• Zarxrax@lemmy.world
    link
    fedilink
    English
    arrow-up
    64
    ·
    2 months ago

    Would it be possible for a browser or extension to just provide false metadata in order to subvert this type of fingerprinting?

    • Dr. Moose@lemmy.world
      link
      fedilink
      English
      arrow-up
      11
      arrow-down
      1
      ·
      2 months ago

      No. Anything that executes Javascript will be fingerprinted.

      That being said it depends who are you fighting. For common commercial tools like Cloudflare fingerprinter it might work to some extent but if you want to safeguard against more sophisticated fingerprinting then TOR and no JS is the only way to combat this.

      The issue is that browsers are so incredibly complex that it’s impossible to patch everything and you’ll just end up getting infinite captchas and break your browsing experience.

    • kipo@lemm.ee
      link
      fedilink
      English
      arrow-up
      8
      arrow-down
      1
      ·
      2 months ago

      Yes. There is a firefox extension called Chameleon that does this.

    • fmstrat@lemmy.nowsci.com
      link
      fedilink
      English
      arrow-up
      2
      ·
      2 months ago

      Others have mentioned what Firefox/etc do, but another option is a PiHole. If you can’t look up the IP for an advertiser URL, you don’t load the JavaScript to begin with.

    • JackAttack@lemmy.dbzer0.com
      link
      fedilink
      English
      arrow-up
      1
      ·
      edit-2
      2 months ago

      So from what I understand, theres 2 common ways that browsers combat this. Someone add to or correct me if I’m wrong.

      1. Browsers such as Mull combat this by looking the same as every other browser. If you all look the same, it’s hard to tell you apart. I believe this is why people recommend using default window size when using Tor.

      Ex: Everyone wearing black pants and hoodies with the facemasks. Extremely hard to tell who is who.

      1. Browsers such as Brave randomize metadata that fingerprinting collects so that it’s more difficult to piece it all together and build a trend/profile on someone.

      Ex: look like a dog in one place, a cat in another place. They get data for a dog but that doesn’t help build anything if the rest of the data is a cat, hamster, whatever. No way to piece it together to be useful.

      In both my examples, there are caveats. Just because everyone dressed the same doesn’t mean someone isn’t taller or shorter, or skinnier or fatter. There can still be tells to help narrow down. Or a cat that barks like a dog suddenly is more linkable to a dog if that makes sense lol.

      In other words it still depends user behavior that can contribute to the effectiveness of these tools.

      EDIT: got distracted. To answer your question I don’t think so. I think it’s more about user behavior blending in or being randomized. I think the only thing an extension would be able to do is possibly randomize the data but I’m unsure of such an extension yet. These aren’t the only options, these are just ones I’ve read about recently. Online behavior, browswr window size, and I’m sure so much more also goes into it. But every little bit helps and is better than nothing.

      EDIT2: Added examples for each for clarity.