Personal Data Hub
This summary page explains what the Personal Data Hub is and how the different pages fit together.
Personal Data Hub series
- Personal Data Hub - You are here
- Foundations
- Data pipelines
- QuickAdd scripts
- Chrome extensions
- Plugins
What this is
I am rubbish at keeping a journal. After years of buying notebooks, downloading apps, and telling myself "this time will be different", I finally admitted it. I am simply not built that way.
At the end of each year I create a year-in-review mindmap. They always take weeks to complete. I find them incredibly valuable though. Connections appear that I had forgotten. Insights emerge that I didn't notice at the time. Patterns reveal themselves only when you see everything laid out on a single canvas.
This year I sat down to create my 2025 review. I had captured zero information. Not a single note about the books I read, the shows I watched, the runs I completed, or how my health tracked across the year.
I needed a process that me, an evidently lazy person when it comes to journaling, could actually follow. Not a system that required daily discipline. Not another app promising to change my habits. A system that worked with my laziness, not against it.
The four criteria
-
Own your data. All information must not be locked into any application or cloud service. I need to own my data in a format I can read without any tool, and move it whenever I want.
-
Don't reinvent the wheel. If something already exists, test it and adopt it if appropriate.
-
Import existing data. Must be able to import my existing data from many different sources and represent it as a cohesive set.
-
Visualise when ready. Only visualise for year-in-review once the majority of data is imported. Not before.
Why Obsidian
My direction was influenced by Steph Ango, the CEO of Obsidian, specifically how he uses his own vault. I had already been using Obsidian in my workflow to create The Muppit Sphere and my private sites, but there was much more to it.
Obsidian stores everything as plain markdown files with no database and no cloud lock-in. Just text files I own and can read with any editor. Long live markdown.
What I built
What started as a simple journaling system for lazy people became a personal data hub. Books, shows, health, fitness, and more in a single queryable knowledge base.
| Domain | Features |
|---|---|
| Books | Google Books API with Apple fallback, visual picker with covers, AI summaries via Ollama, batch CSV import |
| Shows | TMDB integration, movie and series notes, episode tracking, watch counts |
| Health | Daily metrics import (steps, HRV, sleep, weight), incremental updates |
| Fitness | Workout tracking with duration, calories, workout type linking |
| Calendar | Custom plugin with month, week, and list views, colour-coded events |
All running from any of my devices. All stored in plain text files I own forever with every change tracked and backed up locally.
When I need AI assistance, I use local models via Ollama running on my Kubernetes cluster. My reading habits, health data, and watch history stay on my infrastructure. No cloud APIs, no data leaving my network. This is private information and I intend to keep it that way.
The numbers
- 4 stages: Books, Shows, Health, Fitness
- 2 Chrome extensions: Prime, Netflix
- 1 custom plugin: Calendar View
- 3 community plugins: Git, QuickAdd, Local REST API
- 100% iOS compatible
What surprised me was how little I needed. Three community plugins and one I built myself. The rest is just JavaScript running inside Obsidian.
There is also a trust issue with community plugins. I do not know what data they collect or where they send it. When I build my own scripts and plugins, the intent is mine and clear because I created it.
How these pages are organised
Each section follows a similar structure:
- Context. Why this piece exists and what problem it solves.
- How it works. Technical explanation with diagrams where helpful.
- Step-by-step. Numbered instructions for reproduction.
- Configuration. Code blocks with settings.
- Troubleshooting. Common issues and solutions.
Where this fits in The Muppit Sphere
The Personal Data Hub is one of the application examples in The Muppit Sphere. It shows how I apply:
- Local-first thinking to personal data management.
- Muppitify principles around purpose, modularity, and automation.
- Custom tooling when existing solutions fall short.
Please use it as a working example when you want to:
- Build your own personal knowledge management system.
- Import scattered data from multiple sources into one place.
- Create custom Obsidian scripts and plugins for automation.
Source code
All scripts, extensions, and plugins are available on GitHub: