A couple points. First, I think you guys might be looking for consistency in the wrong place, on a level where it's impossible. I mean the level of specific moderation calls. There are too many factors, such as plain randomness, inevitable subjectivity, and (yes) moderator error, for the set of specific moderation calls to ever be consistent. And that's without even considering that the sample set that each reader chooses to decide "is moderation consistent or not?" is biased (because people tend only to notice the cases they dislike - see [1] for what I mean by that, if interested).
The right place to look for consistency is at the level of moderation principles, and there I do think we've been pretty consistent over the years. Do we apply the principles optimally? No. Do we mostly apply them ok? I hope so (if not, I probably shouldn't be doing this job). Certainly I've spent much time and energy trying to explain what the principles are.
There's another point which is important here. Unfortunately it's more subtle and I'm not sure I can explain it well but I'll try:
There is a temporal decay of interestingness in any sequence of related stories. Curiosity withers under repetition, so we can't have too much repetition [2], and that means we can't have too many predictable sequences [3].
When you have a sequence of related stories (S1, S2, ... Sn), once S1 has had significant attention, S2 becomes less interesting (in HN's sense of the word) until enough time has gone by. This, for example, is why we downweight follow-ups [4]. Time counteracts repetition ("everything old is new again"), so letting enough time go by is one solution [5], but it's not always possible and anyway takes longer than people usually want it to.
What this means is that we can't treat related stories consistently, because how interesting they are doesn't only depend on the story—it also depends on what else has been discussed recently. In itself, S2 might be a better story than S1 is, but if they're related enough and S1 was discussed recently, then S2 becomes less interesting, qua HN topic, than it otherwise would have been. If you take seriously the principle of avoiding repetition, that is how we have to moderate. If we didn't, then the same few themes (the hottest ones) would dominate the site.
It is something of a lottery which story (S1 or S2) shows up first and thereby "wins". But if you only consider the articles, and not the sequence, this is inconsistent! "Why is S1 on-topic while S2 is not?" is thus a common question.
As moderators we're more concerned about the overall functioning of the site (e.g. not having too much repetition) than we are about specific stories. Users, on the other hand, are concerned with specific stories, and rightly so—why should they care about the global state of the site? It should just be there and be good enough.
This disconnect is mostly a background thing, but it flares up when users are personally interested in S2 and don't see why S1 got to "win" and now S2 has to suffer. This is a consequence of mod attention and user attention being scoped at different levels. It's our job to care about the global state while users' job is to care about what interests them (specific stories). To a reader who cares specifically about S2 (and we all have our S2s), this feels like unfair prejudice.
To treat all stories consistently, we'd have to go back and rearrange the sequence (S1, S2, ..., Sn) over time. That's not doable, and from a moderation point of view, not so important either. There is an endless stream of stories in every category. Few matter much in the long run. We try to make sure that the major ones get discussed (e.g. right now, the launching of the $500B data center project and the Ross Ulbricht pardon) but that too is subjective. I'm sure that some commenters in this thread feel like the Musk video is more important than those.
What does all this mean? Maybe it means that people are right that the mods are inconsistent, error-prone, and biased, but a bit less so than at first appears.
this makes total sense to me. but the key takeaway for me is this: There is an endless stream of stories in every category. Few matter much in the long run.
everyone has their own favorite topics and if i see posts with my preferred topics not get traction i am disappointed. but there are so many reasons why that may have happened, it's not worth losing sleep over, much less blame moderation.
there is no agenda here to promote the right stories and hide the wrong ones. the only goal is to promote engaging discussions. those discussions are why i am coming here. i will also admit to often checking comments first exactly because i want to see if there is an engaging discussion that i would want to join.
for this particular topic it is morbid curiosity to see if an engaging discussion will ever happen. so far it hasn't, which matches my expectations. (well, except for this sub thread about moderation)
The right place to look for consistency is at the level of moderation principles, and there I do think we've been pretty consistent over the years. Do we apply the principles optimally? No. Do we mostly apply them ok? I hope so (if not, I probably shouldn't be doing this job). Certainly I've spent much time and energy trying to explain what the principles are.
There's another point which is important here. Unfortunately it's more subtle and I'm not sure I can explain it well but I'll try:
There is a temporal decay of interestingness in any sequence of related stories. Curiosity withers under repetition, so we can't have too much repetition [2], and that means we can't have too many predictable sequences [3].
When you have a sequence of related stories (S1, S2, ... Sn), once S1 has had significant attention, S2 becomes less interesting (in HN's sense of the word) until enough time has gone by. This, for example, is why we downweight follow-ups [4]. Time counteracts repetition ("everything old is new again"), so letting enough time go by is one solution [5], but it's not always possible and anyway takes longer than people usually want it to.
What this means is that we can't treat related stories consistently, because how interesting they are doesn't only depend on the story—it also depends on what else has been discussed recently. In itself, S2 might be a better story than S1 is, but if they're related enough and S1 was discussed recently, then S2 becomes less interesting, qua HN topic, than it otherwise would have been. If you take seriously the principle of avoiding repetition, that is how we have to moderate. If we didn't, then the same few themes (the hottest ones) would dominate the site.
It is something of a lottery which story (S1 or S2) shows up first and thereby "wins". But if you only consider the articles, and not the sequence, this is inconsistent! "Why is S1 on-topic while S2 is not?" is thus a common question.
As moderators we're more concerned about the overall functioning of the site (e.g. not having too much repetition) than we are about specific stories. Users, on the other hand, are concerned with specific stories, and rightly so—why should they care about the global state of the site? It should just be there and be good enough.
This disconnect is mostly a background thing, but it flares up when users are personally interested in S2 and don't see why S1 got to "win" and now S2 has to suffer. This is a consequence of mod attention and user attention being scoped at different levels. It's our job to care about the global state while users' job is to care about what interests them (specific stories). To a reader who cares specifically about S2 (and we all have our S2s), this feels like unfair prejudice.
To treat all stories consistently, we'd have to go back and rearrange the sequence (S1, S2, ..., Sn) over time. That's not doable, and from a moderation point of view, not so important either. There is an endless stream of stories in every category. Few matter much in the long run. We try to make sure that the major ones get discussed (e.g. right now, the launching of the $500B data center project and the Ross Ulbricht pardon) but that too is subjective. I'm sure that some commenters in this thread feel like the Musk video is more important than those.
What does all this mean? Maybe it means that people are right that the mods are inconsistent, error-prone, and biased, but a bit less so than at first appears.
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
[2] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...
[3] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
[4] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
[5] That's why HN allows reposts after a year or so (https://news.ycombinator.com/newsfaq.html).