• 0 Posts
  • 7 Comments
Joined 3 years ago
cake
Cake day: June 14th, 2023

help-circle
  • i2p is a more workable and in many ways more private alternative to tor when it comes to hidden websites and stuff, but it’s intentionally different and separate from traditional internet traffic, instead of trying to be compatible with it. It has a more lively community in my opinion, and most of its software and stuff is pretty much as simple as you describe, although a lot of it is hidden away on the i2p network itself.


  • Email’s not nearly as hard to host as people make it out to be. It is intimidating, yes, and there are a few very specific gotchas that make it infeasible or impractical from most if not all home internet connections, but if you’re willing to outsource just a little bit on that one particular point so that you have a static, commercial-grade IP address somewhere that you can get reverse DNS and open up SMTP ports on, the rest of it is mostly just fighting with your usual, standard self-hosting struggles of getting it configured and working at all, then making sure it’s secure and stays secure, and making sure your data is protected. The setup might be slightly more advanced than some other services just due to the number of moving parts, but it’s not fundamentally any different than a webserver. There are some other issues that might crop up a little later in the game, managing your mail server’s reputation for not sending spam (especially if, at some point, someone DOES get access and starts sending spam, you need to figure out how to recover your reputation). But overall it’s manageable, I’ve been self-hosting my email for like 30 years. It sucks, but it’s worth it.


  • Sort of, there are some similar aspects to it. I refer to it as my “internet appliance gulag”. They can all live there in their big open-air prison and be as totally insecure as they want, spying on and hacking each other all day long and they can try to access my network or the internet all they want, none of them are getting out unless I have explicitly reviewed whatever the fuck it’s trying to do and where it’s trying to go and why.

    Do not phone home, do not pass go, do not collect $200, go directly to Gulag.


  • No they weren’t, how did you get that idea? Nowhere did they ask for a specific fix, nor did they provide the details necessary for a specific fix. The OP asked how it works, and according to my reading of both their post and their comment, they seemed quite curious about what was going on (which is a good mindset to have when attempting to solve a problem) so my intention was to answer some of that curiosity by explaining “how”. You’re the one who decided someone should just provide a fix (which, in this case, would’ve been basically impossible)

    Especially when a problem arises because of misunderstanding or ignorance (which is not a shameful thing, it’s simply the default state of understanding), an improved and better understanding of the situation often helps lead one quite quickly to the solution to the problem, and in this case, it seems like OP did solve their own problem once they had a better understanding, so… seems like a successful effort to me? Any particular reason we need to argue about this?


  • It works the same way it normally works. You generate the certificate, give it to them, tell them who you are, they verify who you are, then they sign it and give it back to you. Very routine and straightforward, except for the challenge part that they use to verify your identity.

    Think of it in terms of delivering a pizza. The signed certificate is the pizza in this analogy, and they don’t want to send it to you until they’re sure it’s going to the right place. Either the store can send it to you (you have to give them your address and have port 80 open for when they arrive) or you can go pick it up at the store (you give them your address, they verify it through DNS, no port 80 required).

    There’s never been anything preventing you from establishing a conversation with the certificate-issuer. You don’t need port 80 open for that. However, unless you have port 80 open the issuer cannot reach out to you, and that’s how they verify you are who you say you are. DNS challenges avoid them having to reach out to you directly, instead they reach out to DNS on your behalf and verify you are who you say you are that way instead.



  • 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)