4 ms·
For this to be useful at scale, you will need to not just design this for good-faith users, but also against bad-faith actors. If this catches on, it will becom
by scott_s 8y ago
For this to be useful at scale, you will need to not just design this for good-faith users, but also against bad-faith actors. If this catches on, it will become a target of coordinated disinformation campaigns. Have you been thinking about what to do against that?
- YazIAm 8y agoIndeed, currently all "verified" reviewers are manually vetted. Long term the goal is to have an open reputation score for every fact checker. I'm currently experimenting with different implementations of this, and seeing good promise.
- scott_s 8y agoHow would the open reputation score mitigate bad-faith actors trying to exploit the system? For example, I believe HN has a vote-ring detector, but I think there's also a decent amount of manual intervention from the mods. Amazon's reputation system is under constant attack through a variety of means (paying teams of people for fake reviews, going to the length of having real transactions on Amazon between fake-reviewer and the seller buying the review).