Tech Rants

Stop wasting time with the classic "verb-noun" commands of Vim. If you're tired of guessing what your complex keystrokes will do, it's time to meet Kakoune. This editor flips the script to a powerful Select-Then-Act model, giving you instant, visual feedback and making multiple, parallel edits a core feature. It's faster, more intuitive, and built for modern coding efficiency...

In this episode, I walk you through the process of installing ONLYOFFICE on Fedora 43 and integrating it with local AI using Ollama. Perfect for those who want to run a powerful self-hosted office suite on Linux without relying on the cloud. Learn how to: Install ONLYOFFICE on Fedora Set up Ollama AI for seamless document editing Run everything locally, ensuring privacy and performance For detaile...

Learn how to install ONLYOFFICE, the open-source office suite, on your own server using Podman. In this detailed guide, we walk you through the process of setting up ONLYOFFICE for secure, self-hosted document editing. Watch the full screencast or check out the blog article for step-by-step instructions. Full tutorial and written guide: https://ojambo.com/how-to-install-onlyoffice-secure-online-of...

Stop accepting low-quality AI video. If you run the powerful Wan 2.2 T2V 14B model, this episode is a game-changer. We dive deep into migrating the setup from ComfyUI's default settings to stable-diffusion.cpp—the key to unlocking dramatically better quality output on Linux. This is the optimal configuration for high-performance GPUs like the AMD Instinct Mi60 32GB HBM2. Learn how a C++ implem...

In this episode, we explore how to set up the Gemma-The-Writer-J.GutenBerg-10B-D_AU-Q6_k LLM using the Llama.cpp backend on your custom web interface. This setup allows you to unlock the power of large language models for content generation, AI applications, and more. Learn the steps, get insights into the Apache License 2.0, and understand the licensing restrictions for the model's weights. F...

In this episode, I discuss my 1-year experience with the Honeywell HCE100B Ceramic Heater. Learn how I used it to lower my home propane usage, keep my office comfortable, and save on energy costs. I cover power settings, efficiency, and whether this compact heater is worth buying in 2025. Full blog article: https://ojambo.com/post-unboxing-honeywell-hce100b-ceramic-heater Long-form video review: h...

In this session, we break down the non-negotiable steps for securing any user registration form. Learn why client-side validation is never enough and how to build a robust PHP backend. We cover strict server-side validation, the critical use of password_hash() to secure credentials, and implementing prepared statements to defend your MariaDB database against SQL Injection. This is essential listen...

In this episode, discover how to configure stable-diffusion.cpp with WAN 2.2 TI2V 5B on Linux, using the AMD Instinct MI60 (32GB HBM2) GPU. Learn the best settings for faster video generation, higher output quality, and efficient model downloads using aria2c. 💻 Watch the full long-form tutorial: https://youtube.com/live/qC9KNSxZ3GM 📝 Read the detailed blog review and setup guide: https://www.oja...

Are you still relying on hefty JavaScript libraries for basic form validation? Stop! This episode reveals the powerful secret weapon: the HTML5 pattern attribute, powered by Regular Expressions (REGEX). We walk through how to enforce complex rules for usernames and strong passwords using pure HTML and simple CSS for beautiful, real-time user feedback. Get ready to clean up your codebase! Links: Fu...

In this tutorial, we dive deep into optimizing stable-diffusion.cpp on Linux using the AMD Instinct Mi60 32GB HBM2 GPU. Learn how to boost image generation speed and quality for AI-driven visuals and videos. We focus on setting up the environment for maximum performance, allowing you to create AI outputs faster and with better results. Full tutorial video: https://youtube.com/live/msG28ynxsNM Blog...

Tired of repetitive 3D modeling tasks? This episode breaks down a powerful workflow for automating the creation of 3D assets using the Blender Python API. Listen as we explain how a single command line script can generate a photorealistic, golden cup trophy, apply PBR materials, optimize the mesh, and export it as a highly efficient .glb file ready for the web. This is the future of 3D content gen...

In this episode, we explore how to fine-tune the Web UI for Llama.cpp running on Linux with AMD Instinct Mi60 GPU to prevent clipping and improve chat quality. We walk through the process of setting up Llama-2-7b-chat and DeepSeek-R1-32B, and we also examine stable-diffusion.cpp as an alternative to ComfyUI for smoother AI workflows. If you're working with powerful models like these, this tuto...

