SelfPrivacy is in “Open beta” and promises to make setup and use of email, messager, password management, video chat and other services simple by leveraging the likes of Hetzner, Cloudflare, and Backblaze.
I stumbled on the app while browsing the F-droid app “store” and had never heard of them. I think the proposition is neat and while I’m comfortable hosting most of these services myself, my curiosity has been piqued. Searching for it elsewhere on the web as far as privacy rating, reviews, etc has left me empty handed. I dont’ know if they’re just too new or not. So I’m curious if anyone has tried them out or looked into it further.
You can do all of that on your own.
OR, you can create a single attack vector that can potentially be exploited and put everything at risk, at the same time.
If you’ve ever worked in, or adjacent to, IT, then you’ve heard the phase “single pane of glass”, meaning you can manage all your infrastructure, or IOT, through a single terminal/UI.
This is basically a single pane of glass that you’re getting through a side loaded repo, to manage your entire digital life. That means it can also become a single pane of glass for anyone able to exploit that application i.e. supply chain attack, phone AND/OR app specific vulnerabilities, etc.
Would the Freedombox fit that description?
Not really, sort of, but different threat models IMO.
The app this thread was about is asking to become a single pane for external services e.g. cloud, which is why it requires your Hertzner API.
For the following, I’m reaching into my memory hole, so definitely check elsewhere to confirm before doing anything.
FreedomBox, if I recall, is basically Debian Linux with a variety of self-hosted tools that are easily configurable e.g. Media servers, torrents, NextCloud, etc. It’s been around for a while and I don’t recall ever hearing anything bad about the project.
Ultimately, sure, you’re still trusting the maintainers to some degree, like with any distro/spin, but that’s a judgement you’ll have to make for yourself.
If you’re going to use Freedom box for all of your most critical and private parts of your digital life, then you should probably weigh the risks more heavily, than if you’re just going to make it a media and torrent box.
I guess my question was geared more towards what would happen if you chose to host the entire box on a VPS server somewhere, or heck, even had it at home from a publicly accessible IP address / port. At that point, wouldn’t it essentially be the same “single pane of glass” setup as you describe above?
I guess when it comes to trusting the distribution maintainers, I take that for granted… After all, if I didn’t, at that point one would have to write off a ton of Linux distributions entirely for server usage… I think. If I understand the argument correctly, anyway.
Anything with “Freedom” in its name sounds sketchy.
I think that software package is an exception to the rule, but I see where you’re coming from.
The other side of that being Security through Obscurity.
If you’re not running all your stuff through a major well-known host like Google or Amazon you’re less likely to be a target than if you’re just self-hosting.
Supposedly Google and Amazon have “good” security, but they still get hacked.
I tried them a couple days ago, got to setting up Hetzner API, had my account rejected a bunch of times, found out Hetzner team is infamous for rejecting new accounts and cancelling old accounts by the whims of their ‘protection systems’, realized the only other hosting option supported by SelfPrivacy is Digital Ocean, noped out of it all
They reject new accounts because they are really cheap and reliable so people abuse their system.
It’s kind of ironic that in order to spin up a bunch of privacy centric services, you need to hand your identity over to a company that will be hosting your data and will, more likely than not, be able to look at its contents in their unencrypted form. I always found this a bit discomforting.
I get the reason behind it, and support it too, but it doesn’t make a good impression when your account gets rejected despite every information being correct just because you signed up using a VPN (I can’t verify that VPN is the reason, but it has been suggested elsewhere to be a cause for suspicion on their part).
Yes but if you think about it, you are not the only one trying to register with the same IP with hetzner that’s why the system triggers the ban.
That wouldn’t make sense either, because the user literally has to provide them all kinds of personal information in order to register. And no matter which IP address is being used to register, the user still has to pay to even use their service. So rejecting accounts simply because the registration was done via VPN is, in the best case scenario, overkill.
Don’t get me wrong though, I have nothing against them; I just don’t think their anti-spam measures are anywhere as good as they need to be, and their responses towards people complaining about them indicate that they wouldn’t bother trying to make it better.
Yes you are right their antispam measure is very aggressive but I think plenty of bad actors have abused the service before that’s why its like that.
