This is Part 1 of my DIGITAL INDEPENDENCE Series. I’m documenting the tools I build, the services I replace, and the lessons I learn as I work toward greater digital independence. Follow along and discover ideas you can put to use yourself.
I Don’t Want to Leave the Cloud. I Just Want to Be Able To.
A few days ago, Google flagged a spreadsheet I’d kept in Drive for more than a decade.
It was called wish list. Nothing particularly interesting: just an old collection of product links I’d saved over the years. I hadn’t updated it in about two years.
Google’s automated systems apparently decided it might violate the company’s phishing policy. I could still open the file myself, and I wasn’t planning to share it, so the restriction didn’t actually interfere with my plans. But I requested a review anyway.
To Google’s credit, the decision was reversed remarkably quickly. An automated review concluded that the spreadsheet did not violate its terms.
That should have been the end of it. Instead, it made me think about how much of my digital life depends on decisions I don’t control.
It’s Not Really About One Spreadsheet
I’ve spent more than twenty years collecting digital information: journals, photographs, videos, business records, research, ideas, and notes from projects I finished—or never quite started. There are things in those archives I probably don’t even remember saving.
Much of it lives in services run by other companies. That’s convenient. I can open a file from my phone, search an old document, or pick up work on another computer. I don’t want to give that up.
But I do wonder: What would happen if I woke up tomorrow and couldn’t get into my Google account?
I’m not suggesting that one mistakenly flagged file means my whole account is in danger. The review process worked in this case. Still, the incident reminded me that an automated decision can be wrong, and that I have limited visibility into how those decisions are made.
I’m not especially bothered by security scanning itself. I understand why providers need to detect abuse. My concern is whether my private information is used in ways I haven’t agreed to—and whether a mistake could ever cut me off from records I care about.
What I Mean by Digital Independence
For me, digital independence doesn’t mean refusing to use Google, Microsoft, or every other cloud service. It means being able to leave one without losing my work or my history.
It means having copies of my files that I can actually restore. It means knowing where my information is stored, who can access it, and what happens if a provider changes its policies or closes my account.
I’m already building a personal workspace called ModernOS, where I can bring together projects, tasks, publishing records, notes, and other parts of my work. The more I use it, the more I find myself wanting the underlying data to be just as portable as the interface.
A Small Server, Not a Giant Data Center
I’ve been exploring the idea of running my own infrastructure on a small, quiet computer. It could host ModernOS, Nextcloud for file access and synchronization, and other Linux applications I want to use.
I’d like a machine with enough memory to run several virtual machines, but I don’t need to buy it tomorrow. I’d rather learn what I need and save for hardware that can grow with the project.
Eventually, I’d love to keep that server in a small private office where I could work during the day. For now, a machine at home would be a practical place to begin.
Three Copies, Not One Point of Failure
Owning the server wouldn’t solve everything. A drive can fail. A machine can be stolen. Files can be deleted accidentally. A fire or power problem can take out equipment I own just as easily as equipment I rent.
So the plan I’m considering has three layers:
- A primary server I control, running my applications and holding my working files.
- Encrypted, automatic backups at Hetzner, with the backup encrypted before it leaves my machine and the recovery keys kept by me.
- A separate external drive, updated weekly, disconnected afterward, and ideally stored somewhere away from the server.
Hetzner’s German location appeals to me partly because of Europe’s privacy protections. But geography alone isn’t a guarantee. A hosting provider can have outages, legal obligations, or account restrictions. Encryption and independent copies matter more than trusting any single company.
And a backup isn’t truly useful just because software says it completed successfully. I want to practice restoring files and databases onto another machine, so I know the process works before I need it.
Making the Archive Useful Again
There’s another possibility that excites me almost as much as protecting the data: being able to explore it.
I’ve kept journals and accumulated research for decades. I can search individual files, but it’s much harder to connect ideas across years of notes and projects.
I’d love to build a private, ChatGPT-like assistant that could answer questions such as:
- What was I working on during the summer of 2012?
- What ideas did I have about affiliate marketing ten or fifteen years ago?
- Which projects did I start but never finish?
- Where did I write about a particular person, place, or event?
The approach I’m exploring is called retrieval-augmented generation, or RAG. Rather than training a model on my entire archive, I could create a searchable index. When I ask a question, the system would find relevant passages and provide them to a locally running language model.
Ideally, it would also point me back to the original documents. I don’t want an assistant that confidently invents memories. I want one that helps me rediscover what I actually wrote.
I’ve started thinking of this as ModernOS Intelligence: a private research window into my own history and work.
Starting Small
None of this needs to happen at once. I can begin by testing a small local AI model on the laptop I already own. I can index a limited set of documents, see what works, and learn where the limits are.
Then I can make a more informed decision about the server, storage, and software. I can move my archives gradually instead of attempting one enormous migration.
I’m not trying to build the perfect system overnight. I’m trying to reduce the number of things that could go wrong all at once.
The Freedom to Leave
In the end, the spreadsheet was restored. No files were lost, and I still use Google Drive.
But the experience clarified something for me. I don’t need to distrust every technology company to recognize the value of keeping independent copies of my own information.
My journals, research, photographs, and business records represent a significant part of my life. I want to preserve them, understand them, and keep using them for years to come.
I don’t want to leave the cloud. I just want to know that I can.
Discover more from JohnHigginbotham.com
Subscribe to get the latest posts sent to your email.