Man, Hacker News was buzzing yesterday, March 10th. Lemme give you the quick rundown of the interesting stuff:

Big Tech News & AI Shenanigans

First up, some sad news: Tony Hoare passed away. You know, the computer science legend. Everyone was talking about his famous "billion-dollar mistake" with null references. Apparently, he even gave a talk about it back in 2009! Pretty wild to think about how much impact one decision can have. You can read more about it here.

Then there's this kinda scary thing about online age-verification tools. They're supposed to protect kids, right? But it turns out they're also doing a whole lot of surveillance on adults. People in the comments were saying it feels like governments are just looking for a "foot-in-the-door" to control the internet more. Check out the article on CNBC.

Speaking of AI, Amazon had some outages and now they're making senior engineers sign off on any AI-assisted code changes. Makes sense, right? But some folks in the comments were actually arguing that reviewing AI-generated code takes *more* time than just writing it yourself because it's often over-engineered or buggy. Others said code reviews are more about spreading knowledge than just finding bugs. Wild stuff. Ars Technica has the story.

And Meta just acquired a company called Moltbook. It's all about AI agents for social networks. Apparently, Zuck's still trying to make big moves. One interesting comment was about how a lot of people are just using AI tools like Claude Code to make "single-serve" software to solve their own problems, even if they don't know how to code themselves. Axios covered the acquisition.

Cool Projects & Big Ideas

Someone posted about how they put their whole life into a single database. Like, everything! It's a super detailed personal tracking project. It reminded some commenters of Robert Shields, who kept a 33-million-word diary back in the day. This guy's blog is called How is Felix Today.

Also, Yann LeCun (from Meta AI) just raised a massive $1 billion to build AI that understands the physical world, not just language. Think about it – AI that can really grasp how things work in reality. People were excited about the idea of "world models" learning things even humans don't fully understand, like how insects fly. Pretty ambitious! Wired has the details.

And on the open-source front, the Redox OS project announced a strict "no-LLM policy" for contributions. Basically, no AI-generated code allowed. This sparked a huge debate in the comments about "AI taint" and whether people would just lie about using AI anyway. It’s a tricky one for open-source projects moving forward. You can see their new CONTRIBUTING.md.

That's the main gist of it, man. Lots of AI talk, as usual, but some cool and thought-provoking stuff in there. Talk soon!