- InsideOutSantaThat's certainly less fancy and looks outdated, but it's a lot more functional.
- I think this is the first time I've seen a website where the download button, which is just a link, requires JavaScript to render.
- People all over the world are already building new bridges to places like China, so even if the old ones are rebuilt, they might get substantially less use.
- >Are there any good robo-vacuum cleaners that will still clean your floor if the internet is down?
It depends on what exactly you want. My Roborock can't connect to my Wi-Fi anymore for some unfathomable reason. It no longer runs automatically, and I can't edit its map or tell it where exactly to clean. I just hit the power button once a day to start it manually, and it cleans everything it can access.
- I would assume that if the model made no assumptions, it would be unable to complete most requests given in natural language.
- I bought two Ventas well over a decade ago, and they still work as well as the day I bought them. They're an expensive initial investment, but IMO worth it over the long run.
They are also mechanically simple, so I trust that if they ever break, I will be able to repair them.
- Iirc, Heisenberg reinvented Matrix calculations to solve a problem in quantum physics. Not being a mathematician, he wasn't aware of the concept. Born recognized what Heisenberg had done and introduced him to his own reinvention.
- It runs on your own hardware. There is nobody else who has access to unencrypted data.
- Both because of the content, and because of the odd perspective shifts in the AI-generated footage. It made me feel like I was drunk.
- That ad is brutal. I feel sorry for people who experience Christmas like this.
- The worst part about the Coke ad is the fake "making of" video they released to show how much manual work went into their ads. The "pencil sketches" ostensibly made by humans in the making of were also AI-generated.
- In my experience, the messed up closing brackets are a surprisingly common issue for LLMs. Both Sonnet 4.5 and Gemini 3 also do this regularly. Seems like something that should be relatively easy to fix, though.
- I gave Devstral 2 in their CLI a shot and let it run over one of my smaller private projects, about 500 KB of code. I asked it to review the codebase, understand the application's functionality, identify issues, and fix them.
It spent about half an hour, correctly identified what the program did, found two small bugs, fixed them, made some minor improvements, and added two new, small but nice features.
It introduced one new bug, but then fixed it on the first try when I pointed it out.
The changes it made to the code were minimal and localized; unlike some more "creative" models, it didn't randomly rewrite stuff it didn't have to.
It's too early to form a conclusion, but so far, it's looking quite competent.
- There are battery-free switches for Hue light systems and, presumably, other similar systems. They're pretty large, though, and pushing the button has a lot of travel, which can be used to generate electricity.
- Yeah, this would have been a neat tool for "middle-of-the-night" thoughts, but I don't want a disposable electronic device. I get why it is that way; not having any charging hardware probably makes the device much smaller. But I'd rather it be a bit bigger and rechargeable.
- The paperclip idea does not require that AI screws up every time. It's enough for AI to screw up once in a hundred million times. In fact, if we give AIs enough power, it's enough if it screws up only one single time.
The fact that LLMs do it once in a thousand times is absolutely terrible odds. And in my experience, it's closer to 1 in 50.
- You're literally telling me that the thing that has happened on my computer in front of my own eyes has not happened.
- But LLMs already do the paperclip thing.
Suppose you tell a coding LLM that your monitoring system has detected that the website is down and that it needs to find the problem and solve it. In that case, there's a non-zero chance that it will conclude that it needs to alter the monitoring system so that it can't detect the website's status anymore and always reports it as being up. That's today. LLMs do that.
Even if it correctly interprets the problem and initially attempts to solve it, if it can't, there is a high chance it will eventually conclude that it can't solve the real problem, and should change the monitoring system instead.
That's the paperclip problem. The LLM achieves the literal goal you set out for it, but in a harmful way.
Yes. A child can understand that this is the wrong solution. But LLMs are not children.
- The Newton had an absolutely brilliant user interface, but its hardware let it down. I wish we had a device similar to the smaller Remarkable tablet, running something like NewtonOS.
(Related: Columbo's Mystery Capers is an amazing and hilarious game from one of the Zork devs that has never been dumped and doesn't exist on any other platform. I hope someone will figure out how to get it off an actual Newton and get it to run on Einstein.)
- That's the crazy part. The people at the top seem to think they're better off if they can get another billion in the bank, regardless of the impact on the rest of society. But they, too, live in that same society that they are destroying.
They seem to think it's better to be a king in the Middle Ages than just a regular rich person in modern society. They forget that the lives of kings in the Middle Ages were absolutely terrible.