Miro Adamy

In July 2023 I sat down to list every place my files lived. The list had eight entries - seven online storages and one line for “external volumes” - and at the bottom a checkbox: “Finish this list”. It stayed unchecked for three years, which tells you everything about the problem. In the earlier post about reorganizing my digital life I described the big picture: the Data Map, the disk catalog, the new NAS. This one zooms in on the clouds, one by one.

Consolidating a digital life onto one NAS has an ugly side effect: everything now burns in the same fire. In the previous posts I described how I moved decades of files, photos and documents onto a single Synology box at home. This post is about the second half of that story - getting a copy of it out of the house, encrypted, in a way that lets me sleep. Picking the provider # The first lesson was that where a company keeps its servers and where it is registered are two different things. Backblaze has European data centers, but it is a Californian company, and the US CLOUD Act follows the company, not the building. For a family archive I wanted a European company in a European data center.

Seven devices, two of them NAS boxes, and a laptop that travels with me. For years, reaching any of them from outside the house was a small project every time. In July 2026 I put all of them on one private network, and the day every machine got one name that works from anywhere, half of my runbooks got shorter. Why a mesh VPN # My home fleet is not big, but it is spread out: two MacBooks, an iPhone, an iPad, a Mac mini that is always on and serves as a small home hub, and two Synology NAS boxes, an old one and a new one.

Last weekend 25th to 27th of September, members of Wikimedia Slovensko met in Dubnica nad Váhom. It was great to meet the fellow Wikipedians in the flesh again (Matej, Patrik and Filip), and to meet Kormi in 3D for the first time. It was mainly a planning, coordination and brainstorming meeting, but part of the programme happened outdoors, with cameras. By Sunday the town’s 18th-century Calvary had its own photos on Wikimedia Commons, a record on Wikidata and the start of a Slovak Wikipedia article. The local Jewish cemetery gained a set of new photographs too. That weekend is a good small example of what this whole thing is about.

I have always been a pack rat when it came to digital assets. I didn’t like the idea of deleting files. What if I want to come back to them later? After all, they don’t really take almost any physical space compared to paper books (certainly not now, in the age of multi-terabyte hard drives and the cloud). So what I did - I kept archiving them. It started in the late ’80s and the ’90s with floppy disks, eventually migrated to external hard drives, CDs, DVDs and with the ascent of the cloud they ended up in a variety of various storage solutions.

Plain markdown files turned out to be the best possible interface between a decade of notes and a language model. No API, no export, no connector, no sync service in the middle: the agent opens the folder and reads it, the same way I do. This post is what happened once I let it. The setup # There is a CLAUDE.md file at the root of my vault. It is twenty-nine lines long, and the last three of them say this:

This is a history of three years - from bookmarking ChatGPT articles to an AI-managed knowledge base; with a certification in the middle and one embarrassing self-assessment along the way. Spectator to User (2023) # I still remember my first interactions with ChatGPT (via the web, using the free version) sometime in early 2023. I wanted to have an episode guide of an excellent HBO series called Patria, a story of a Basque family during the ETA years. So I asked for it, and what ChatGPT produced was completely, utterly, unbelievably hallucinated: 12 episodes (the series has 8), a made-up story, randomly assigned actors, two thirds of them non-existent. One huge hallucination. I did not keep the transcript, but I remember it clearly. It was very disappointing, and my conclusion was: well, this is fun, but useless.

The pitch I once wrote for a conference talk still says it best: Obsidian has a cult following “not because it’s clever software, but because of what it isn’t: your notes are just plain markdown files in a folder on your computer. No database, no cloud lock-in, no proprietary format. Just files.” Nothing ages as well as plain text; it has been the same since, well, files were invented. Markdown is just a very thin layer that gives the plain text some structure, and allows it to be rendered and displayed in an aesthetically pleasing way.

I have always taken notes. At first on paper - there are still many handwritten notebooks somewhere in the archives - but for the last thirty years or so most of them, if not all, have been digital. And digital always means an application and a format. I have tried a lot of them. While doing so I made pretty much every mistake one can make when it comes to choosing an application and settling on a workflow. What follows is a tour of the graveyard: a long line of tools, most of them now dead, and the lesson each one charged me on the way out.

My company, Klassify s.r.o., has had a domain for a long time before it was even registered (I registered klassify.com way back in early 2000 in Canada, but never actually did anything with it) - and a web page since this afternoon. For all that time klassify.com carried only my email - there was no A record at all, nothing was served, and whoever looked up the company from an invoice or a contract got a browser error and, I suppose, some doubts about whether the company exists. So here is how it got fixed in one afternoon, in the order it happened and with the reasoning. Nothing here is clever or sophisticated or the only/the best way. The point is the sequence of creation: there are two places in it where it is quite easy to break something that matters more than a web page.