Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Disproportionate reach control #13

Open
TristanGrayCC opened this issue Jun 24, 2023 · 3 comments
Open

Disproportionate reach control #13

TristanGrayCC opened this issue Jun 24, 2023 · 3 comments

Comments

@TristanGrayCC
Copy link

TristanGrayCC commented Jun 24, 2023

One issue faced across social networks is when a very large account, used to dealing with huge quantities of interactions, quote-posts or otherwise promotes their response to a much smaller account that is then overwhelmed with a level of interaction that is completely beyond their capacity to handle. This is especially bad when it's used to target individuals for harassment by the fans of the far larger account. This tactic is frequently used to bully critics off other platforms.

Would it be possible to set a client-level control on how larger a discrepancy in size between accounts can be before quote-posts are disabled? Eg a client can set it so quote-posts are unavailable if your account has 100x the followers of another account.

@davekasten
Copy link

Oooooh I like this. It's only a speedbump and thus preserves freedom of speech -- people can still screenshot, write out the post, include a direct link, etc. -- but that extra bit of friction discourages dogpiling harassment behavior.

@sneakers-the-rat
Copy link

Similar to #18 , how would this be enforceable in a federated context? an adversarial client that ignores your limit can still issue the post that references yours, even if your client doesn't recognize it/drops it/etc. it could still be visible to an unknowable amount of badly behaved clients

@TristanGrayCC
Copy link
Author

It wouldn't, but clients should still be able to implement their own rules on how their users interact with one another regardless. If a client believes another client is frequently being used to direct harassment in this way without those rules then clients can enter a discussion about that and in a last resort cut one another off completely.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants