Posted by ingve 16 hours ago
That article needs to have about 80% of the words cut out of it.
When the author straight up tells you: I'm posting this in an attempt to increase my subscribership, you know you're in for some blathering.
In spite of that, personally I think algorithmic feeds have had a terrible effect on many people.
I've never participated, and never will...
> As in real life.
No, your proposal is very different to real life. In real life, the things you say will eventually be forgotten. You won't be fired for things you said or did years ago, because people will have moved on.
Having a convenient index of everything anyone has ever shared is very different to real life.
You realize that the evidence is against you on that one. Just recently, who was that UK ambassador that Prime Minister Keir Starmer just fired?
For instance a political vote needs to be anonymous. Access to public space typically is (you're not required to identify to walk the street) even if that anonymity can be lifted etc.
Real life is complex, and for good reasons, if we want to take it as a model we should integrate it's full complexity as well.
If you’re out in public, you’re also not fully anonymous. You display metadata such as race, gender, age, behavior. Now you could wear a ski mask during broad daylight but I doubt if you’d be allowed inside a bank. And the bank has a right to judge you for that.
That cannot be right, that's the fundamental core of the voting process in our democracies. You might be thinking about the party registrations or voluntary polls ?
> You display metadata
What you show to the world has no requirement to be accurate. If you look like a rich 70 old Asian lady when going to the park there will be no check that's actually what you are (unless the police comes at you for an identity check...). That's particularly impacting for gender, you're typically not required to represent your official assignment, and how you behave isn't stuck to your official identity.
Rather than sacrifice the cover that anonymity grants vulnerable people, journalists, and activists, I think we should come at this issue by placing restrictions on how social media platforms direct people to information. The impulse to restrict and censor individuals rather than restrict powerful organizations profiting from algorithmic promotion of the content you deem harmful is deeply troubling.
The first step here is simple: identify social media platforms over some size threshold, and require that any content promotion or algorithmic feed mechanism they use is dead-simple to understand and doesn't target individuals. That avoids the radicalization rabbithole problem. Make the system trivial and auditable. If they fail the audit then they're not allowed to have any recommendation system for a year. Just follows and a linear feed (sorting and filtering are allowed so long as they're exposed to the user).
To reiterate: none of this applies if you're below some user cutoff.
Q: Will this kill innovation in social media? A: What fucking innovation?
Quite the contrary, a core journalism principle is accountability and transparency. Readers must know who the reporter is to assess credibility, context, and potential conflicts of interest. Attribution builds trust, allows audiences to verify the source, and distinguishes reporting from anonymous or propagandistic material. This is different from covering source anonymity, but the audience is still relying on the journalist’s _known_ integrity that they’re not just making up some bullshit source.
When you remove the incentive to engage users, the companies will engage in less abusive practices to push engagement.
I've never seen this proposed, and I'm confused why.
- social incoherence because silos cannot communicate laterally is still there
- the ads will likely go native to become "content" and more revenue will shift to influencers
Just saying it's not quite that easy, but yes, ad monetization is a great force of evil.