The story the OP is complaining about was flagged by users. Moderators never saw it (edit: wrong, we put 2010 on the title by mistake, see downthread [1]). Had we seen it, we would have turned off the flags. There's a long tradition of people looking at HN data and posting about it. Edit #2: since the 2010 thing was our mistake (an accident of sleep deprivation by the looks of it!) I've invited foob to repost the original article using the second-chance mechanism described at
https://news.ycombinator.com/item?id=11662380 and the links there. I think he's planning to do that tomorrow.
The [flagged] annotation only shows up on stories that are heavily flagged, i.e. enough to kill the post. User flags have downweighting effects long before that.
Story rank on HN is determined by upvotes, flags, software, and moderators. Moderators downweight stories to the degree that they don't fit the site guidelines. This doesn't happen by upvotes alone, unfortunately; certain stories routinely get tons of upvotes regardless of how good they are for HN—e.g. anything sensational, indignant, or meta. If we didn't have a compensating factor, those stories would dominate the front page every day and HN would no longer follow its primary rule: "anything that gratifies one's intellectual curiosity". Of course that means HN is subject to our interpretation of what counts as "intellectual curiosity". HN has always worked that way; before we did it, pg did, and he trained us to do it so it would work as before. There's no way around the need for moderator intervention on a site like HN—the clue is in the word 'moderator' itself: left to its own devices the system runs to extremes and it needs a negative feedback loop to dampen it.
When YC is involved, we do this less than usual as a matter of principle. When HN itself is involved it's a little bit different, because the hypnotic power of all things meta causes HN upvoters to go into an upvoting trance. Meta on forums is basically crack, so we routinely downweight such posts—but only so much, to compensate for the crack effect. That's what I've done here, which is why the post is now at #7 rather than #1. It should probably be lower, but I want to illustrate the point that we intervene less, not more, when judgments about ourselves are involved. As a further example, a moderator actually turned off software penalties and user flags on this post this morning, which is probably why it went to #1 in the first place. That's more than I would have done but it shows how seriously we take that principle.
None of this is new information, btw. I've posted about it plenty over the years and am always happy to answer questions.
https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=...
https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=...
https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=...
https://hn.algolia.com/?sort=byPopularity&prefix&page=0&date...
1. https://news.ycombinator.com/item?id=13858850
The [flagged] annotation only shows up on stories that are heavily flagged, i.e. enough to kill the post. User flags have downweighting effects long before that.
Story rank on HN is determined by upvotes, flags, software, and moderators. Moderators downweight stories to the degree that they don't fit the site guidelines. This doesn't happen by upvotes alone, unfortunately; certain stories routinely get tons of upvotes regardless of how good they are for HN—e.g. anything sensational, indignant, or meta. If we didn't have a compensating factor, those stories would dominate the front page every day and HN would no longer follow its primary rule: "anything that gratifies one's intellectual curiosity". Of course that means HN is subject to our interpretation of what counts as "intellectual curiosity". HN has always worked that way; before we did it, pg did, and he trained us to do it so it would work as before. There's no way around the need for moderator intervention on a site like HN—the clue is in the word 'moderator' itself: left to its own devices the system runs to extremes and it needs a negative feedback loop to dampen it.
When YC is involved, we do this less than usual as a matter of principle. When HN itself is involved it's a little bit different, because the hypnotic power of all things meta causes HN upvoters to go into an upvoting trance. Meta on forums is basically crack, so we routinely downweight such posts—but only so much, to compensate for the crack effect. That's what I've done here, which is why the post is now at #7 rather than #1. It should probably be lower, but I want to illustrate the point that we intervene less, not more, when judgments about ourselves are involved. As a further example, a moderator actually turned off software penalties and user flags on this post this morning, which is probably why it went to #1 in the first place. That's more than I would have done but it shows how seriously we take that principle.
None of this is new information, btw. I've posted about it plenty over the years and am always happy to answer questions.
https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=...
https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=...
https://hn.algolia.com/?sort=byDate&prefix&page=0&dateRange=...
https://hn.algolia.com/?sort=byPopularity&prefix&page=0&date...
1. https://news.ycombinator.com/item?id=13858850