Loading...

Well, I've been slowly teaching myself how to use it the last couple of months migrating away from Microsoft OneNote I've been a OneNote user for about a few years with you know with my job Anyhow, so I've been using a CD and more and more for the last few months and I go ahead and spun up a Docker container, so I host it on my on my little pie now, and I have it direct sync to both of my mobile devices, my phone and my tablet, and I also have it to sync to my work Windows PC. So now I have complete syncability between all four of the major things that I use every day. My next step, and I was researching it last night, and I dreamed of it last night, I'll be honest with you, I guess maybe I researched too long on it. I am planning on integrating it with my chat GPT, my Google, my Gemini, those two for sure. Anyhow, I'm going to integrate it and everything that I use those LLM for will automatically sync into my Obsidian folder. So you know like, okay, LLM's, their chat threads are limited to this chat thread, right? So if you've got 20 different chats in, say, ChatCPT, well, it's really only going to reference the one that you're in unless you specifically tell it to go reference another or unless you chat and paste the contents of another chat into the current chat window. A lot of people don't understand that they don't really cross chat windows. Kind of like Ghostbusters, don't cross the stream. That's how OpenAI set up ChatCPT. And Jim and I set up much the same way. And I find that to be very limiting. So I'm planning on solving that with Obsidian. I'm going to set up a special folder specifically for this inside of my Obsidian vault. And it's going to be a data dump. A data dump note vault, where everything I use on those LLMs will just get dropped there. That way, if I ever need the rest version of it again, I can say, hey, search my Obsidian Vault for this, this, and this, we talked about it six months ago. And it can actually go and do that and then bring it back to me rather than me having to reteach it or copy and paste every time I open up a new chat window in one of those LLMs. I think I described that clear enough. If not, let me know and I'll try to find a better way to describe what I'm doing. KJ5 ILQ. Now that's very clear, not something I thought about. I run Open Web UI at home for my chat, or AI chats, and I save all of the chats just in the sidebar. I've been doing some more some reading on the what are they calling now MCPs and tying in multiple LLMs into into one and then it picks and chooses based on what you're asking which model to call and it also allows the models to hand off to each other so that sounds really cool too now I don't I don't take notes of my phone so I don't worry about syncing I keep my home network is other than a couple of services that I proxy in is totally isolated and a good chunk of it is even air-gapped under physically a separate switch in the house so all of the IOT stuff is non accessible from the internet and it can't access the internet likes to keep my local stuff local you know what I mean