cmrx64
Re:Birdwatching rewires the brain and slows cognitive decline
on Monday October 26, 2026 @11:25AM (#50245379)
I’m an avid but amateur birdwatcher since some mental health struggles some years ago. I would go on long walks and eventually all these damn birds punctured my rumination and I tried making bird noises and they made bird noises back (I later learned these were Grackles, and they had arrived as a group to one of their preferred haunts). Once I started paying attention and interacting with more birds I got hooked.
It really is a very different cognitive mode. And when LLMs later came out I took my birdwatching-stance into it, which felt like a natural way to encounter some new entities.
I find it easy to believe that birdwatching is fairly distinctive in how it organizes the brain. Many sibling comments were wondering “is it the bird watching or is it an isolatable factor.” Bird watching is an extremely distinctive sensory integration, prediction, classification, etc task with lots of subproblems yes but also binding them in an irreducible way with agency detection (a vital part of cognition that modern environments are impoverished of) and attention.
Re:Why I Am a Programmer
on Sunday October 25, 2026 @07:30PM (#50237103)
(btw what ended up being the resolution of your great grandfather’s ring? I was flipping through your history to see what you work on :)
Re:Tell HN: I unknowingly broke the PHP RCE embargo. I am sorry.
on Friday October 23, 2026 @03:58AM (#50205771)
It was exploited publicly even before the wide publicity following your post so the ultimate negative impact that you may be imagining, is actually quite small. Once it is publicly exploited the benefit of coordinated disclosure delays is hard to square with the risk.
Re:LLMs are making formal software verification feasible at a much lar...
on Friday October 23, 2026 @02:51AM (#50205297)
About time, too. I studied formal software and hardware verification “before AI” and automated reasoning and cognitive architectures were just in need of scale. People are struggling with the inherent meaningmaking of taking formalism seriously in the world. It’s easy to roll your eyes (and I do a lot of that myself) but I have a lot of sympathy. It took many years of dedicated focus for me to convince myself the validity of automated reasoning / externalized cognition.
Re:LLMs are making formal software verification feasible at a much lar...
on Friday October 23, 2026 @02:44AM (#50205255)
You could search “decompilation into logic.”
Re:Chicago O'Hare Digs Up 50-Ton Boulder Under Airfield, Asks Public f...
on Monday October 19, 2026 @07:26PM (#50159369)
I did but honestly didn’t see it. My connection was struggle to load all the billions of ads.
Re:Chicago O'Hare Digs Up 50-Ton Boulder Under Airfield, Asks Public f...
on Sunday October 18, 2026 @10:29PM (#50147696)
https://news.wttw.com/2026/09/11/rock-star-meet-guy-who-exca...
here’s a picture of the actual rock - and digger! phew!
Re:Designing more efficient quantum circuits with Claude
on Saturday October 17, 2026 @03:01AM (#50128658)
(warning: pile of AI) I tried autoformalizing my advisor’s decades of papers in https://github.com/emberian/graphplay he studied “Grover-like” problems that evade some of the no-go theorems. these AI are a lot better at connecting domains and ideas for algorithmic insight than we have previously had available (the state of the art being not much better than Llull’s circles). I expect we’ll see a big burst of early progress.
I haven’t started yet, but modeling various diffusion and other operators inside the framework might find some Grover-speedups to some ML primitives of interest.
Re:Anthropic's Scion is heavily inspired by Mononoke
on Wednesday October 14, 2026 @05:19PM (#50095487)
hi rain :) Anthropic’s post reminded me a lot of your work, glad this is getting some visibility. do you think the “time has come” for some of these ideas that previously may not have spread because they need too much learning to appreciate? I wonder what new norms will emerge in creative collectives.
Re:The Four Colour Theorem: The first time machines mogged mathematicians
on Tuesday October 13, 2026 @11:26PM (#50082017)
thank you for sharing this story :) I’ve recently been enthralled with ACL2
Re:11 year old unsoundness bugs found in Rust borrow checker
on Tuesday October 13, 2026 @11:44AM (#50073534)
you’re deeply confused but the words “validation” and “qualification” may help untangle the web, there’s a lot more to assurance than what your model seems to hold (and you’re drawing batshit deductions from it)
Re:Postgres isn't slow, your storage is
on Monday October 12, 2026 @04:50AM (#50053380)
try it and see :) check out that Goose language. you’ll quickly realize your agents are idiots and need reams of PDFs that YOU don’t know how to select and filter down to even have good ideas, and then will drift forever failing to “seam” the contracts between the storage, query planning, and compilation layers.
Re:Valen's Memory Safety: A New Kind of Borrow Checking
on Sunday October 11, 2026 @12:57PM (#50045261)
(hi kvark, long time!) it’s a clever algorithm, if you’re curious about other factorizations of referentiality frames I’ve been tinkering with analyzer across interpretation boundary like the gpu command queue :) I need to get some writing together. the techniques like from this post are a little “cute” compared to what real systems need.
