HNHacker News
TopNewBestAskShowJobs

dingdongditchme

152 karma · joined September 12, 2025

submissionscomments
dingdongditchme··on Three new discoveries in mathematics, computer science, and physics
From my experience in research this was already the case 15 years ago (and from older peers, way before). Science has always has "in-groups". I forget the name of some philosopher in the 50's who was the opinion science is just a bunch of fads and it takes the old generation to die before a new idea can really be accepted.
dingdongditchme··on The Exponential Time Hypothesis Is False [pdf]
I believe this stems from an incomplete understanding of the problem space on my part (excuse the multiple puns); but, I am skeptical Knuth really said P=NP. There are so many mathemtical examples of equations that clearly show P≠NP. Computing in itself was originally formed from using iterative methods to solve hard problems. Least-squares methods, sin() cos(), traveling salesman, so many to list? To say P=NP for all problems seems like a fools errand to me.
dingdongditchme··on Ask HN: If the models are so good, why doesn't OpenAI just start 100 companies?
Leopold Aschenbrenner tried this and got eaten by the wolves as far as I can tell...
dingdongditchme··on Rabbits were a medieval royal luxury in England until the Black Death
I have tried it in a French restaurant and it was delicious. Just don't think about how cute it would have been to cuddle instead. :-(
dingdongditchme··on Rabbits were a medieval royal luxury in England until the Black Death
How is that an example of burying the lead? (I presume lede is the correct spelling, but burying lead for others to find in their drinking water amuses me.)

EDIT: saw the picture and yes. That certainly was catchy, but I the hn title is missing the subtitle: "until the black death made them commonplace".

dingdongditchme··on 90% of US working hours are spent on tasks not covered well by AI benchmarks
Or the productivity gains, albeit substantial, are not as earth shattering as the llm providers would have you believe. As with any new technology adoption is a laggard. The internet "existed" in the late 80's, early 90's. Many companies are still reaping productivity gains from plugging into this first.
dingdongditchme··on 4 Years of Ludicon
Very interesting. LLMs at the state today certainly do feel like a Cambrian explosion on software with more code is being written/generated than ever. Throughout my career it felt like solving one problem opened the door for many more. To infinity and beyond! With all the trepidation coming from the maths community (Terrence Tao in particular), I am optimistic that many previously overlooked opportunities in software will now be filled.
dingdongditchme··on Americans' social activity has steeply declined in the past 30 years
Being interested is also a lot more easy than to be interesting. The latter really depends on the eye of the beholder. Whereas I make it a game to find out what is interesting about the person across from me. Sometimes I strike gold, sometimes it's a dumpsterfire and we implicitly agree never to talk again. But once you strike gold and meet people you genuinely are really interested in, you get addicted.
dingdongditchme··on Vote on which of Hacker News' challenges for AI have been met
I like your list of "challenges" but have a personal issue with two of them:

1. "improve uniteds' plane shedule" -> the word "improve" does a lot of heavy lifting there..

2. Turing test for me is solved: "AI expert" is just moving the goal posts imho. The original turing test to my knowledge is about passing notes under a door. Obviously, llms can't communicate via hand-written notes and if that is the bar it will take a long time (or a specifically designed hand-writting machine) to really do this. As far as how much writing back and forth you can do it is clear that turing test is beat: I am wondering often enough on text sent to me from colleagues if it is generated, the same goes for comments here or any ol' website. In a standard llm session I don't really communicate differently than with a human and would not be able to tell the difference in an hour texting session or so. Of course if I ask it to count words or do something ridiculous I can find sus it out; but for all intents and purposes the chat-bot exists.

dingdongditchme··on Git 3.0's upcoming SHA-256 default will be a costly mistake
I completely agree with everything you say here. I felt the critique in the article was way more strawman falacy than everything else. I use(d) git much more locally, and would never run into this problem because I would only use sha-256 on new repos. Rewriting the entire history is also not so much of a pain depending on organisation size and how it is done (if an org decided to do it). It may be a pain for git-hosts, but even there i think reasonable solutions can be found. Still not sure if I care about the switch though sha-1 worked and the security implications seem minor to my use-cases. I am much more concerned with code leaking online...

sidenote: i hate the term "forge", is it too late to settle for something less... cringe??

dingdongditchme··on Agents.md speaks Unix, and you should too
lul what? As a linux fanboy who "speaks unix" (presumably, do I? not sure...). I find it humorous that we have gone full circle. The newest tech shit is recommending I learn almost 50 year old technology? Time for me to log off.
dingdongditchme··on I said no and Apple said yes
these days to me a device is bricked if I need to spend +1h recovering it. Happy linux user though through and through.
dingdongditchme··on Learning Programming in an Age of LLMs
certainly! There are pro's and con's to fire still today. Us fireless plebs prefer a life much closer to nature, eating our meat raw and bathing in the mud.
dingdongditchme··on Learning Programming in an Age of LLMs
I think bill gates summarized it pretty well in a recent letter [1]. There are pro's and con's to every new technology. Learning can be greatly accelerated with the use of llm's but you have to use them the right way. Just like calculators help further down the line, they do not help you when you are still trying to learn the basic concepts of arithmetic. I personally think I have found a way of working with llms that really accelerates getting stuff done while i am still able to learn. It means reading more, and (although I hate this in part) reading generated text. What is infuriating is when I suspect people writing to me with generated text, it is insulting and should be banned. Makes me want to spend more time offline (probably a good thing in my case).

