For some time, I've been grappling with the ever-growing stack of important papers that seems to be a natural part of getting older. Living in Germany, one quickly accumulates a significant collection of documents, some of which may be needed decades later. Having always been more organized digitally than in the physical world, I've been searching for the best method to digitize and organize all my documents.
Paperless
My first discovery was the paperless-ngx project, a document management system (DMS) specifically designed to facilitate the digitization of documents or transitioning to a "paperless" lifestyle. It features OCR (optical character recognition) for PDFs and images, allows for tagging, and most importantly, enables full-text search across all your documents 🙌 Being open-source and requiring self-hosting was ideal for me, given my interest in homelabbing and self-hosting, while having my own server and a few Raspberry Pis in my living room. The workflow was straightforward: whenever I received a new document, I scanned it using Genius Scan, which automatically exported the PDFs to my home server via FTP and simultaneously to my cloud provider via WebDAV.
This setup worked wonderfully and I would highly recommend it to those serious about self-hosting and seeking a robust system for managing their documents. However, I encountered a downside. My home server is more of an experimental playground where I frequently try new things, often leading to setups being reset or reverting to backups. This meant regularly backing up the paperless-ngx project or setting it up anew. It was a bit too cumbersome for my needs and my home server's somewhat erratic and unprofessional architecture for storing many important documents.
Johnny.Decimal and Obsidian
However, I didn't want to revert to a paper-based system, so I looked for another solution. I needed something that could operate on top of the filesystem, allowing me to utilize my cloud provider and its synchronization feature. As I was scanning my documents, they were not only exported to my homelab but also to my cloud provider. This resulted in a large collection of unsorted scanned documents. I needed a well-structured system to organize them, leading me to the Johnny.Decimal system. I believe I first learned about it on Reddit and was immediately intrigued. The Johnny.Decimal system organizes data into ten broad areas, each subdivided into ten categories, accommodating up to 100 categories in total. Each category is assigned a two-digit number, and items within these categories are given a unique ID in the format: [category number].[item ID]. This method simplifies finding and managing documents, minimizes duplication, and improves data organization, offering a clear, numbered structure for easy access. This system worked excellently for me, and I had all my documents synced across all my devices.
However, I found myself missing features like full-text search across my documents. This is where Obsidian and its plugins came into play. Not only does it allow me to enhance my PDF collection with Markdown notes, grouping them and creating links, but it also supports a wide range of community plugins. The omnisearch plugin allows you to search through the content of files, which solves the last big feature I was missing with the current system. And with that the system was complete.
And last, for my fellow coders: Their is also the doubleshift plugin which lets you open the omisearch with a double press to the shift key, which makes it feel just like home ;)
#blog #homelab #server