The authoritarianism is exercised differently in both countries. Mainly, the US state is openly more physically violent. In China, surveillance, media control, censorship, and control of their population in general is stronger (but the US is rapidly trying to catch up). As one example, the US wouldn’t have been able to effectively silence Naomi Wu and presumably remove her way of making a living at the time, if she was a US citizen in the US.

Specifically, I’d love to hear your thoughts on how to better optimize the context window limits when dealing with massive raw outputs (like huge DNS dumps or nmap scans) before feeding them back into the LLM’s memory.

Launch a sub-agent that reports its findings back to the main orchestration agent. If it’s even too long to fit in the sub-agent’s context, you can chunk it up, have a sub-sub-agent per chunk report back to the sub-agent for a shorter summary, just the interesting lines per chunk, or “no relevant lines found” or whatever. Can get even fancier by allowing the sub agent to use tools like grep, head, tail, etc on the text to search it instead of reading the whole thing directly.

Surprised you’re not using LangChain/LangGraph as it makes some of the things you’re doing easier. But it looks like you’re vibecoding this anyways, so it’s just doing whatever Claude Code or whatever decides to do. My suggestion would be to code it yourself with minimal AI assistance, as this will just turn into an unmaintainable mess as time goes on, and eventually, the AI coding agent will get stuck and be unable to really progress.

Now that I think if it, you could probably get Claude Code or OpenCode to do everything this project can do by just installing all the tools needed in your environment, creating a new empty project, telling it what tools are available in your environment, allow it to download any other tools necessary, and prompting it to do the recon (may need to use an abliterate, heretic, or otherwise uncensored model to do some things).

TIL LibreWolf randomizes some fingerprinting targets.

I recently noticed that Postgres has “sounds-like” functions too, which I thought was neat: https://www.postgresql.org/docs/current/fuzzystrmatch.html.

Tons of people are getting fired because the owners think AI can replace them. Doesn’t mean the AI can do the work properly, but they are getting fired anyway.

I know a person that works at an AI startup and they convinced a company to replace their HR department with AI a while back. Funnily enough, that startup’s “AI” is largely “Actually Indians.” Their service is an agent that writes its own “tools” to solve problems/complete tasks, but the tools often don’t work, so they have a large team of devs in India rewrite them or do the tasks.

It’s weird that the mass layoffs are happening during the bubble inflation this time now. That “techno-feudalism” idea really seems like where the world is heading. Doesn’t matter if the products and services work or not, they still make money from the irrational stock market hype, lock-in and rent extraction, and governments.

AstroNvim is nice. It’s an opinionated and pre-configured nvim distro, so you don’t have to spend as much time finding the right plugins and configurations.

Yeah, for personal projects, I’ve been going back to just HTML, CSS, and some “HTML Web Components” for progressive enhancement where needed. The complicated build process and bloated frameworks of modern JS web development just feels wrong to me. JS has enough features now, where you don’t even need libraries like JQuery.

Ive been hired at every job ive applied to

That’s crazy. Straight out of high-school, I spent about six months driving around and applying to a job or 2 every day before I got a job (applying to pretty much every business in the small towns I lived near; McDonald’s, Wendy’s, Walmart, factories, everything). Now, I’m an unemployed software engineer, and I’ve probably applied to > 500 jobs, and still no luck (I think every job posting gets > 1000 applications in this job market). The vast majority of the time, I never actually get a chance to even talk to a person to impress them with my awkwardness and anxiety :)

IDK, this post says “GPT-5.3-min” which is not an open-weight model AFAIK. Either way, everything is public on the Fediverse, so IDC very much. Specifically asking for political classification is a little bit weird though.

I’m mostly just surprised that a mod would pay for tokens to moderate. The Fediverse is radically public by design, so I don’t have any expectation of privacy. I’d bet at least someone is gobbling up the entire Fediverse to train AI, since companies are so desperate for new human-generated data.

I doubt it’s the “culture.” It’s just that they used to specialize in making things for the lowest cost possible. That’s changing bit by bit. DJI is best in class for example; no other consumer product comes close. They are also leading in many scientific fields.

Can states not sue companies that accept payment from citizens within their state?

I’ve seen reports that Trump was going to go for Greenland and his advisors shifted his attention to Iran instead. There was also that time Trump leaked Macron’s message where Macron said he’d support Trump with Iran if he backed off Greenland.

Cuba’s current government won’t let US businesses extract Cuba’s wealth and exploit their citizens.

These markets were illegal for so long because they can easily turn into assassination markets with plausible deniability.

I think offshoring/nearshoring is booming right now. Tons of “Global Capability Centers” being built. It’s even cheaper than H-1B.

I never heard of pirating NES/SNES/Genesis back then, so I don’t think it was very common. Renting was pretty cheap, and you could always trade or borrow games with friends. We didn’t get a computer until the late 90s when the WWW was taking off, but yeah, pirating games is probably one of the first things I did. Downloading a large number of RAR files from warez sites over dial-up, taking multiple days, IIRC. And ROMs and emulation. Someone in my school would put cracked PC games on a shared network drive, so you could play them in the library. Later, I went to a technology focused high school, and we’d all play (cracked) Half-Life or Unreal Tournament multiplayer before class started every day (usually with the teacher).

Mostly just cost. Broke as hell. Would be nice to visit if I could justify the expense at some point in my life.

IDK. I don’t trust Chinese (closed-source) software either. I believe TikTok does gather a ton of data (though it’s not Chinese now). On the other hand, I will likely never go to China, so there’s not much the CCP could do to me (maybe blackmail me if I was a politician or something); while the opposite is true for the US.