Social Media Design

Today I blogged some thoughts on how to manage and support natural conversation flow (parallel and meandering) on group chat servers: https://www.brainonfire.net/blog/2021/11/28/chat-conversation-splitting/

Curious to hear what people have to say.

I've been designing a distributed social media protocol, and one thing I've been puzzling over is comments. Specifically, how to handle plausible deniability.

For example If I comment on your journal, there are at least three ways I could possibly do it in some imagined protocol:

1. Hand you a comment, and you publish it to your journal. This is like how commenting works on Wordpress. Simple, but you could make fake comments and pass them off as mine. Maybe this is not a problem, though!

2. Hand you a cryptographically signed comment, which you publish to your journal. Anyone could verify that it's really from me. But on the other hand... it can now be proven at any point in the future that I said that thing, which might be highly embarrassing, or misleading out of context. It's a mismatch for the way casual interaction usually works; we don't go around getting our chit-chat friggin' notarized.

3. Publish the comment myself, and send you a link to it; you display the comment and include metadata indicating where it can be retrieved on my site. People's clients can verify it comes from me by asking my journal about it. If I ever want to take it down, I can just delete or hide it. Downside: More complicated, more fragile.

Option 3 might be for the best since it's the closest match to "casual privacy", despite the complexity.

(There's also still the problem of how to publish a journal post and ensure that people can verify that I wrote it at the time they retrieve it, but *not* allow them to convey that proof to others. The solution to that might also be the solution to the comments issue, or possibly not.)

One of my goals in social media design is to promote the use of distributed systems, moving away from centralized ones.

( the problem of naming in distributed systems (cut for length) )

Petnames

What's interesting to note here is that in everyday life, we don't go around calling each other "Mrs. H8owhw85g3w492" or "Mx. 00813t3gh2ig3". There's no use of cryptographic identities, and there's no central naming authority. In offline social life, we don't bother with uniqueness of names, because we rely on social context (Terry-at-work vs. Terry-with-the-moustache vs. Terry-who-I-know-through-Alex). And so one rather radical idea (that isn't radical at all) is to not bother with secure, unique names at the user interface level. Under the covers, a distributed system should probably assign everyone cryptographic keys, so that users can be assured in who they are talking to. But those cryptographic keys should be hidden underneath "petnames". If your friend sets their name to be "John Smith" or "xXxDaRkMeTaLxXx", by gosh, that's what should be shown to you! And if someone else has the same name, the application can disambiguate, or ask you to assign locally-unique names to each one. (Tall John, Work John.)

It can go beyond petnames, too. If you have a John Smith in a custom contact grouping called "coworkers", another one is a contact of your friend Alice, and you encounter a third one you've never met in the comments in someone else's journal, then your client should have enough information to disambiguate them without your assistance. The names could be displayed like John Smith [coworkers], John Smith [via Alice S.] or John Smith [?]. And of course, you could always assign your own petnames. Petnames being only locally unique is no trouble. If you mention someone in your journal (e.g. timmc) the software can embed their unique cryptographic key identity out of site, and just display your petname to you—and when someone reads your post, they'll see whatever their client shows, including possibly a different petname.

I think this could work. Our brains are already adapted to the notion of non-unique names, and social media software would already have enough information to be of assistance to disambiguating names and preventing spoofing.

I'm curious to hear your thoughts on whether this approach would be sufficient, or what problems it might have.

This is a community for people who are interested in the design of social media systems—both analyzing existing ones, and designing new ones that will better affirm their users' privacy and security, respect the value of their time and resources, and embrace their unique abilities and needs.

As we've seen from the past few years, perhaps the past decade, it's time for a change in social media. Earlier systems relied heavily on good faith, users who brought more curiosity than cynicism, a certain type of homogeneity of worldview, and a relative lack of interference from nation states and surveillance capitalists. These systems have not fared well as the realities of the world have caught up with them.

This is an exploratory space to imagine better futures together.

The goal is not to design one new system to Solve It All, but to encourage discussion, experimentation, analysis, and collaboration. Some sample themes and topics to give a sense of the intended scope:

• What is the role of public discussion vs. private discussion? Global conversations vs. living room and café chats?
• How can we leverage the vast expanse of human experience on this planet while minimizing harassment, dogpiling, disinformation campaigns, censorship, and chilling effects?
• To what extent is it useful to combine tools for instant messaging, journaling, organizing, scheduling, and collaboration? Which of these use-cases deserve dedicated tools, or even a proliferation of tools?
• How has moderation succeeded and/or failed on the big social sites (Reddit, Twitter, Facebook, Tumblr)? How does moderation differ among these sites? What might work better?
• To what extent is it useful to combine tools for instant messaging, journaling, organizing, scheduling, and collaboration? Which of these use-cases deserve dedicated tools, or even a proliferation of tools?

This is also a place for brainstorming protocols and implementations, where people can feel free to ask for guidance or feedback on:

• Moderation toolkits
• Peer to peer networking protocols
• User interfaces and UX matters
• Resilience and robustness in distributed systems
• Prototypes

The boring details: Membership is open to all. Posts will generally be publicly visible. Conduct guidelines are, for the moment, left at "don't be a jerk" (which includes harassing behavior). If you are unable to resolve a conflict by discussion, or even if you're just unsure, please contact one of the community administrators (or if appropriate, Dreamwidth staff.) Self-promotion is fine if it is on-topic and tastefully infrequent. Linking to prototypes and interesting experiments in the social space is encouraged, especially if accompanied by discussion of how they're different and interesting.