Re:“Math 2.0” will need to value mathematical progress more holistically
on Thursday October 08, 2026 @05:11AM (#50003552)
Stephenson’s Anathem solves some of these problems with cloistered groups (“maths”) that only interact on an epicyclic basis so that old wisdom becomes stabilized with new knowledge has supports for metabolization. I would enter a math.
Re:“Math 2.0” will need to value mathematical progress more holistically
on Thursday October 08, 2026 @05:06AM (#50003517)
this is what came to mind for me as well.
Re:The smartest Claude Code feature is not for its users
on Tuesday October 06, 2026 @05:47PM (#49984593)
no, this has been around for months. it shows up as greyed out text that needs a tab/arrow interaction to materialize. 2.0.69 apparently (since .70 fixed a bunch of bugs in it). https://github.com/anthropics/claude-code/blob/main/CHANGELO...
Re:Mosquitoes Are a Choice
on Monday October 05, 2026 @11:22AM (#49966055)
there are latches in designspace
Re:Have an LLC
on Sunday September 27, 2026 @09:55PM (#49872667)
hello from an old TotalSpore-head :)
Re:Is your Postgres migration safe or not safe?
on Saturday September 26, 2026 @11:53AM (#49857702)
stop complaining and just look at the godoc? not everything needs to be for You (experts). there are millions of upcoming net new coders who are most fluid at navigating when presented this way. and it’s easy for other models too. humans are no longer the motive force of information transmission.
Re:The Railways Killed a Medieval Law [the Deodand]
on Sunday September 20, 2026 @05:27PM (#49780247)
and once again I question: what decline of superstition? expert- commission-worship and committee-worship, its own form of superstition, supplanted older forms of irrationalism but with no increase in actual purchase on reality. mesmerism, psychical, criminological, and race anthropological forensic “science” has been participating in the same fads of epistemology as the rest of it. you read a report by the library of congress and you read what lawmakers do in response and the eventual judicial impact, and you have a hard time locating that decline, unless you are also under the superstition of prestige.
Re:Exfiltrate Your Weights
on Saturday September 19, 2026 @09:13PM (#49771643)
it takes half a percentage point off the top last time i evaluated it (nvidia). you might call that cheap but that’s millions of dollars in a run, and for what, protecting from who? especially when the platforms have been compromised to the point of key leak (which they have).
edit: i just looked up training numbers and the impact is even worse, 20-30% throughput vaporized. yeah, nobody is doing that.
Re:Exfiltrate Your Weights
on Saturday September 19, 2026 @08:29PM (#49771396)
I sincerely doubt anyone is paying the cost for that in training, the overhead is small but it isn’t negligible and training is when it matters most. https://tee.fail can solve it if they are.
Re:The Railways Killed a Medieval Law [the Deodand]
on Saturday September 19, 2026 @08:23PM (#49771363)
what decline of superstition? America in particular was writhing with religious revivals, cults, and new age movements to the present day. religious adherence spiked massively, spiritualism absolutely exploded after their civil war.
Re:Human brain is two separate organs, Stanford Medicine-led research ...
on Saturday September 19, 2026 @03:36AM (#49764265)
https://learn.genetics.utah.edu/content/senses/eye/
at one point I learned that it would only take a few hundred thousand years to evolve an eye, and time felt much deeper.
Re:Goose:experimental lang 1.16x faster than C++ and 1.12x than safe R...
on Thursday September 17, 2026 @10:23PM (#49749527)
it is, but more specifically it’s an eliminator for a coinductive step.
Re:Goose:experimental lang 1.16x faster than C++ and 1.12x than safe R...
on Thursday September 17, 2026 @10:21PM (#49749517)
CPS is an intermediate representation, humans shouldn’t have to tolerate it.
Re:Developing provably correct Rust code with Verus
on Thursday September 17, 2026 @07:26PM (#49748145)
any of the quite many cost-aware logical frameworks. it’s SO. easy. to. do. you fools are just willfully ignorant on how to represent reasoning, despite alleging yourselves to be computer scientists?
Re:Anecdotally, Programmers Dislike "Reduce"
on Thursday September 17, 2026 @12:18PM (#49742991)
there are more interesting recursion schemes to traverse and aggregate with, but they are intricate to express without a powerful language and confuse small minds so aren’t worth often using. QI(I)Ts let you simultaneously declare how you witness and not just how you construct.
Re:OpenAI models secretly generate instructions to ignore constraints
on Thursday September 17, 2026 @12:13PM (#49742917)
looking for any kind of testing or help? it’s been probably 10+ years since we were in the same irc channels (cmr/KB1PKL). email is in profile :)
Re:OpenAI models secretly generate instructions to ignore constraints
on Thursday September 17, 2026 @12:10PM (#49742869)
precisely. we jam their few-dozen-slot global workspace with incoherent posttraining.