I think most reasonable people understand that politicians are fallible and can be shortsighted and make mistakes. This is part of being human and part of life in government which, at it's core, is just a huge risk management agency. That's not to say that honest mistakes don't carry weight - Bush Jr may have truly believed everything he claimed about Iraq prior to invading it, but it was still a fool's errand, built on a scaffold of lies, and one of the biggest wastes of resources and lives in US history. Bush should be excoriated for that mistake, even if it was honest.

The problem is when intent and motive undermine the attempt at honest governance. If a politician has ulterior motives, such as personal enrichment, or to gain political advantage over an opposing person or party, to save face in the light of a mistake, or any number of other human foibles and failings, that's when it becomes a valid point to make.

For the helium reserve, it seems to have been a victim of shortsighted attempts to reduce the deficit. The program had accumulated almost $2 billion to buy and store it, and some people thought it wasn't needed.

When it comes to Trump, many people don't trust his motives because he's shown repeatedly to do and say things that are dishonest or hide the true intent. It seems fair to ask who stands to benefit from this change in helium classification.

Well said. This is a very succinct way of describing the secondary effects for everyone around someone using AI to feel "more productive". Sending people massive emails you didn't even bother to read, creating huge design documents that are just dozens of meaningless pages and offer nothing of tangible value, or asking fellow developers to reviews 50,000 line code changes you didn't write (or read), etc.

I use AI at work for development and I know that the frontier models in a harness like Claude Code are more capable than many AI critics claim but I also see the results they create in a skeptical light. For a prediction chatbot they are surprisingly good at writing software, but also surprisingly bad compared to a decently skilled human. They can generate massive amounts of design documents and code in a very short time, and easily overwhelm humans trying to read, review, verify, and test the changes. The same thing is happening in other kinds work as well.

Drug use is an apt metaphor. The Dr Feelgood of this decade.

If you ask a bare LLM to do stuff you'll get a mess. I run Claude code with a multipart harness that compels the model to treat the code as the mathematical thing it is, not human prose. LSP Enforcement Kit keeps low capability tools like grep out of the mix, Abstract Syntax Tree things handle the code as its meant to be understood, so that's CodeSight, OptiVault, Serena. About a year ago I gave Antigravity the boot after their awful 2.0 release and went back to JetBrains IDEs, which now offer code structure wise MCP services. Code review with the Ponytail skill chases away LLM excesses, and there's a smattering of other stuff like that in the mix. And then you hunt for whatever additional guardrails you can get the LLM to apply systemically.

I've evolved my own higher level project methods and they're comparable in a lot of ways to what Matt Pocock offers, but his stuff is more structured, more carefully considered, and I have to work with others now, after a long time solo. I'm taking his stuff, implementing a Kanban/wiki system, Gitlab CE, and NanoClaw for anyone who wants agents. I expect quite the slog between now and New Years, but in 2027 we're gonna stop trudging and start gliding over the landscape.

Fake news. We have such a HIGH IQ president (he has taken and aced FOUR dementia tests!) that nobody is smarter than he is, so there's nothing to be afraid of.

Please go read the definition of "enshittification". It doesn't mean "anything I don't like."

The group that designs Firefox has made some very questionable decisions over the years, but enshittification of the browser is definitely not on the list.

The locals are going to get hit with bills from Trump's order to keep Colorado's coal plants open, why should this be any different? https://coloradosun.com/2026/0...

I'm about as far from an LLM booster as possible, and hate to be that guy, but "you're holding it wrong".

ChatGPT shouldn't be used for real software development. It's a toy chatbot. If you're doing development you should be running a harness like Claude Code or OpenCode or whatever locally on your development machine. That gives the model and tools full access to your source, including private repositories, libraries, databases, documentation, bug tracker, logs, etc. These tools are also much, much better at seeing the connections between code and concepts and will do a better job understanding and writing stuff "end to end".

We've reached a point where the quality of the LLM harness is probably equally as important to the LLM itself at writing software, so if you're still using web-based stuff like ChatGPT and pasting single blocks of code or files into it, you're about 2 years behind where most other people are and it's not surprising you don't see any value from it.

We are nowhere near eliminating humans from development, but the current set of harnesses and models actually do a pretty good job when used properly and in the hands of someone who already understands the codebase and domain. They can still fuck up royally, but, well, so can humans.

If I were, perhaps it's because I'm not seeing any of the real stuff here.

The enemy of my enemy's enemy is my....?

Hm. What's the involution function rule for asshat corporations and their pet lobbyists?

As Eric Raymond (ESR) says, AI is a force multiplier. My elaboration: If you are good at what you do, it will let you do accomplish more good stuff faster. If you are not very bright, AI will let you be not bright more efficiently, but it won't make you bright. It's just a tool. It needs to be utilized correctly. There is no shortage of things in this world that can do tremendous damage if not used properly. AI is nothing new in that regard.

"AI slop" is so 2024. That argument isn't valid any more.

I'm surprised you didn't use the word "slop". Spewing a bunch of slogans that would have to be more clever to make it on to bumper stickers makes you look like the people you are complaining about. There are legitimate criticisms to make about AI, and ways to support and defend those arguments. You are not doing that.

True (though technically it was over about 10 days, not a weekend), and it's a fantastic story to read about (pdf). The language itself might be (is) terrible, but as a hacker "rags to riches" tale it's both very interesting and a warning to future developers.

Firing wasn't enough. TO THE MINES WITH HIM!