Sorry about your wallet then, haha. :) He’s got a lot of videos and a lot of reviews. I assumed almost everyone already knew about the nerd castle. He’s a pretty prominent and prolific reviewer of indie games (about to hit 1 million subs on Youtube) with a slight but noticeable lean and appreciation for the grittier, gorier, creepier and crunchier side of indie gaming that a lot of the usual “cozy indie” communities lack. It’s a good place to be if you appreciate that kind of stuff.

I’m surprised splattercat hasn’t played it yet? The description makes it seem right up his alley. You can tell from the videos he makes about them that he particularly loves twisted dystopian space looters like Quasimorphs and Ostranauts and weird, equally crunchy and mechanic-heavy scavenger shit like Roboscav and Mirklurk and they all tend to be the kind of games he’s either played a bunch even before making videos about them or games he keeps coming back to and making additional videos about. I eagerly look forward to seeing a video about it from him sooner or later.

One might even say that Jeffries and Schumer are some of the powerful people who need to be held accountable. Hashtag NotAllDemocrats /s

As I understand it, SearXNG is a decentralized and anonymized peer-to-peer network of search-engine scrapers. So you’re not self-hosting the search engines, you’re self-hosting a node in the network that scrapes real search engines. It’s basically search-piracy, except I don’t think they’ve managed or bothered to make it illegal yet, so, enjoy it as long as we can.

As a SaaS vendor, I guarantee Google knows how to use feature flags and conditionals to limit access to certain code paths. I know they do, because I also work for a SaaS vendor, and ex-Google employees are the ones who taught us how to do it. There is also no possible way it is as intertwined in the code as they are implying it is. It is awfully convenient for them to claim that, but it’s nonsense. The stated goal of preventing students from accessing AI results is quite simple, from client side technical analysis alone I can tell, at least in most ways, how literally tacked on it is. While it wouldn’t be an ideal solution, I can personally literally block it with a CSS rule. So could the schools, technically, and it would more or less achieve their goal. But most importantly, so could Google. The fact that they won’t tells you everything you need to know about the honesty of their position.

Stop believing and repeating what they’re claiming, you’re giving near-trillion-dollar companies the benefit of the doubt but they’ve done nothing to earn that benefit. They’re corporate liars, they’re deliberately destroying the world and smashing holes in our civilization like a bull in a china shop, and we keep dancing around like we are afraid of being unfair to them, even while they are in the act of being unfair to us and literally standing in front of us telling us something we can see with our own eyes is not true. Just because they have lots and lots of money and lots and lots of platforms to make sure we have no choice but to know all the things they are saying, doesn’t mean the things they are saying are at all trustworthy. We need to get over this habitual deference and stop giving equal weighting (or any weight at all) to corporate lies from untrustworthy sources.

Since jet fuel is a very similar grade of fuel to diesel, is this where all the diesel is going and why diesel prices are astronomical now? I know the wars are also not helping, but I wouldn’t be surprised if this is contributing too.

Of course they can’t, they don’t know how to code. Claude can probably do it for them though, but it’s going to cost a lot of tokens for such a substantial rewrite.

Money is a shared delusion. It’s been fun. But it’s time to break the delusion and start asking ourselves what our money is actually for, what we actually need, and whether money is still the only way we can get those things.

That’s some mighty fine police work, Lou.

Absolutely. There are tons of open-licenced, open-weight (the equivalent of open-source for AI models) models capable of what is called “tool usage”. The key thing to understand is that they’re never quite perfect, and they don’t all “use tools” quite as effectively or in the same way as each other. This is common to LLMs and it is critical to understand that at the end of the day they are just text generators, they do not “use tools” themselves. They create specific structured text that triggers some other software, typically called a harness but could also be called a client or frontend, to call those tools on your system. Openclaw is an example of such a harness (and not a great or particularly safe one in my opinion but if you want to be a lunatic and give an AI model free reign it seems to be the best choice) You can use commercial harnesses too by configuring or tricking them into connecting to a local model instead of their commercial one, although I don’t recommend this for a variety of reasons if you really want to use claude code itself people have done it but I don’t find it works very well since all its prompts and tool calling is optimized for Claude models. Besides OpenClaw, Other popular harnesses for local models include OpenCode (as close as you’re going to get to claude for local models) or Cursor, even Ollama has their own CLI harness now. Personally I use OpenCode a lot but I am starting to lean towards pi-mono (it’s just called pi but that’s ungoogleable) it is very minimal and modular, making it intentionally easy to customize with plugins and skills you can automatically install to make it exactly as safe or capable or visual as you wish it to be.

As a minor diversion we should also discuss what a “tool” is, in this context there are some common basic tools that some or most tool-use models will have or understand some variation of, out of the box. Things like editing files, running command-line tools, opening documents, searching the web, are common built-in skills that pretty much any model advertising itself capable of “tool use” or “tool calling” will support, although some agents will be able to use these skills more capably and effectively than others. Just like some people know the Linux commandline fluently and can completely operate their system with it, while others only know basic commands like ls or cat and need a GUI or guidance for anything more complex, AI models are similar, some (and the latest models in particular) are incredibly capable with even just their basic built-in tools. However they’re not limited by what’s built in, as like I said, they can accept guidance on what to use and how to use it. You can guide them explicitly if you happen to be fluent in their tools, but there are kind of two competing models for how to give them that guidance automatically. These are MCP (model context protocol) which is a separate server they can access that provides structured listings of different kinds of tools they can learn to use and how they work, basically allowing them to connect to a huge variety of APIs in almost any software or service. Some harnesses have an MCP built-in. The other approach is called “skills” and seems to be (to me) a more sensible and flexible approach to giving the AI model enough understanding to become more capable and expand the tools it can use. Again, providing skills is usually something handled by the harness you’re using.

To make this a little less abstract you can put it in perspective of Claude: Anthropic provides several different Claude models like Haiku, Sonnet, and Opus. These are the text-generation models and they have been trained to produce a particular tool usage format, but Opus tends to have more built-in capability than something like Haiku for example. Regardless of which model you choose though (and you can switch at any time) you’ll be using a harness, typically “claude code” which is typically the CLI tool most people use to interact with Claude in an agentic, tool calling capacity.

On the open and local side of the landscape, we don’t have anything quite as fast or capable as Claude code unfortunately, but we can do surprisingly okay considering we’re running small local models on consumer hardware, not massive data center farms being enticingly given away or rented for pennies on the dollar of what they’re actually costing these companies on the hopes of successful marketshare-capture and vendor-lock-in leading to future profits.

Here are some pretty capable tool-use models I would recommend (most should be available for download through ollama and other sources like huggingface)

• gemma4 (the latest and greatest hotness, MIT licensed using TurboQuant to deliver pretty incredible capability, performance and results even with limited VRAM)
• qwen3.5 (from Alibaba, a consistent and traditional leader in open models so far with good capability and modest performance)
• qwen3-coder-next (a pretty huge coding-focused model you might struggle to run unless you have a very beefy system and GPU)
• glm4.7-flash (a modestly capable and reasonably fast option)
• devstral-small-2 (an older, not-so-small variant of mistral, the French open-weight AI model if you’re looking for a non-Chinese, non-US based model which are few and far between)

All this bullshit about phones with folding screens nowadays when what I really want is a phone with a folding mechanical 104-key :P