3 ms·

I don't really disagree with that. I'm just challenging that many of those points are how people signal to their in crowd. Also notable that twitter, specific

by taeric 2y ago

I don't really disagree with that. I'm just challenging that many of those points are how people signal to their in crowd. Also notable that twitter, specifically, is overwhelmingly signalling to the crowd. I think this has been true for a long time, but it does feel that it is more extreme now. My guess is that this would be a more constructive argument against algorithms driving engagement. They are overwhelmingly hijacked by in group signals in ways that curated data, oddly, was not. Not that curation always worked, mind. You could build up more trust that curators at least put some reputation on the line with regards to other factors.

munificent 2y ago
> Also notable that twitter, specifically, is overwhelmingly signalling to the crowd. I think this has been true for a long time, but it does feel that it is more extreme now. There is an interesting positive feedback loop: 1. Some people's messages are intended to be in-group signals and others' are not. 2. Because of the former, people sometimes intepret the latter as implicitly signally membership in the opposing group. 3. Once people start making that interpretation, they start using that deliberately, so now messages that don't look like group signals do look like group signals for the other group. Eventually, the result is that nearly all messages are tribal markers and it's nearly impossible to say anything that isn't just a shibboleth.