dr_analog
top 1% of underdog fetishists
No bio...
User ID: 583
Your comment convinced me @ffrreerree2 was right on the money to bring this up as a Culture War topic.
It's basically Epstein Island but for blue collar and/or immigrant types instead of the elite, the victim is older instead of underage and the videos of the rapes exist.
dr_analog
top 1% of underdog fetishists
2d ago
Thinking-style models I tried the same question on gave the 'wrong' answer, but had Thinking components specifically highlighting that the question was strange and must have involved unstated assumptions (either picking up materials from the car wash to do the work at home, or the car already being there).
Mine got hung up wondering why you would try to optimize 50 meters, it's too inconsequential a distance to matter for greenhouse gas emissions or exercise.
dr_analog
top 1% of underdog fetishists
2d ago
Congrats!
Stock some Pepto bismol. The sulfur burps can be brutal while you're coming to speed.
Lift and consume high protein.
My wife said I had bad BO during my major losing months. So I had to make an effort to shower more.
Good luck!
dr_analog
top 1% of underdog fetishists
2d ago
I linked this down-thread but there's notes about each of the convicted rapists in this case here: https://en.wikipedia.org/wiki/Pelicot_rape_case#Convicted
Do you see a pattern?
I'll admit some of these fit some stereotypes.
Nizar Hamida
Had eight prior convictions, including domestic violence and attempted abduction of his child with a former partner.[65] Said he went for a sexual encounter to celebrate the end of his bachelor days as his wife-to-be was arriving shortly from Tunisia.[64]
Mohamed Rafaa
Had prior conviction for raping his 17-year-old daughter, for which he was sentenced to five years' imprisonment. Raped Gisèle while the Pelicots were at their daughter's holiday home on the Île de Ré.[64]
Hassan Ouamou
Convicted in absentia having fled the country, travelling between Morocco and Romania claiming no intention of returning to France. Thirteen prior convictions in connection with theft, violence, drugs, and possession of weapons.
But to me it's not that solid a pattern. Plenty of these people have proper French sounding names and like they'd be familiar with Christianity.
The most solid theme for me is: losers and imbeciles with a splash of psychopathy.
dr_analog
top 1% of underdog fetishists
3d ago
Calling LLMs “wordcel technology” is backwards in 2026.
You can paste in a screenshot of a math problem that 99%+ of adults would fail, calculus, linear algebra, probability, geometry and it will solve it step by step, showing its work.
Not just arithmetic. Structured reasoning over formal systems. The same goes for logic puzzles, physics derivations, statistics problems.
They'll even teach it to you.
If your definition of wordcel now includes ‘solves multistep math from an image and explains it', then we're just not going to agree on the term.
I can't really imagine all of those men thought they were knowingly raping this guy's wife against her will. I have to believe he told them it's her kink to be drugged up and raped and then she wants to watch the videos later. Or something. This is a lame excuse but I can see at least a few pathetic horn dogs falling for it.
Now let me see if any of the men offered any defense.
Update: I'm back. A few of the men said they thought it was a role play situation and that they were led to believe this by her husband. But most of them sound utterly indefensible. So, I underappreciated the horror by quite a bit.
There's a full table here with notes about each man and their defenses, if available.
https://en.wikipedia.org/wiki/Pelicot_rape_case#Convicted
They kind of sound like the dregs of society to me.
But yes, what is there to say? You can find dirt bags to do sex crime if you look long enough. The were all publicly shamed and convicted. The right thing happened, at least. My update is France takes these crimes more seriously than I would have expected. Also the right thing.
I'm slightly reassured it was only 72 men over 9 years and not, like, a thousand. Suggests the candidate pool is not a vast ocean.
I was unaware of this article.
My attitude towards AI tools for the last 18 months had been "yeah they're useful but if you try to get too ambitious with them they waste more time than they save" and I was like AI 2027? Ha, try AI 2035.
But something fundamentally changed with the models in the last month. I'm low key freaking out at how goddamn useful they are now through Claude Code and OpenAI Codex.
Forget METR evals. My personal real world evals are that they're 6/6 on doing 2-4 week long tasks in 1-2 hours.
I'm just doing this on the $20 OpenAI Plus monthly plan. They haven't throttled me yet. Feels like year 2020 Uber rides across town for $4 stuff.
