And what is the social difference between a publisher and a platform in terms of expected social responsibility? This distinction only now exists because people attempted to use a legal definition that isn't real. So, again, I think this entire line of thinking falls flat upon inception.
> But too often do we give Facebook a pass since it doesn't make the content, it only curates it.
But that is true, no? Do we honestly believe that between ML and a real person Facebook is validating and approving every comment on Facebook? I sure don't. Do we believe that Google is publisher because mailing lists can exist or because they allow an email to be sent while blocking perceived malicious emails? Is that not some form of validation that happens that we could imply makes Google a publisher of every email sent from GSuite?
I just don't agree that an algorithm is undeniably curation in the same sense as a newspaper. The process isn't the same, the intended outcome isn't the same, and the inherent approval of the content isn't the same. So while they may be like conceptually, they aren't the same functionally whether you measure it by the spirit or letter of the systems.
The idea of user moderation vs service based moderation is purely a feature set. If we don't like that feature set or it doesn't meet the needs of the way we use the service, the service sucks at it's job. The correct solution isn't to try and have all these convoluted/philosophical discussion around where to draw the line. The solution is to have a service that does it better get the userbase. Today people like Facebook because it already achieved a wide enough userbase, which lead to default integration into other services. It seems we are willing to trade convenience for expectations, which is an issue for the user, not the service.
Now, if we want to say that they practice anti-competitive market strategies and are monopolies, I can entertain that argument. It still does not require this discussion at all.