4 ms·

great question - something i've been thinking about a lot. my framework is, - moderation needs two levels (top level and grassroots) - top level - has to com

by viksit 3y ago

great question - something i've been thinking about a lot. my framework is, - moderation needs two levels (top level and grassroots) - top level - has to come from communities that are not in an "accepted" list which shouldn't be allowed to form. (the usual suspects like hate speech, CSAM). one way to think about is having community server owners "stake" some sort of investment and risk losing it in case communities on their servers get banned. this would happen for instance if you run a server for this protocol, and someone on your server creates a community against policies set by the protocol steering committee. - grassroots level - moderators are those who actively contribute to the community, can be elected, have audits on their logs, and can also be incentivized for their work (eg, given a "salary" every month) through that staked fund we talked about earlier. ultimately, no one is doing any of this free and we should assume that monetary incentives in some way that are aligned with the community's have to be factored in.

flagrant_taco 3y ago
> top level - has to come from communities that are not in an "accepted" list which shouldn't be allowed to form. (the usual suspects like hate speech, CSAM) Even this becomes tricky because you first have to make a common list of banned content and define the lines between what each category of content is. Sure a huge majority of people would like to see CSAM on the list, but they definitely won't agree on what is/isn't CSAM in their opinion I've recently been working through whether any level of content centralization is possible without the fundamental challenge of moderation. If anyone can follow me by consuming content I post to my own site then the main questions are around discoverability. If I'm posting content to a server and service that someone else is directly responsible for, even if federated, they have to consider moderation and the array of different content laws globally.

by viksit 3y ago

flagrant_taco 3y ago

> top level - has to come from communities that are not in an "accepted" list which shouldn't be allowed to form. (the usual suspects like hate speech, CSAM) Even this becomes tricky because you first have to make a common list of banned content and define the lines between what each category of content is. Sure a huge majority of people would like to see CSAM on the list, but they definitely won't agree on what is/isn't CSAM in their opinion I've recently been working through whether any level of content centralization is possible without the fundamental challenge of moderation. If anyone can follow me by consuming content I post to my own site then the main questions are around discoverability. If I'm posting content to a server and service that someone else is directly responsible for, even if federated, they have to consider moderation and the array of different content laws globally.