Is there a need to formalize Botiquette?
socialhub.activitypub.rocks
What are well-behaved bots? On the Fediverse we have some informal conventions when it comes to dealing with bot accounts. A #nobots on our profile means we don’t want to be bothered by them. Though it is hard to get an overview, there are many different bots on the Fediverse. A lot of them are very useful, or fun to use. There are also many bots for which it is unclear what they do, who created them, who manage them, and to delve into the source code to figure out their inner workings. ...

Do we have a need for a good #Botiquette that can also be enforced or encouraged by the various social apps of the Fediverse?

What are your ideas and considerations?

Add them to the referenced Fediverse Futures SocialHub topic if you want them to be part of technical elaboration, if/when it comes to that.

See also my toot at: https://mastodon.social/@humanetech/108225431329625395

  • electrodynamica
    2·
    3 years ago

    In a word, no.

    Bots have been around forever and the most benign of them do the work that no one else wants to bother doing.

    Bots that are malicious or annoying will never conform to whatever you come up with.

    I’m saying this as someone who loves semantic data: I just don’t see a point, or benefit, or problem solved by this.

    • smallcircles@lemmy.mlOPM
      11·
      3 years ago

      Yes, you cannot avoid a malign bot. It may pose as a normal person and can just ignore any conventions. Sometimes you can detect them, for instance because their interactions are inhumanly fast.

      But here I am talking about Bots and bot developers that intend to be well-behaved. There we can have botiquette. Besides that, in the SocialHub post I am not only talking about polite conventions, but also things that can be enforced. Like replacing the #nobots thingy on the profile, with a real profile property that allows the server to restrict access based on a setting. Or disallow access based on missing source code or maintainer contact point.

      I should have chosen a better title though, as with ‘formalization’ I also mean extensions at the Fediverse protocol level.

      Another such extension might be block- and allow-lists specifically for bots (different than regular moderation blocks). If well-behaved bots are properly described, then I could filter an allowlist from the long list of known bots (‘known’ being the ones my instance knows about).

      Examples of accounts marked as ‘bots’ that are not well-behaved:

      • I get followed by bots and there is zero information on their profile. Why do they follow me? What do they do with my info / data?
      • On Matrix you have the Matrix Traveller Bot. It enters your chatroom and sits there idle for a couple of days and then leaves. What does it do? You have to go to matrix.org to figure that out, and it is not easy to find out exactly either.