I am new to Librewolf and downloaded Multi Account Containers addon along with libredirect, skip redirect and ublock origin. I am currently signed in a container with a Google account and set the URL to always open in that container. However I do not want to allow any Google-affiliated redirecting links to automatically open within that container and sign me in with that account, letting Google know what I had been browsing or searching for even when I would not want them to. This issue is restricting me from making Librewolf my default browser.

How do I make it so that all these links, whether clicked on from somewhere within the browser or an external application, open in a separate container or without a container or in a private window, if possible?

  • lemmyvore@feddit.nl
    link
    fedilink
    English
    arrow-up
    3
    ·
    edit-2
    1 year ago

    no history, no cookies, no login information

    If your concern is online tracking then it doesn’t really matter, browser fingerprint + your exit public IP are enough to identify you uniquely pretty thoroughly even without any persistent information (cookies, local storage etc.) And yes, LibreWolf includes anti-fingerprinting protection, but it’s not fullproof and it cannot be without ruining most JavaScript functionality. Do the EFF test to see.

    If your concern is privacy from people using your computer then I suggest using multiple OS accounts and a screen lock.

    • jetA
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      Using two separate browsers solves the fingerprint problem. If they only log in to their services using their LibraFox container then that will be a distinct fingerprint than they’re logged out browser

      • lemmyvore@feddit.nl
        link
        fedilink
        English
        arrow-up
        2
        ·
        1 year ago

        You’re thinking about canvas fingerprinting, but there are APIs which cannot be faked without messing with functionality, and there are enough of them to obtain a unique fingerprint even if you omit the canvas. Which advertisers can easily do if they figure out you’re faking your canvas.

        A lower entropy fingerprint plus the IP can still identify you across browsers. If you compare the results of the EFF test in the two browsers you’ll see that most of the values will be the same.

        • jetA
          link
          fedilink
          English
          arrow-up
          2
          ·
          1 year ago

          I’m sorry. I must have misunderstood. You just said that different browsers from the same IP address can be linked to the same person.

          I find that difficult to believe, here’s a counter example, a VPN endpoint. There’s many different people with that originating IP address using different browsers. Being able to identify two that an individual is using would be very difficult.

          Browser fingerprinting fingerprints a browser, not a user, not a computer. A browser. I would love to see data contradicting that, but I am not aware of that data

          • lemmyvore@feddit.nl
            link
            fedilink
            English
            arrow-up
            2
            ·
            1 year ago

            Browser fingerprinting fingerprints a browser, not a user, not a computer.

            But how much of the fingerprint data comes from stuff that’s common across browsers? Do the EFF test in a completely different browser and see how much will be the same in Chrome vs Firefox vs Librewolf for example.

            • jetA
              link
              fedilink
              English
              arrow-up
              2
              ·
              1 year ago

              That’s a good point and it’s always worth testing any setup you have.

              My mullvad browser passes the fingerprint.com test with flying colors, it doesn’t share system fonts and the like. That plus a VPN would make it really hard to say the librewolf browser and tbe mullvsd browser belong to the same person.

              I still think there’s value in using separate browsers for the scenario the original poster outlayed