One doesn’t need to trust it. Its open source software you can conduct the tests yourself if you know what you are doing. Regardless I tossed it in to point out that we trust and think things are going what we ask. That’s a false sense of security. Never think that until you’ve actually proven it is the way I operate on Linux in particular. I guess its something you pick up when you develop software
I dont use flatpak, appimage, and snap. Im currently moving away from docker. In my view the tools needed for app isolation are there. You have firejail, firewalld, selinux (or apparmor), cgroups etc. Setting it up native feels cleaner, simpler and less prone to breakage than overriding system library in a sandbox. Thats just my approach though. In no way should one feel obligated to feel the same way
I can review all of flatpak myself? I wish there was a standard trusted 3rd party that did this with software to check claims. It could be a great go-fund-me-like platform… I would actually consider making this, and way before I consider sending a month reviewing a code base just to see if I want to run it or trust a hard-to-verify claim like “sandbox” (which there’s plenty of contact aka “surface area” between something like flatpak and the OS).
UPDATE: confirmed it’s perfectly accessible through the web browser, also running into the same issue using Chrome on another device (not previously configured)
Thanks for all the pointers. I’m just finding one unexpected behavior, I’m trying to connect to my self-hosted bitwarden container, and it’s giving me a “NetworkError when attempting to reach resource”. In the login options I have the exact same thing as chrome, where it’s working flawlessly.
Any pointer of which of the settings or extensions may be interfering?
Privacy related, you may look into the kind of addons or *monkey scripts that automatically redirect you to an alternative in case you click a link to a variety of big offender websites.
Use it, it’s pretty good. Even available on Windows now. It works like Firefox, but was hardened, so it is mostly harder to fingerprint. Mullvad is another good contender. The following is a good overview of the features each offers and I 'll link it again.
I don’t think that doable anymore. You can maintain a fork of Firefox or Chromium with a few people but developing an entire browser probably not. Browsers need to support like a million different standards for displaying all the different website code, videos, audio and interactive menus all while they have to be efficient (i.e. not a resource hog) at doing that and most importantly they have to be secure. The last part is paramount since the browser is basically the one piece of software on your computer whose main tasks it is to execute other peoples untrusted code on your computer.
The collective amount of man hours spent on coding and auditing code for web browsers and all the underlying components has probably overtaken the linux kernel long ago.
You probably need to spend hundreds of USD billions at this point if you want a browser that can also do what Mozilla and Chrome does and make it multiplatform.
In news that should surprise no one, the psychopaths over at google/chrome are now brainstorming an anti-competitive and anti-privacy scheme where users have to prove the “integrity” of their software and hardware stacks in order to access websites, so google can stuff their harmful ads down your throat without those pesky alternative/rooted operating systems, browsers and adblockers.
This is in order to protect you, of course. It allows google to uniquely identify you without having to rely on so many fingerprinting techniques. Which clearly improves your privacy, duh!
That’s why I think ^ this is so important. We are spread too thin to fix chromium releases properly unless we get all of the scattered chromium privacy forks (which are essentially doing the same thing) to collaborate, and at least use a common base fork.
involved with the above efforts made a karmic blog post where