Seeking Freedom from Cloud Lock-in: A PKM Community Discussion

daviddelven

Fellow PKM enthusiasts,

I find myself caught in what seems to be a cyclical dilemma in the world of personal knowledge management, and I'm curious if others are experiencing the same.

My journey began with enthusiasm for local-first environments like Obsidian and Logseq. The promise of data ownership and flexibility was alluring. However, as time passed, I found myself gravitating back to cloud-based solutions.

Now, I'm feeling the pull towards local files and Markdown again. It's as if I'm caught in a perpetual swing between the convenience of the cloud and the freedom of local storage.

This pattern has me wondering:

• Is this back-and-forth between local-first and cloud-based systems common in the PKM community?
• For those who've made multiple switches, what ultimately made you settle on one approach (if you have)?
• Are there solutions that effectively bridge this gap, offering the best of both worlds?
• How do others manage the trade-offs between sync convenience and data sovereignty?
• I'm starting to think this isn't just a fleeting concern, but a fundamental tension in how we approach PKM. The desire for true ownership of our data and the freedom to use multiple tools seems to persist, even after trying various solutions.

Has anyone else found themselves in this PKM pendulum? How have you reconciled these competing needs in your own systems?

andang76

I don't have your doubts :-)
Mi vision is very clear.
I have my first disappointing situation many years ago, I remember near the 2000, when I had my personal address book on a service online, and one day I lose all.
It was a silly harm (a bunch of email address...), but this episode signed my general relationship with online and cloud services.
When I can, I want my data on my pc. Under my total control. No doubts.
If I lose something, this happens for my fault.

This is even more relevant with my note system. It's a work of almost two years

KillerWhale

No tension either. I'm using apps that either allow direct file access or clean export in Markdown that can be imported elsewhere, with privacy-first cloud systems. I do not consider any system that does not offer those safeguards.

zettelsan

I’ve settled in a specific computing environment for a long time (Linux, Emacs, etc.), so from the outset I limit myself to options available in that niche. These days, however, being purely on Linux isn’t much of a limitation than it used to be, due to the maturity of cloud solutions and the availability of cross-platform environment for apps.

The data portability/exportability is very important, since you never know how long the (proprietary) service of your choice last. So using a simple, portable data format like Markdown is a must. Over the years, I’ve used a few different ways to keep my lasting notes, and it always has helped that those data were in a convertible format. Text, HTML, XML, Markdown, Org, etc. Though not fully and often without some loss, they can be converted to each other, so my notes have survived a few iterations of platform switch over the years.

In my opinion, the local vs. cloud really depends on how much you trust existing cloud solutions for your needs. With the cloud solutions, you give up privacy, security, and the (full) control over your data, but you gain in releasing you from responsibility of data redundancy, data available over the network for sharing, data backup, etc. Personally, if my data are neither sensitive nor important, I wouldn’t mind using the cloud solutions even just for convenience. And, unfortunately, I’m not that important, and so are my data; if the cloud solutions help me keep things sufficiently simple to maintain, I am willing to sacrifice a bit of my inflated sense of self-worth.

With all that said, my setup is mostly local-based (Emacs, Org Roam), though I delegate data storage to cloud-based solutions right now (GitHub and AWS for Syncthing). I could make all my stuff run fully locally, but I then have to invest my time on putting all that stuff together. Is that effort worth my time? That’s a factor to consider in addition to data ownership and the freedom to choose (client) solutions.

GeoEng51

I suspect you can have your cake and eat it too, in this area.

I'm working with two apps: The Archive and Logseq.

I store the data files for The Archive in a Dropbox folder, which syncs frequently when changes are made. Sometimes I access those files using 1Writer or iA Writer from my iPhone, but for the most part, the access and editing is done on a Mac. I have never had a problem with the Dropbox service, by the way, with over 20 years of experience on both personal and work files. That doesn't mean it couldn't happen, but Dropbox has proved its reliability (to me).

I store the data files for Logseq in the Documents folder on my Mac. Logseq has its own sync service, which I use because it gives easy access on my iPhone. The files are on my personal computer and in the Apple cloud, and I suppose there is some vulnerability to data loss. Again, I haven't experienced that but it doesn't mean Apple is perfect.

This might sound like I am at the mercy of several cloud-based services, at least at a "low risk" level, but my backup plan is using Time Machine on the Mac, which saves both data sets hourly onto an external hard drive. So if I ran into any problems with the cloud-base services, I have the backup files in Time Machine from which to restore a previous state of the data.

My "deep" backup plan is to make a separate copy of both sets of data to a second hard drive about once a week (manually). While this is not that frequent, it is enough to recover from a hard drive failure if that also occurs. If I was processing more data, I would do this deep backup more frequently.