All said and done… people should have personally controlled access to their data. For physical things, some people have safes, others use safety deposit boxes at banks. But we don’t have a digital equivalent. And the problem is that the complexity is too high for a lot of people. So something like this would be good for some people, it still won’t get the majority. What we need is the digital equivalent of a fiduciary. Someone who is legally bound to look out for a person’s digital interests. That would allow people to trust such a person to vet simpler wrappers around set ups like this, or anything.
Most peoples’ computers and phones have encrypted data storage by default now.
They are the digital equivalent of a safe.But most people do not want to own, manage, back up and store their own data.
Just like most people do not own or think they need a safe.
More things could be just files
Congratulations, you’ve invented Unix.
It feels like the author could become a Plan9 fan in no time.
The only detail really is that at least 2 of the N machines you are using have to be on at the time so where ever a change was made is synced to another machine that is on and this continues so that you never end up booting a machine to use when nothing else with the latest files is available. This is where having a centralised low power machine is valuable and saves having a desktop or a laptop on when it doesn’t need to be.
I really wish the desktop version of the world had not become so marginalised as local programs are considerably better to use than websites, they are so much quicker, accessible and easier to use.
That centralized low-power machine can even be your phone, if it has enough storage for your needs.
This is exactly how I use Syncthing, and as the author says, it sure would be nice if more things were just files. Really, most things are stored locally as files, but not always in a way that plays nice with syncing. Like, I can sync my Firefox profile between machines (it’s all in one folder), but I found it prone to conflicts, with little to resolve those conflicts.
In a similar vein, local-first apps built with Conflict-free replicated data types (CRDTs) can be another way to avoid server dependency. I haven’t seen any significant apps built this way yet (just occasional blog posts about it). I imagine the CRDT approach would work better for individual apps, since conflict resolution can be written in a way that works best for a given app, but I also imagine that such apps would not play nicely with a generic sync solution like Syncthing.
You just need to sync the entire os install. /S
This is a vision I could get behind, and I already do a ton of self-hosting for that stuff! The theoretical simplification, easier conceptual understanding, and easier portability would all be very nice plusses over what we have now.
Buy it does feel a bit to me like creating just one more perfect standard, and then we will have our fracultured world, but with 14 standards…
I mean, Syncthing is much more than that. The great thing about it is that it works no matter where you are - home wifi, over the internet etc.
But that means that someone else’s server is used whenever you leave your home network.
But that means that someone else’s server is used whenever you leave your home network.
I’m pretty sure syncthing does NAT hole punching, so someone else’s server is only used for initial connection, after that, your data goes directly to your devices.
Your detail is correct, but I feel like the point is - it would not work if there would be no server