Looking to take back control of your source code? This episode breaks down the simplest, fastest, and most resource-friendly way to host your own Git service using Gogs and Podman-Compose. We discuss why Gogs is an excellent open source alternative to GitHub or GitLab for personal projects or small teams. Learn about the necessary container definitions, the initial setup wizard, and the essential...

In this episode, we dive into the essentials of optimizing local AI setups. Learn how to select the best backend to make the most out of your system's specifications—whether you're focusing on performance, memory usage, or quality. If you're running AI models locally, this tutorial is for you! Catch the full video and blog for even more insights: Blog: https://www.ojambo.com/review-gen...

In this episode, discover why Lite XL 2.1.8 is becoming the favorite lightweight code editor for developers who value speed, simplicity, and open-source freedom. I explain how to install it on Fedora Linux, explore its best features, and share tips for customizing it for your workflow. 📄 Full blog article: https://ojambo.com/lite-xl-2-1-8-advanced-editor-review 🎥 Watch the full in-depth screenca...

This is the audio breakdown of our highly-requested GPU benchmark: Ollama versus Llama.cpp for local Large Language Model (LLM) inference on the powerful AMD Instinct MI60. We're testing a single, massive 70-billion-parameter model to measure which framework delivers the absolute highest tokens per second (t/s). The results are not what you expect! If you're running local AI or looking at...

Learn how to install and run Mattermost, the open source alternative to Slack and Microsoft Teams, using Podman or Podman Compose. This tutorial walks you through creating the configuration file, launching containers, and accessing your self-hosted chat app in minutes. Watch the full long-form video here: https://youtube.com/live/0yI8RcOqHkQ Read the full blog article here: https://ojambo.com/how-...

Want to run a powerful Large Language Model (LLM) privately on your machine? We discuss how to move the DeepSeek-R1 32B model from the command line to a sleek, custom web interface on Fedora Linux. Learn about the Python logic and web components needed to connect to the local Ollama API. Resources: Full Tutorial & Code: https://ojambo.com/web-ui-for-generative-ai-deepseek-r1-32b-model Video Wa...

Did you use my popular AJAX Video Upload tutorial but needed full file management? This episode is the essential update! We dive into the code to add the missing Replace (Update) and Delete (Destroy) actions for a production-ready PHP and MariaDB video uploader. Learn how to securely manage file system cleanup and database record updates using a single, efficient PHP endpoint. Links & Resource...

Ready to secure your web applications? This is the essential backend episode! We detail how to use PHP to process data from the Fetch API, implement a bulletproof strategy for saving video files to a secure folder outside the web root, and manage metadata efficiently in a MariaDB database. Listen to learn the professional workflow for handling CRUD operations, including safely deleting old files t...

Dive into the client-side magic of video uploading! In this audio breakdown, we discuss how to use HTML5, JavaScript, and the Canvas API to create a stunning user interface (UI). Learn the core logic behind implementing a smooth drag-and-drop file zone, giving users real-time control over video resizing, and letting them "scrub" to capture the perfect placeholder image. This is Part 1, s...

Stop waiting hours for your video exports, especially on Linux! This episode is all about the specific H.264 and AAC settings you need to master in Blender's Video Sequence Editor (VSE) to get faster, higher-quality, and more compatible final renders for platforms like YouTube and Odysee. We break down the magic numbers: Keyframe Interval (GOP) of 120, Max B-frames of 6, and a rock-solid 192k...

Level up your creative automation! 🚀 Learn how to use the Blender Python API to automatically model and animate a flickering 3D candle, then instantly deploy it to any website using the <model-viewer> component. This episode breaks down the script, the command-line execution (running Blender without the GUI!), and the simple HTML required for web deployment. If you want to automate your 3D...

In this episode, I explain why I switched from Docker to Podman and Podman-Compose — and why you might want to as well. Discover how Podman’s rootless design, Dockerfile compatibility, and easy integration with Jenkins make it a modern, secure alternative for containerized development. Watch the full video on YouTube: https://youtube.com/live/8hX4RnvWI08 Read the related blog article: https://www....

In this episode, I walk you through setting up Jenkins using Podman containers for a seamless continuous integration and deployment pipeline. Learn how to automate your development workflow with Jenkins, Podman, and containerization in minutes! For a full step-by-step guide, check out my blog post and full video tutorial: Blog Post: https://ojambo.com/getting-started-with-jenkins-and-podman Full V...