hhmx.de

🅱🅻🆄🅴🅱️

· Föderation EN Sa 22.02.2025 20:54:24

@oldgeek

Ah, I didn't realize there was more that he had spoken about.

That is one thing (of many) I am frustrated by with social media. The sort of disjointed nature of it. He had made the arguments in the past, but it wasn't in THAT post.

alcinnz

Föderation EN So 23.02.2025 00:18:56

@BlueBee @oldgeek I'll remark that I've seen someone (can't recall who) *explicitly* state that the goal was to chase him off for being a popular YouTuber.

That prompted me to state: This is not on!

🅱🅻🆄🅴🅱️

Föderation EN So 23.02.2025 00:31:20

@alcinnz @oldgeek

I'm not sure what else can be done except begin regulating posts, but a central regulator is obviously bad. Too much power in too few hands.

I think we should probably have our own review structure of people kind of like steam does for games. But you get to say who's reviews you trust and you ignore everyone else's.

Where you and me and other people can say if we trust someone or not and then if you trust them their opinion of people affects the trust levels of others. And then allow people to be sortable by your personal trust levels for them.

I don't know if I'm explaining that well, but I don't want to spend too much time on it right now. Hopefully it gives the jist of the idea. If someone is a troll, I don't want to see them. But I don't trust a giant corporation to tell me who is and isn't a troll. YouTube surfaced Jordan Peterson to me and that man is an idiot.

So much of the pain people feel reading comments would be alleviated if we had a bit more proof required before allowing people to take up air time. I'm not saying turn people off, but label them differently, trolls shouldn't show up the same as someone who will engage with you with care.

alcinnz

Föderation EN So 23.02.2025 00:32:09

@BlueBee @oldgeek Rate limiting was a suggestion...

Maybe sentiment analysis software...

🅱🅻🆄🅴🅱️

Föderation EN So 23.02.2025 00:36:46

@alcinnz @oldgeek

I really only trust who I trust.

If you say sentiment analysis I'm now trusting the software and the writer of the software.

I would prefer people's opinions, and especially those who I trust.

Just allow me to say if I trust people or not and allow others to subscribe to that trust or not.

And allow me on the fly to disagree with someone else's trust. Someone shows up as gold trust with a flurry of bad takes. Let me override and say I don't need to hear them anymore.

alcinnz

Föderation EN So 23.02.2025 00:38:57

@BlueBee @oldgeek These tools would be for folks like the top-of-thread who can't respond to all the fan/hate-mail they get!

Fame has its downsides...

🅱🅻🆄🅴🅱️

Föderation EN So 23.02.2025 00:39:16

@alcinnz @oldgeek

Write me software that reviews video games. You can't.

alcinnz

Föderation EN So 23.02.2025 00:42:06

@BlueBee @oldgeek I think we're in the context-collapse territory top-of-thread complained about in his video.

Also: Who said anything about video games?

I'll end the conversation here!

🅱🅻🆄🅴🅱️

Föderation EN So 23.02.2025 00:46:12

@alcinnz @oldgeek

This isn't context collapse lol.

And the reason why I'm talking about video games is because video games are less complex to review than people, and if you can't write a program that can review them, then you can't write a program that can review people.

On how this relates to his issue, he is saying that the problem is he gets 1 to 1 good to bad takes. A big part of this is that he has no method of sorting these takes. They are all viewed equally. So no matter if it's a well respected scientist or a troll, they all look the same.

Some method of determining trust is required, but if you give this to a single entity like with Blue sky it's too much power for an individual entity to have. But it still needs to be solved as the original poster has stated.

We have a giant pool of very capable reviewers who could help the larger creators sort through people and find who they should probably be listening to, but they lack the levers to provide that service.

alcinnz

Föderation EN So 23.02.2025 00:52:45

@BlueBee @oldgeek Maybe people are hard to review, but text is quite easy to approximate an automated solution for. I was thinking of a system which judges based on word choice, it works reasonably well for eMail.

Anyways I'm not all that interested in this meta-talk...