3 ms·

If Twitter has taught us anything, it's that everyone should not have an equal voice and an equal platform. A person or company's reach or audience should be a

by sub7 6y ago

If Twitter has taught us anything, it's that everyone should not have an equal voice and an equal platform. A person or company's reach or audience should be a function of 1) what they're talking about and 2) what credibility they have re: 1. I'm all for anyone being allowed to say anything but I'm completely against anyone being allowed to amplify anything. Also, the Republicans would do well to introduce legislation expanding data portability and making it much harder for these completely unethical platforms to keep your data locked into their walled gardens. Friends names, birthdays and emails as a csv is basically impossible to export from FB for example.

lliamander 6y ago
> If Twitter has taught us anything, it's that everyone should not have an equal voice and an equal platform. A person or company's reach or audience should be a function of 1) what they're talking about and 2) what credibility they have re: 1. How do you propose we should establish credibility of sources? Also, how do propose to factor in the originality of what they say? > Also, the Republicans would do well to introduce legislation expanding data portability and making it much harder for these completely unethical platforms to keep your data locked into their walled gardens. Friends names, birthdays and emails as a csv is basically impossible to export from FB for example. Agreed.
- sub7 6y ago
  Let's put aside originality for now. I think with a few small deep structural changes, you can get 80% of the way there. I would have all posts require a tagged context, let's call it a topic (in reality it'd be multi facet hierarchical object). I would also have all profiles list the topics they'd like to be credible in. The credibility score for a given topic would be an algorithm that would use followers credibility scores to "rub off" on your score for that topic. You seed the profiles you know to be credible and then propagate the graph. It'd initially be a guesstimate and then gets refined as more humans rate, interact and judge the post. Over a few posts, I think you'd be able to get a broad credibility bucket for a given profile/topic pair. Btw I've built a social network with tens of millions of users and we used this approach and it works pretty well. It does require a bit of moderation and bad actor weeding out but for the most part it works. Also, FB and Twitter aren't full of retards (shockingly) They know they can do this but they also know they would tank a shitload of their core quantitative metric for a vague qualitative metric. It'd just never happen from within.

by sub7 6y ago

lliamander 6y ago

> If Twitter has taught us anything, it's that everyone should not have an equal voice and an equal platform. A person or company's reach or audience should be a function of 1) what they're talking about and 2) what credibility they have re: 1. How do you propose we should establish credibility of sources? Also, how do propose to factor in the originality of what they say? > Also, the Republicans would do well to introduce legislation expanding data portability and making it much harder for these completely unethical platforms to keep your data locked into their walled gardens. Friends names, birthdays and emails as a csv is basically impossible to export from FB for example. Agreed.

sub7 6y ago
Let's put aside originality for now. I think with a few small deep structural changes, you can get 80% of the way there. I would have all posts require a tagged context, let's call it a topic (in reality it'd be multi facet hierarchical object). I would also have all profiles list the topics they'd like to be credible in. The credibility score for a given topic would be an algorithm that would use followers credibility scores to "rub off" on your score for that topic. You seed the profiles you know to be credible and then propagate the graph. It'd initially be a guesstimate and then gets refined as more humans rate, interact and judge the post. Over a few posts, I think you'd be able to get a broad credibility bucket for a given profile/topic pair. Btw I've built a social network with tens of millions of users and we used this approach and it works pretty well. It does require a bit of moderation and bad actor weeding out but for the most part it works. Also, FB and Twitter aren't full of retards (shockingly) They know they can do this but they also know they would tank a shitload of their core quantitative metric for a vague qualitative metric. It'd just never happen from within.