597: Cache My OS

Just had a random thought it would be cool if instead of discontinuing the podcast, it could just be a merger of sorts with LUP just did a self hosted section as part of that podcast. Obviously, I may not understand some of the things going on in the background that would keep this from being a thing.

Sad to hear the show is coming to an end I am so thankful for the content and energy you have all invested into Selfhosted. In reply to Hybrid Sarcasms boost, OpenWebUI actually has a partial solution to a searchable obsidian vault of documentation with their Knowledge implementation. It's really easy to set up and works okay. Setting up a knowledge base in OpenWebUI can be done in the workspace. You can upload your entire obsidian vault directly through the WebUI and attach that Knowledge to a custom model based on whichever LLM you would like, ollama or API based. I only wish there was an implementation that allows syncing this from obsidian.

Following up on last weeks boost, it was a typo and it's called cheat. Double checked for typos this time. Thanks for the great work from JB!

Conversation regarding Snippety reminded me of a MacOS app I use called Clips, it provides historical clipboard using command+c and command+shift+v. I have found it extremely useful when copying and pasting commands usernames and passwords. I can't really speak regarding the security of having something with a complete history of your clipboard but it's great to be able to select anything in your clipboard history from a context menu that appears upon pressing the combination of keys.

Like many, I’ve accumulated numerous Docker containers. I want to ensure some stateless containers remain available if my primary server goes offline. Considering NixOS, Kubernetes, and Proxmox HA, which would best achieve high availability while using code as documentation? Open to other solutions I am unaware of as well. What are your thoughts? Keep up the good work. Tebby

I find people have become lazy due to having video. Often times I'll be listening to a podcast and they will refer to something that the viewers can see on video. They don't describe what they're holding so the listener has context and then as a second thought will describe it "for the audio listeners".

The podscape runs on lightning