3 ms·
Hey Dang, I appreciate the transparency and general policy in such situations. As well as undowning the story Do you think there will ever be in band signaling
by zamadatix 2y ago
Hey Dang, I appreciate the transparency and general policy in such situations. As well as undowning the story
Do you think there will ever be in band signaling of when (or perhaps even why) a post was artificially depressed by moderation? Particularly aimed at cases where people are already using tools like https://news.social-protocols.org/stats?id=40710417 https://news.social-protocols.org/stats?id=40710417 on posts like this where more people have already seen the story drop off and confirmation that it was via moderation than will probably ever see this comment in its later revival explaining how or why. Something like a "downed" with hover alt text next to the flag tag even. Apart from being immediately clearer to everyone without waiting for a comment/email that helps solidify against "and then 5 hours later they back tracked it slightly but not all the way with some other excuse why after being called out" type thinking/arguments. I mean you'll never get rid of that completely but it's a lot more powerful that the message existed there at T=30 minutes than T=5 hours.
Would typically email but this seemed like something good for an open message :).
- CobrastanJorji 2y agoYeah, as much as the transparency is very much appreciated here, "yes, a mod totally did try to subtly bury this story exactly as you predicted and despite our official mod policies" sounds pretty bad, even if you undid it. Some visible "[Moderator Downvoting]" warning would help earn trust that mods are not in fact doing this to YC-related stories.
- dang 2y agoI checked with the mod and they weren't trying to bury the story but rather to downweight it less than we normally would, in keeping with the #1 rule I described upthread (https://news.ycombinator.com/item?id=40712900 https://news.ycombinator.com/item?id=40712900). The intention was to downweight it enough that it wouldn't be near the top of the front page but not so much that it would leave the front page. That's the same thing I would have done, btw. It looks like what happened was that more user flags came in and that, together with the moderation downweight, moved the story off the front page. Eventually we noticed this and restored it to the lower part of the front page, which is the standard practice for stories of this kind, which we'd never allow to be on HN's front page if they were about anybody else. (Though, as someone pointed out to me, this story wasn't negative about YC.)
- dang 2y agoI understand the theory—surely publishing more data would reassure everyone and help settle things—but this is based on an assumption which in my experience is false: that these dynamics are rational. I don't think that's how they work at all, so I don't have faith that earnestly publishing a moderation log would improve things. On the contrary, I find it easier to imagine it making things worse. Why? Because there would still be all the same suspicions and resentments, and now they'd have more material to work with. I'm tired and doubtful that I've expressed this very well, but I've posted quite a bit about this over the years, so if the present comment doesn't help, perhaps https://news.ycombinator.com/item?id=39234189 https://news.ycombinator.com/item?id=39234189 will, or one of the other comments linked back from there.
- zamadatix 2y agoReading through the linked comment (and it's linked algolia feed) I'll add that public mod actions are less aimed at the making interactions with small and loud subset who will never engage rationally themselves any better and more aimed at helping the rest of the community easily ignore said subset. If the typical rational user will have seen many posts about YC critical news not related to celebrity gossip they will immediately know such claims are unfounded and have public references to it. Right now it's instead much easier to find 10 examples of this type of public comment thread where things may look gray or look at the public data analytics and find 10 examples that appear to confirm the viewpoint that topic $x articles are silently downweighted more often than they should be than it is for anyone but the mod team or those who send constant emails to provide even one bit of pre-existing hard proof to the contrary that doesn't carry the risk of looking like it was formed after the fact. Specifically on a full moderation log your links also show how it would be troublesome in some ways I perhaps hadn't thought as deeply about. After all even if you moderate 9999 things right the same crowd is going to want to parade the 1 - now with more details as to why. At the same time I still think the same set of people are going to complain either way and the data isn't really helpful to them as much as the rest of the people anyways. I realize I'm extraordinarily unlikely to budge your views on this due to volume of experience/thought/interaction with this topic you've had but just count me in the selfish "I'd rather the log/actions were public for us average user and then your half a day of emailing was explaining the details to the troublesome bunch rather than with the average user spending time trying to understand". I'll close with a smaller push of "and even if the site never does a full moderation log, not signaling moderated post/comment actions at any level only feeds that small group on believing everything unpopular they say is being censored 100% of the time instead of x% of the time". With that I'll leave you some peace on the matter :). Thanks for the time and reading through (yet another) user's view on this and for the good moderation of the site as always.