3 ms·
This article made me wonder about the range of things that these companies are dealing with under the rubric of "moderation" or "trust and safety", particularly
by schoen 11y ago
This article made me wonder about the range of things that these companies are dealing with under the rubric of "moderation" or "trust and safety", particularly the way that companies can have so many different kinds of incentives to remove content from a platform. (Three that come to mind are "this may create legal liability", "this may damage our brand", and "this isn't what many of our users would like to see".)
If the Internet or the way that people commonly use it is going to continue to get more centralized, I hope platforms improve their ability to distinguish between the problem of "things people don't want to see" and "things people don't want other people to see", or, to put it another way, between "please don't show me things like this" and "please don't allow people to publish things like this". While users themselves might not consistently draw the distinction between the two, platforms, in principle, could.
The article touched in this issue in its discussion of the extreme variation in cultural norms, and the likelihood that one culture's vulgarity is another culture's lyr—er, that one culture's distaste for something will lead to considerable pressure for platforms to squelch it for everybody.
A common idea is that platforms have a right and even a responsibility to define their own community standards and then people can choose platforms that they prefer or that best suit them, much as people can choose the newspaper whose editorial policy and biases they find most agreeable. I think this notion has a lot to recommend it but it seems less comfortable as the patterns of people's usage of the Internet becomes ever more centralized; it also raises the question of what things can be considered infrastructural enough that people can reasonably expect (or at least accept) complete content-neutrality from them.
The article also provided an interesting reminder that most people are unlikely to be comfortable using communications systems that have no moderation or filtering at all. Among other things, that's an interesting challenge for decentralized and censorship-resistant systems; to become more popular and practical, they'll need to be paired with some ways that people can avoid unwanted communication, beyond spam.