Skip to main content

Posts

Create a react project from scratch - the template for a new start

How to Create a React Project on Ubuntu 24.04 How to Create a React Project on Ubuntu 24.04 Step-by-Step Guide 1. Create a Folder for the Project Start by creating a dedicated folder where your project will reside. Open a terminal and run: mkdir react-project && cd react-project 2. Check if Node.js and npm Are Installed Node.js and npm are essential for creating React projects. Verify their presence with: node -v npm -v If they aren’t installed, use nvm (Node Version Manager) for easy installation: curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.4/install.sh | bash source ~/.bashrc nvm install --lts After installing, verify the installation again with the commands above. Learn more about nvm here. 3. Use Vite to Create the Project Vite is a fast build tool optimized for modern web apps. T...
Recent posts

Making an AppImage discoverable in ubuntu and configuring URL handling in Firefox

OS: Ubuntu 24.04 Using software in AppImage format can lead to Ubuntu not recognizing it properly when handling links. Here’s how to configure Ubuntu to manage an AppImage application seamlessly. We'll use Obsidian as an example. Step 1: Creating a .desktop File To make the Obsidian AppImage discoverable by the system, create a .desktop file—a shortcut that defines how the application should be launched. Here’s the content for the file: [Desktop Entry] Exec=/path_to_the_appimage/Obsidian-1.6.5.AppImage --no-sandbox %U Name=Obsidian Terminal=false Type=Application MimeType=x-scheme-handler/obsidian; Explanation of Key Entries: Exec : The command to run the AppImage, with --no-sandbox to avoid sandboxing errors. Name : The display name of the app. Terminal : Set to false to avoid opening a terminal window. Type : Identifies this as an application. MimeType : Registers obsidian:// links to be handled by this application. After creating t...

Learning for the future

While we spend a great part of our professional life learning, little time or thought is spent improving the learning process in itself. I think it's relevant to separate the learning process in these 2 pathways: Preparations for examinations, such as certifications Preparation for real life work, adapting previous knowledge and acquite new knowledge to specific circumstances of our work Learning is by itself a very personal and lonely process. It helps having a structure, an obvious path to follow, and that is often the case when learning in order to obtain a certification. On the other hand, that structure is non-existent when we have to learn for real life work, when we have to create are own path for an unique goal.  To circumvert the negative aspects of learning we must use well-known evidence-based learning techniques with the right technology - better and cheaper. An adaptative learning system should be created. Adaptative in the sense that it adapts to the learner and in th...

A workflow to support personal learning from raw sources

 Unlocking Efficient Learning with AI-Powered Flashcards in Obsidian As learners, we're constantly seeking ways to optimize our learning workflow and process new information more effectively. In this post, I'll explore a powerful combination of tools that can help you achieve just that: Obsidian and the SystemSculpt and the Spaced Repetition plugins. Together, these tools can help you generate flashcards that support a structured learning workflow, making it easier to create knowledge from your source materials.   The Power of Obsidian Obsidian allows you to organize your thoughts and ideas in a flexible and customizable way. Its unique features, such as tags, folders, and links, enable you to create a knowledge graph that connects your notes and ideas. In this workflow, I'll be using Obsidian as the central hub for my learning process.   SystemSculpt and Spaced Repetition: The AI-Powered Flashcard Generators SystemSculpt is an AI-powered plugin that integrates seamless...

Ubuntu & Ollama & Obsidian: Your Local AI Powerhouse

My pc runs on Ubuntu 24.04 and one of my favorite things to do in my pc is take notes and annotations using Obsidian . Even though I have a modest pc with a low end graphics card, instead of using a paid cloud AI service, I run ollama with local LLMs - it might be slow but is nevertheless useful. But then I ran into a problem when trying to use obsidian plugins that run AI on ollama local LLMs. After running   ollama serve in the terminal, the following error would show up: Error: listen tcp 127.0.0.1:11434: bind: address already in use I tried some tweaking here and there, like changing the port address in the command line, but it didn't work as I needed it to. Turns out the solution is simpler than I thought. Here is how it works. When you install ollama on ubuntu 24.04 it will install itself as a systemd service. Ollama runs in the background Ubuntu 24.04 uses systemd, so Ollama automatically starts as a service when the pc starts. Think of it like a silent workhorse, always ...