I’m unsure how it will play out in practice. I think it’s possible that different infra could wind up being deployed in jurisdictions that differ too significantly. Certainly that could happen outside of Bluesky.
Bluesky itself is US-based.
Bluesky itself is US-based.
--
https://www.europarl.europa.eu/RegData/etudes/ATAG/2020/6581...
As you can use their guidelines:
European Union (EU), there are several types of social media content that are considered illegal.
---
* Incitement to Terrorism: Any content that encourages or promotes terrorist acts, violence, or extremism is prohibited.
* Illegal Hate Speech: Social media posts that spread hate based on race, ethnicity, religion, gender, sexual orientation, or other protected characteristics are not allowed.
* Child Sexual Abuse Material: Sharing, distributing, or creating content related to child sexual abuse is strictly illegal.
* Infringements of Intellectual Property Rights: Posting copyrighted material without proper authorization violates intellectual property rights.
* Consumer Protection Violations: Misleading advertisements, scams, or fraudulent content that harms consumers are prohibited.
--
These rules are further strengthened by stricter regulations for four specific types of content, which have been harmonized at the EU level:
* Counter-Terrorism Directive: Addresses terrorist content.
* Child Sexual Abuse and Exploitation Directive: Focuses on combating child sexual abuse material.
* Counter-Racism Framework: Aims to prevent and combat racism and xenophobia.
* Copyright in Digital Single Market Directive: Deals with copyright infringement online1.
---
You could ideally put the EU illegal categories in a drop down like old Slashdot Style - but each mod who selects the same selection from the drop down adds a point to the tag - the post is removed upon some thresholds of the points.
This could be different for each reagion - and a post could be flagged with points for each region... so have a region selection, then the illegal tag list sets to that region. A post maybe could be tagged by multiple regions infractions based on where the mod sits?
Also - you can keep metrics for all posts flagged for what infraction types for - plus you can move them to the "if law enforcement needs this post S3 bucket" -- based on whatever time period the laws require.
Coming from Silicon Valley companies, all the federation stuff aren't sincerely-intended ideas to promote free society values. They're an escape hatch. It's a free-speech zone to divert unwanted activists onto—away from the company and its brand, and away from its users—where they can blow off steam quietly. And it's something their PR can point to to deflect accountability for the awful things they will definitely end up doing, in their uncompromising pursuit of market share. ("Oh, we're not really censoring that; it's only censored on the main corporate instance, which is the only one people use").
Federated social media could be used for very bad purposes and it would be impossible to stop. Consider this situation
https://www.amnesty.org/en/latest/news/2022/09/myanmar-faceb...
in the centralized case we can point to Facebook as a responsible entity. The thing is Facebook later got kicked out of Myanmar not as accountability for those crimes but because they were insufficiently pliant to the government. In the future a country like that might just run its own Mastodon network where it is all genocide all the time.