The best part of the fediverse is that anyone can run their own server. The downside of this is that anyone can easily create hordes of fake accounts, as I will now demonstrate.
Fighting fake accounts is hard and most implementations do not currently have an effective way of filtering out fake accounts. I’m sure that the developers will step in if this becomes a bigger problem. Until then, remember that votes are just a number.
Web of trust is the solution. Show me vote totals that only count people I trust, 90% of people they trust, 81% of people they trust, etc. (0.9 multiplier should be configurable if possible!)
deleted by creator
Fwiw, search engines need to figure out what is “reliable”. The original implementations were, well if BananaPie.com is referenced by 10% of the web, it must be super trustworthy! So people created huge networks of websites that all linked each other and a website they wanted to promote in order to gain reliability.
If this was implemented on the server, that implies a significant amount of information about said web of trust to be stored by the server admins. Furthermore, it would imply that that trust web is also federated out as if you’re a first tier trust of mine, for me (or a server calculating on my behalf) to evaluate the value of the likes and dislikes it would need your web of trust too… and transitively out.
If this was implemented on the client, that means effectively revealing the origins of all the likes and dislikes on an object. Aside from the “this can be a lot of data to send over the wire whenever someone looks at an active post” it also means that you wouldn’t need to be a server admin to see that data.
Either way, this approach with ActivityPub being the underlying protocol, would entail privacy violations and opportunities for bullying that anything Threads (from other threads of concern) could do would pale in comparison to existing bad actors.
P.S. Don’t trust me because I trust JohnDoe and he works for marketing at BigCo and might be persuaded to list some of his paid clients highly.
It could be implemented on both the server and the client, with the client trusting the server most of the time and spot checking occasionally to keep the server honest.
The origins of upvotes and downvotes are already revealed on objects on Lemmy and most other fediverse platforms. However, this is not an absolute requirement; there are cryptographic solutions that allow verifying vote aggregation without identifying vote origins, but they are mathematically expensive.
Given that Lemmy isn’t that popular yet, how big would that payload and computational cost be when considering the votes on the highly active threads of !fediverse@lemmy.world … 1.5k votes with 960 comments. Or the highly active https://lemmy.world/post/1033769 (3k votes, with 1081 comments) from earlier this week.
It’s nothing. You don’t recompute everything for each page refresh. Your sucks well the data, compute reputation total over time and discard old raw data when your local cache is full.
Historical daily data gets packaged, compressed, and cross signed by multiple high reputation entities.
When there are doubts about a user’s history, your client drills down those historical packages and reconstitute their history to recalculate their reputation
Whenever a client does that work, they publish the result and sign it with their private keys and that becomes a web of trust data point for the entire network.
Only clients and the network matter, servers are just untrustworthy temporary caches.
Any solution that only works because the platform is small and that doesn’t scale is a bad solution though.
Client must computer all raw data. All individual moderation action (vote,block, subscribe) would be made public by default and stealth optional.
Only user led moderation has a future, it all has to be transparent, public, client sided, optional and consensual
That sounds a bit hyperbolic.
You can externalize the web of trust with a decentralized system, and then just link it to accounts at whatever service you’re using. You could use a browser extension, for example, that shows you whether you trust a commenter or poster.
That list wouldn’t get federated out, it could live in its own ecosystem, and update your local instance so it provides a separate list of votes for people in your web of trust. So only your admin (which could be you!) would know who you trust, and it would send two sets of vote totals to your client (or maybe three if you wanted to know how many votes it got from your instance alone).
So no, I don’t think it needs to be invasive at all.
The single layer web of trust on the server wouldn’t be terribly difficult.
A single layer web of trust on a client would mean that the client is getting sufficient information about all the votes to be able to weight them. This means that instead of “+4 -1” for the information that the client gets instead it would get that “shagie liked the object, JohnDoe liked the object, BadGuy liked the object, SomeoneElse liked it, and YetAnotherPerson disliked it.” That implies a lot more information being revealed to a client than many would be comfortable with.
Granted all of that is available if you federate with a system and poke in the database. It’s there. But this makes it really easy to get that information.
A transitive web of trust implies not only are you getting those votes and considering that “shagie liked the object” but also that you trust me and so that I trust JohnDoe is available to whatever is making that vote weighting calculation.
And while that single layer on the server isn’t too eyebrow raising, getting the transitive listing gets into the Facebook level of social graph building - but for all to see. I’m not sure that people would be comfortable with that degree of nakedness of personal information.
Consider also the data payload sizes. This post (rather mundane and not viral) has 243 comments. Some of them have over a hundred votes. How big of a payload do you want to get to send to the vote weigher (and back)?
Consider the load for… say… https://lemm.ee/post/843533
And for bad actors, all they have to do is cast a couple hundred votes on each comment (until they’re defederated and the database cleaned up by the admin) to DDOS the vote weigher.
My point is you can have a mixed system. For example:
That’s not a ton of data, and the “special interest” users wouldn’t need to be synchronized to any other instance. The client would store the WoT data and update the server as needed (this way the server doesn’t need any transitive logic, the client handles it).
Facebook and Twitter have always had their equivalent of upvotes be public.
What if the web of trust is calculated with upvotes and downvotes? We already trust server admins to store those.
I think that could work well. At the very least, I want the feature where I can see how many times I’ve upvoted/down voted a given individual when they post.
That wouldn’t/shouldn’t give you transitive data imo, because voting for something doesn’t mean you trust them, just that the content is valuable (e.g. it could be a useful bot).
Your client has to compute the raw data, not the server or else it will just be your server manipulating what you see and think.
Love that type of solution.
I’ve been thinking about an admin that votes on example posts to define the policy, and then getting users scored against it, then using high scorers to represent user copies of the admins spirit of moderation, and then make systems that use that for automoderation.
e.g. I vote yes, no, yes. I then run the script that checks my users that have voted in all three, and the ones with the highest matching votes that i define(must be 100% matching to my votes) gets counted as “matching my spirit of moderation”. If a spirit of moderation user downvotes or reports then it can be auto flagged into an admin console for me to then rapidly view instead of sifting through user complaints, and if things get critically spicy i can promote them to emergency mods, or automate their reports so that if a spirit user and a random user both report, it gets auto removed.
For each vote, read user post content and vote history and age
This should happen in the client and easily controllable by the user. As well as to investigate why one particular post or current was selected by the local content discovery algorithm. So you can quickly find fraudulent accounts and block them.
And this public, user led moderation actions then go on to inform the content discovery algorithm of other users until we have consensus user led content discovery and moderation.
And just like that we eliminate the need for shadowy humans of the moderator priesthood to play human spamfilter / human thought manipulator
This was a great feature of reddit enhancement suite.