F40 Change Request: Privacy-preserving Telemetry for Fedora Workstation (System-Wide)
discussion.fedoraproject.org
This is a proposed Change for Fedora Linux. This document represents a proposed Change. As part of the Changes process, proposals are publicly announced in order to receive community feedback. This proposal will only be implemented if approved by the Fedora Engineering Steering Committee. Break-Out Topics! Because this a big conversation, we have identified several important themes and created separate break-out conversations for those. For these topics in particular, please follow the appropr...

the proposal is from a red-hat team member and is proposing addition of “privacy-respecting” telemetry. here’s the link to the hyperkitty thread

  • ono@lemmy.caEnglish
    5·
    1 year ago

    The problem with opt-in telemetry is that it messes with the scope of the research.

    Too bad. That does not make it okay to collect data without consent.

    Not ever.

    In other words, unbiased telemetry is not possible to do ethically. (Or to say it differently, ethical telemetry necessarily has bias.)

    • joojmachine@lemmy.mlEnglish
      03·
      1 year ago

      And once again, it isn’t “without consent”, it just means that the default state of the checkbox is on. Users will still be presented with a confirm option before any data is sent.

      In other words, unbiased telemetry is not possible to do ethically.

      Say that to the opentelemetry and Plausible folks, who have been on the vanguard of doing exactly that for years now.

      • SymbolicLink@lemmy.caEnglish
        4·
        1 year ago

        Having the default box being “on” is only for the purpose of hoping people click through without realizing.

        There is literally no other argument here. “Consent” is: “Hey do you want this, yes or no?”. Not “We are assuming yes unless you explicitly tell us otherwise”.

      • ono@lemmy.caEnglish
        2·
        1 year ago

        the default state of the checkbox is on.

        That’s a very strange thing to mistake for consent.

        • joojmachine@lemmy.mlEnglish
          22·
          1 year ago

          It’s not mistaken for context, you’re just missing the point. The switch is just part of it, the user would still have to consent to send their data before it is sent and the proposal proposes to have it detailing the data that is going to be sent and explaining the process.

          Having it as a default guarantees it doesn’t scare non-power users away from it. It’s not about just having people clicking next and accepting it without consent.

          • SymbolicLink@lemmy.caEnglish
            2·
            1 year ago

            Disagree, no matter the level of detail, having “yes” automatically selected is an assumption. What purpose would it have other than hoping people will just select the defaults and ignore it?

            Having it as a default guarantees it doesn’t scare non-power users away from it. It’s not about just having people clicking next and accepting it without consent.

            Scare away from what? Data collection? I mean even in that wording you are saying there is something to be scared of. It should be up the user. If you are saying “non-power users won’t fully understand what is being collected and might get scared away if it isn’t the default option” then that is even worse TBH. Preying on people not fully understanding what’s going on.

        • Aman Das@rammy.siteEnglish
          01·
          1 year ago

          It’s not mistaken for consent, only if you dont switch it off the data is sent.

          • joojmachine@lemmy.mlEnglish
            1·
            1 year ago

            Wrong again, the “switch” only sets the setting itself, but before finishing the setup process the user would be provided with a confirmation detailing the data that would be sent and provided with the option to send it or not. The process would have user consent.