Originally, I had kind of given up on claude code a few months ago because it was wasting more time than it was saving me. It would lose the plot pretty quickly even though my instructions and the goal were still well within the context window. I considered this an architectural limit of LLMs.
But as of the last two weeks, holy shit. claude code (and codex), they just grind away at problems. They don't lose the plot. They back out and try different approaches. They run micro experiment to test assumptions. They'll run the tools with --help and --version and check the man page. They'll step through the code of the installed version of meson to see why the config file is not behaving the way it expects. I just give them like one simple prompt and it'll chug away for 15-30 minutes just trying shit like an overly caffeinated engineer. They'll run builds and look at errors and fix them until it's clean.
And again, they don't lose sight of the goal. It's amazing.
I'm not really convinced I need to be more than "experienced computer user with good taste" in the end. Probably not much more than a product manager.
I was mentioning io_uring or JITs etc as examples of extreme customizations the agents can do. But if you just tell it what you want and then complain "it's not fast enough, make it faster" it will, on its own, come up with and implement optimizations. The standard ones like better algorithms or pre-computing stuff, but they will get increasingly sophisticated and extreme if you keep saying "make it faster".
My specialized knowledge helps us not get painted into a corner as much, but does it matter when the paint dries instantly and it can repaint the entire house in minutes?
My biggest advantage for now may be that I can approximate in my head the theoretical limit for how fast something could be physically driven on the hardware, so I'll know when to stop saying "make it faster". I'll also know that when the coding agent says "I'm going to bake in hardware assumptions and weaken consistency models" that it might be worth stopping as well.
But this edge won't last for too long.
dr_analog
top 1% of underdog fetishists
4d ago
I normally use Claude Code but this week I've been using GPT5.2 with Codex on the $20 plus plan. They haven't throttled me yet, probably because they're blowing VC money like ZIRP is back in style.
It's so fast and IMO better than Claude and needs less direction. I'm going to be devastated when and if the party ends and I need to pay actual costs.
Well. I don't know what the rest of you do but every elite dev I know that had been skeptical and unimpressed by AI agents has been converted in the last month.
Android apps. Firmware for wearables. UNIX TUI clients. Code analysis tools. Web apps. Flight simulators. Gaussian splatters. One guy writing a functional formula language for a network message bus. 3D games for PCs. A Signal clone that doesn't require phone numbers. Bots to run trading strategies.
If you're in a niche where this hasn't happened to you yet, bless. It's probably better for your mental health to not cross this threshold.
I'm not vouching for Cursor or Copilot or the general chat experience. But Claude Code and OpenAI Codex, within the last two weeks, running the latest best models, are what are scaring the shit out of me. Before this I was mostly in "meh, loses coherence too fast, maybe in 5 years they'll figure it out" mode. No. It's here now.
dr_analog
top 1% of underdog fetishists
5d ago
.I had previously gathered the impression that AI coders were still at the level of an enthusiastic but sloppy apprentice.
Not anymore. They're basically as good as a senior software engineer now, except they finish 100x faster. And never need to rest.
At this point anyone not using them is resisting out of inertia. Or fear. Well placed fear, perhaps.
I am beginning to low key worry about how good the latest coding agents are, claude code and codex, within the last two weeks. I am routinely building apps in 2 hours now that might have taken me a month if I had to use my own brain.
They'll tailor make stuff to any level of customization or weirdness. Use io_uring? Do this part in x86-64 assembly? Want a JIT for you app? Want to see if we can make this algorithm run on your GPU? Sure it's all good.
If I spot a bug it will take less than five minutes to fix. They never give up.
The slowest part of the loop here is me. I can't test changes and describe features fast enough.
This latest app I've put down about 3500 lines of code and haven't looked at any of it. It may as well have been written by a total stranger in a language I don't understand, it just slows things down too much if I try to read the code.
I am guiding it using my skills and experience but it doesn't really matter. When it can rewrite the entire app in an hour there are not that many bad architectural decisions that can't be undone.
Obviously in a mature product with users and deployed infrastructure, radically changing your approach is harder. But even already it tries to steer me away from crazy stuff.
I'm sorry but software engineer grunts are cooked. If you can't design or product manage yourself, you're going to be unemployed. What does it mean to be a SWE in a world where software is built as fast as you can describe it?
sorry I edited this in a few minutes after my reply
More options
Context Copy link