arxiv.org/abs/quant-ph... . It is not about entanglement.
If I believed in the simulation hypothesis (see en.wikipedia.org/wiki/Simulat... for my opinion that somehow made it to Wikipedia) I’d say this is obviously how magic works arxiv.org/abs/2511.15304
Adversarial Poetry as a Universal Single-Turn Jailbreak Mechanism in Large Language Models
We present evidence that adversarial poetry functions as a universal single-turn jailbreak technique for Large Language Models (LLMs). Across 25 frontier proprietary and open-weight models, curated po...
arxiv.org
November 24, 2025 at 3:58 PM
If I believed in the simulation hypothesis (see en.wikipedia.org/wiki/Simulat... for my opinion that somehow made it to Wikipedia) I’d say this is obviously how magic works arxiv.org/abs/2511.15304
The riverlane report is excellent, thanks to the authors for correcting a minor minor minor bug :).
We said that things were moving fast in #QEC in The #QuantumErrorCorrection Report 2025 and that progress was happening from every angle.
So, thanks to @dabacon.tachyon.institute for feeding back on the report (www.riverlane.com/quantum-erro...) and pointing out that...
So, thanks to @dabacon.tachyon.institute for feeding back on the report (www.riverlane.com/quantum-erro...) and pointing out that...
November 24, 2025 at 3:36 PM
The riverlane report is excellent, thanks to the authors for correcting a minor minor minor bug :).
Reposted by Dave Bacon
Chris Umans, Siki Wang
A number-theoretic conjecture implying faster algorithms for polynomial factorization and integer factorization
https://arxiv.org/abs/2511.10851
A number-theoretic conjecture implying faster algorithms for polynomial factorization and integer factorization
https://arxiv.org/abs/2511.10851
November 17, 2025 at 6:56 AM
Chris Umans, Siki Wang
A number-theoretic conjecture implying faster algorithms for polynomial factorization and integer factorization
https://arxiv.org/abs/2511.10851
A number-theoretic conjecture implying faster algorithms for polynomial factorization and integer factorization
https://arxiv.org/abs/2511.10851
Ancient Egyptian Causality
I am Yesterday, I know Tomorrow. (BDB)
November 17, 2025 at 4:26 PM
Ancient Egyptian Causality
Reposted by Dave Bacon
Apply now for SFI’s 2026 CSSS Journalism Fellowship (June 9–July 8). Each year, two accomplished journalists join our summer school program for a deep dive into complexity science. Stipend, housing, and travel covered.
Applications open until Feb. 4, 2026.
Apply: www.santafe.edu/news-center/...
Applications open until Feb. 4, 2026.
Apply: www.santafe.edu/news-center/...
November 14, 2025 at 8:11 PM
Apply now for SFI’s 2026 CSSS Journalism Fellowship (June 9–July 8). Each year, two accomplished journalists join our summer school program for a deep dive into complexity science. Stipend, housing, and travel covered.
Applications open until Feb. 4, 2026.
Apply: www.santafe.edu/news-center/...
Applications open until Feb. 4, 2026.
Apply: www.santafe.edu/news-center/...
Reposted by Dave Bacon
We taught a quantum computer to learn from its own mistakes-- without stopping.
Our RL framework repurposes QEC detection events as learning signals to stabilize the system.
Result: Improved Logical Error Rates for both Surface and Color codes on Willow!
see arxiv: arxiv.org/abs/2511.08493
Our RL framework repurposes QEC detection events as learning signals to stabilize the system.
Result: Improved Logical Error Rates for both Surface and Color codes on Willow!
see arxiv: arxiv.org/abs/2511.08493
November 12, 2025 at 4:44 AM
We taught a quantum computer to learn from its own mistakes-- without stopping.
Our RL framework repurposes QEC detection events as learning signals to stabilize the system.
Result: Improved Logical Error Rates for both Surface and Color codes on Willow!
see arxiv: arxiv.org/abs/2511.08493
Our RL framework repurposes QEC detection events as learning signals to stabilize the system.
Result: Improved Logical Error Rates for both Surface and Color codes on Willow!
see arxiv: arxiv.org/abs/2511.08493
Reposted by Dave Bacon
The list of accepted papers for #QIP2026 is now online at qip2026.lu.lv/programme/ac...
Accepted papers
qip2026.lu.lv
November 11, 2025 at 1:28 PM
The list of accepted papers for #QIP2026 is now online at qip2026.lu.lv/programme/ac...