[Discussion] privacy issues regarding Lemmy - I use arch linux FYI
iusearchlinux.fyi
I want to preface this with the fact that I’m quite new to lemmy and using the fediverse as a whole (this is my first account on a fediverse server), but I have had my eyes on the concept for a while now. What I want to address is some of the issues regarding Lemmy and I want to discuss solutions to addressing those. I do want to note that I haven’t looked into the source for lemmy and my personal source of information is this thread on mastodon [https://mastodon.social/@HistoPol/110522257089980041] which also talks about the political issues with this platform (I’m going to be sidestepping the political/humanitarian views the devs seem to have for this thread as I want to keep discussion here purely about the privacy of the platform) the issues I want to address here are related to how lemmy deals with “deleted” content. 1. comments deleted by users are still visible to admins 2. deleted user accounts still have their data remain on the instance. both of these are serious issues concerning user privacy and also violate several laws around the world regarding privacy (most notably the EU’s GDPR). additionally these seem to be easy (if not trivial) to address from a code perspective so I don’t see why these changes could not be made by someone familiar with the project and then used by instances before they get merged in the main project (if they will ever do so). I want instance admins to be aware of this particular issue (not just for my home instance, but as many as we can) and thus want this thread to be one of open discussion about how we would go about this. I’m just a random guy on the internet; but I want to know what we can do about this at the very least and maybe actually make some sort of difference.

thought I’d post it here as well for visibility

  • Spoichiche@burggit.moeEnglish
    1·
    1 year ago

    Honestly, while i can see why someone would be bothered by this, i don’t think there’s any solution to their concerns.

    The whole point of this fediverse idea, as i see it, is that all the data is fundamentally controlled by a collection of entities instead of a centralised one. And you, as a mere user, have to choose which of these entities will have absolute full control of your data. Regardless of the source code of the webapp, if you’re not fine with others having control of the data you willingly uploaded to their servers, the only way out is to self-host your data.

    If you upload stuff on remote servers you don’t control, you should always assume that you have no power over how that data is used. You can only hope for others to not be assholes, but you’ll never have a true garantee.

    • Scraft161@iusearchlinux.fyiOPEnglish
      0·
      1 year ago

      my problem doesn’t lie in having to trust an instance to not store your data; but rather the platform itself keeping this data without a clear way of turning it off when setting up your instance (aside from modifying the source)

      I agree that there’s always trust needed and who you trust in the end is up to you so you as the end user should make an informed decision if you care about this.

      • Spoichiche@burggit.moeEnglish
        2·
        1 year ago

        I think part of the reason why i don’t care that much, is because i’m looking at it from a software developper’s perspective.

        Even without touching the source code, which would require quite a bit of effort simply to familiarize yourself with the codebase, the moment you get full access to the database it becomes ‘trivial’ (as long as you know some SQL) to do absolutely anything with the data.

        Also, a somewhat common thing when working with databases is to never truly, permanently delete data, especially when the deletion comes from user-controlled actions. You can’t trust users to not delete data they didn’t want to delete and user accounts can also be compromised. Depending on the data itself, allowing total and permanent deletion can be very harmful and irrecoverable. When you don’t fully delete the data but simply ‘mark it as deleted’, it’s a lot easier to revert such problems. Even with the GDPR and the right to be forgoten, i don’t have any stats on this but i’m pretty sure there’s a lot of sites that simply mark things as deleted instead of a true permanent deletion.

        edit: Forgot to mention that since bugs are also a thing, not actually deleting data can save your butt more than once.