Verification will very quickly become an issue on Fedi, I believe.

Even now, we have multiple “Linus Torvalds” accounts.

Some have thousands of followers and a few posts, but do we know if any of them are actually him? Including the newest one on .social…?

Obviously Linus should know how to paste a link into a website he owns for verification.

However, if normal users can’t do it, it’s not good enough. What happens when/if celebrities start signing up? They’re not gonna be linking to a website they own for verification.

#Fedi #Fediverse #Mastodon

  • nickwitha_k (he/him)@lemmy.sdf.org
    link
    fedilink
    arrow-up
    1
    ·
    16 days ago

    My approach is to treat it like an old school forum. That is, everyone is only verifiable as their username, not what their username says. If they are on an instance with a domain owned by a known org (ex. npr.org), there may be some greater assumption of officiality.

    However, if normal users can’t do it, it’s not good enough. What happens when/if celebrities start signing up? They’re not gonna be linking to a website they own for verification.

    This is the same principle. Without something tying to a state or other organization that can verify identity, there isn’t a good way of doing this. Further, I’d say that extant social media did a terrible job as it is. I feel it much better to not worry about verification and instead treat everything as a username - celebrities are just people anyway and there is little justification to treat them any differently on a platform level.

    Is the Margot Robbie account really Academy Award-nominated character actor and producer Margot Robbie? It doesn’t matter in a manner that is material to the software that makes up the Fediverse. It’s a username for an account that has taken part in good discussions.

    Is Stamets really a Star Trek character somehow dwelling in the real world like one of the episodes where a character escapes from the holodeck? I doubt it but there is no way of knowing for certain without verification by a trusted third-party. I do know that the account has been instrumental in me seeing more Star Trek shitposts.

    • BeAware :fediverse:@social.beaware.liveOP
      link
      fedilink
      arrow-up
      3
      ·
      16 days ago

      @[email protected] the issue is, misinformation. If nobody can verify who is who. Then anyone can say anything and it wouldn’t mean anything.

      Celebrities saying things on social media, frequently ends up in the news/tabloids. These things need to be verifiable.

      • BearOfaTime@lemm.ee
        link
        fedilink
        arrow-up
        3
        ·
        edit-2
        16 days ago

        That’s the tabloid’s problem then, and who believes shit in tabloids anyway?

        “Pregnant Man Gives birth, that’s news, it’s in the Weekly World News”

        Frankly I don’t give two shits whether it’s the person they claim to be…i don’t even look at usernames most of the time. I care even less about what some public figure, especially an actor, says.

        If a figurehead decides to be active on any social media, I don’t pay attention to that garbage, because that’s what it is.

      • nickwitha_k (he/him)@lemmy.sdf.org
        link
        fedilink
        arrow-up
        1
        ·
        16 days ago

        I absolutely agree that misinformation (and disinformation) is a problem. However, I do not think that it is something with a good software solution at this time. The traditional/corporate social media effectively implemented a non-solution to the problem and declared it solved. Not to mention, verification of identity is an issue that governments also struggle with, relying on substantial amounts of good faith.

        There ARE means of tying a software identity to a verifiable and hard to counterfeit computing object. For example, a local cryptographic key pair or a hardware security module. However, without involvement of a trusted third-party, it is not currently possible to tie this computing object to a specific human being.

        My thinking is that attempting to implement an identity verification component in the Fediverse, therefore, is likely a misallocation of effort that could be better spent fixing bugs, extending features, and improving stability and interoperability. There is a lot still to do in virtually every project that I’ve peeked at, whether it’s mod tools, IaC, or just plain code cleanup. I think that at this juncture, what is required is more on the social side of things, ensuring that people are aware that one cannot believe everything that they see on the Internet or the identity claims of those that they interact with, unless they undertake further verification themselves. This is what was done in pre-social media forums and BBS systems with a good degree of effectiveness.

        Don’t get me wrong, I would be ecstatic to be proven incorrect but would much prefer that identity verification be shelved until such a time as crytographers are able to solve it with their dark mathematical arts than for any Fediverse project to waste time from people’s lives trying to implement a similar non-solution.