[1]https://www.gatesnotes.com/home/home-page-topic/reader/a-tur...

dingdongditchme··on Pion, an agent designed to run any company autonomously
slop confirmed.
dingdongditchme··on Pion, an agent designed to run any company autonomously
I like the analogy but can't believe it is not generated slop. Just curious how deep the integration of agentic LLMs is in your social media presence?
dingdongditchme··on Reverse-Engineering Claude Web's MicroVM: Uncovering Anthropic's Hidden Antspace
Clearly the author is not a native english speaker (from the other blog posts) i found skimming it interesting nonetheless, i hate slop as much as the next guy but appreciate the language barrier braking qualities of llms.
dingdongditchme··on Reverse-Engineering Claude Web's MicroVM: Uncovering Anthropic's Hidden Antspace
sounds like a douche
dingdongditchme··on Reverse-Engineering Claude Web's MicroVM: Uncovering Anthropic's Hidden Antspace
I believe that what the previous commment is trying to say is that even when we identify phrases highly attributable to llm output we can never be certain. Which is the nefarious thing about broad use of computer generated text. I noticed comments here on hacker news containing conspicous mistakes which llms hardly ever do anymore and was wondering whether it was people purposefully adding mistakes to identify as noslop or whether llms got worse or if it was bots trying to fool me. My misspelling of conspicuous was left in on purpose in this case. All i can say is "trust me bro, I am not ai".
dingdongditchme··on What a time to be alive
oh we do! At least they google is quite good at adhering to robots.txt.
dingdongditchme··on OpenAI bots knew about the RubyGems caching vulnerability
Who the fuck is going to hold these AI companies responsible for running these gigantic semi-autonomous botnets on investors dime?
dingdongditchme··on Blizzard Revives 'StarCraft' Franchise with Open-World Sci-Fi Shooter
I haven't touched it in ages. But I always felt that it SC2 was more successful in taking out APM and "Deathball" mechanics. I remember a lot of cheese options and nuances. I guess there is just a big spread between balancing the pro players and the bronze league fuckers like myself. I have more nostalgia for SC:BW but only played online thanks to SC2 matchmaking. I think making an RTS that ticks all the boxes is incredibly difficult.
dingdongditchme··on Fable 5.1 Solves the Cyphral Distich, a 370-year-old cipher
Congrats on your map! I am leaning more towards the optimism side. Like you I've had some real breakthroughs myself. Personal akin your garden map, no where near solving millennium problems. My take is that all these (Fable solves this, OpenAI does that, AI will take over the world) Marketing Stunts will fade some day when the true cost of tokens hits the market. You and I may have to contend with lesser models, but will probably get work done just fine.
dingdongditchme··on Fable 5.1 Solves the Cyphral Distich, a 370-year-old cipher
You know, it could have been more than one person. At this point I don't think it really matters, unless he/she starts selling their coins.
dingdongditchme··on GUIs should be fully keyboard-driven
A game changer for me was the link hints extension on firefox and chrome: instantly enabled a lot of websites to be keyboard-driven.
dingdongditchme··on Just the rumour of a bug is enough to find an exploit these days
Michael Burry managed just fine, I believe. Secondly, I think the "in every case" is too strongly worded. I am working in an org right now where it has not happened yet. Perhaps it is only a matter of time, but let me have my optimism.
dingdongditchme··on Just the rumour of a bug is enough to find an exploit these days
My take is that when the bubble bursts good software will remain. Right now all the big players are hiding the cost. Anthropic (in my mind) is especially egregious at tokenmaxxing without telling you whats going on. A simple request to edit a text file launches multiple agents which takes ages, and burns tokens. All the while it calles it "Sautéeing" or whatever random verb they spew out as you wait for your answer.
dingdongditchme··on Just the rumour of a bug is enough to find an exploit these days
Thank you for the insight! Are you going back to the changelog and editing the CVE-PENDING? I think this would be a justifiable edit of the release history that I'm not sure is possible on github.
dingdongditchme··on Just the rumour of a bug is enough to find an exploit these days
The star only confuses me more...?
dingdongditchme··on Boot a Virtual iPhone via Apple's Virtualization.framework
Is the simulator faster than this? I would suspect so.
← PreviousPage 2 of 4Next →