23

The mechanics

Hot Network Questions (HNQ) are questions shown in the bottom right sidebar featuring from other sites of the Stack Exchange network that are considered “hot” by a heuristic that takes into account recent visits, answers, and votes. There can be issues when one of our questions becomes an HNQ because its increased exposure leads to considerably higher number of visitors – many of whom are unfamiliar with our site and our community standards.

Moderators can now remove individual questions from the Hot Network Questions list (announcement). This process is irreversible. After removing a question, we cannot decide that we want to allow it to go hot again.

How can we use this?

Having a meta discussion about whether a specific question should be removed from the HNQs is not a good idea, because by the time the discussion has reached any sort of consensus, the potential damage has probably already been done. So, we moderators have to decide on relatively quickly and on a case-by-case basis whether a question should be removed.

If you feel that any question should not be an HNQ, please flag it for moderator attention and elaborate why. This naturally makes the most sense for questions that are already HNQs (which you can now see in the revision history), but if you think it is likely that a question will become an HNQ, but shouldn’t, feel free to flag it. Please do not do this for questions that you consider in need of editing, cleaning up, closing, or similar. Instead perform or suggest the respective edits, flag the problematic comments or answers for deletion, or flag for closure.

What questions should not be HNQs?

Occasionally, we will remove a question from the HNQs, when there is a serious problem, and given their individual nature, some of these problems are unforeseeable. However, we as a community can agree that we do not want certain types of question to be HNQs and try to remove them as quickly as possible.

Therefore I am asking:

Are there any categories of questions that we generally do not want to be HNQs? If you think so, please suggest categories in the answers:

  • Suggest one category per answer.

  • Do not suggest categories of questions that should be closed or can be salvaged with an edit.

  • Provide a rationale why such questions being HNQs is a problem.

  • Remember that at the end of the day we have to decide on a case-by-case basis, by answering the question: "Do we have this kind of problem on our site?" Therefore a good rationale is more important than precisely defining the category.

  • Use votes to indicate agreement or disagreement with proposed categories.

14
  • 1
    What I really don't like of the post above is the sentence "Occasionally, we will remove a question from the HNQs": Would you mind rephrase it in a way that leaves open the possibility that you won't remove any question, should the community decide in such direction?
    – Massimo Ortolano
    Mar 12, 2019 at 7:15
  • @MassimoOrtolano: While this tool should certainly only be used in exceptional cases (unless we opt to ban certain categories), do you really want to issue a blanket ban on this for everything that may possibly come in the future? (Also see my comment on your answer.)
    – Wrzlprmft Mod
    Mar 12, 2019 at 8:19
  • 1
    Why is this process irreversible? Is this just a bug or is there a meaning behind it?
    – Hatschu
    Mar 12, 2019 at 12:06
  • @Hatschu: That question is better asked at the Meta SE announcement.
    – Wrzlprmft Mod
    Mar 12, 2019 at 12:22
  • Do I have to register there? I tried to comment, but it did not work.
    – Hatschu
    Mar 12, 2019 at 12:23
  • @Hatschu: You need 5 reputation to comment on Meta SE. You also have to make an account there.
    – Wrzlprmft Mod
    Mar 12, 2019 at 12:25
  • Oh no:( Asking questions here feels like a Catch-22:(((
    – Hatschu
    Mar 12, 2019 at 15:36
  • 1
    @Hatschu: Well, usually users that engage in site politics have enough reputation to have the association bonus, so it doesn’t matter. Also, since that Meta announcement is well, meta, you may post an answer about your concern (meta works differently than normal sites in that respect), which requires no reputation. Either way, here is the wrong place, since we did not make this feature. Unless somebody from SE pays us a visit, nobody can answer your question here.
    – Wrzlprmft Mod
    Mar 12, 2019 at 16:54
  • Actually -- idea. A lot of people are saying "it would be great to be able to interact with users when they post the questions, but unfortunately we can't". Would it be possible (or, how difficult would it be) to have a list of tags which are strong de-HNQ candidates, so that when a user is about to post a question with the tag in question, he gets prompted along the lines of "Questions tagged with X have often received (unwanted, unexpected amounts of?) attention from other SE sites as they frequently make the HNQ list. Since the question might be of a sensitive nature, we are now
    – penelope
    Mar 12, 2019 at 18:03
  • (contd.) pre-emptively excluding them from the HNQ list. If you wish for your question to be a candidate for the HNQ list, which might attract the attention from unrelated SE sites, please tick the box [BOX] before posting your question. ... basically... would it be possible for us to be there at "exactly the right moment" after all?
    – penelope
    Mar 12, 2019 at 18:06
  • 2
    @penelope: If you want any technical changes, that’s something for Meta Stack Exchange, though I doubt that you proposal will be realised. It is a lot to take in for an edge case of something that may happen to an asker’s question. Without technical changes, this is even more difficult as (most) people do not read tag descriptions.
    – Wrzlprmft Mod
    Mar 12, 2019 at 19:27
  • 1
    @penelope: For a new user, this text would be defenitely to difficult to understand and evaluate the pros and cons of this HNQ thing.
    – Hatschu
    Mar 12, 2019 at 22:58
  • Do we have stats about which questions (or general topics/tags) have hit the HNQ and become dumpster fires of bad answers/comments, frivolous flaging, etc.? Some sort of analysis of where the problem lies can help determine where the solution needs to be applied.
    – R.M.
    Mar 25, 2019 at 19:28
  • @R.M.: That data would be difficult to acquire because it requires access to flag data, deleted comments, etc.
    – Wrzlprmft Mod
    Mar 25, 2019 at 20:38

6 Answers 6

Reset to default
41

I propose to remove all questions around the topic of suicide or severe psychic health problems from the HNQs.

Rationale:

  • For many HNQ visitors this is a dire topic they do not want to be confronted with, to the extent that it could trigger problems itself. Non-HNQ visitors of our site implicitly accept a certain risk of running into this topic by visiting a site on social topics. I strongly suspect that titles mentioning suicide and similar are filtered out anyway, so this is for the questions where these filters fail.

  • If the asker is suffering from such problems themselves, they almost certainly do not enjoy the extra attention HNQs give them. Also there is an increased chance that some idiot leaves a hurtful comment.

2
  • 3
    As well as an increased chance of some idiot leaving a full-blown, deeply misinformed answer.
    – E.P.
    Mar 17, 2019 at 0:57
  • This is an excellent case for it. We don't want to be seen as source of knowledge for these topics IMO.
    – user105707
    Mar 18, 2019 at 12:48
25

I propose to remove all questions where the asker is a victim of sexual discrimination or misconduct from the HNQs. Note that this a more narrow category than what Strong Bad proposed. For example, I would leave questions on how to battle sexual discrimination, avoiding conflicts of interest, and dealing with such issues as a third party.

Rationale: While the topic of sexual discrimination and misconduct tends to attract trolling and other problems in general, I think this is a price we should pay for the positive effect of raising awareness on these issues. Silencing the entire topic is exactly what some of the aggressors want. However for questions by victim, the benefit of awareness is outweighted by the chance of hurtful comments that doubt the asker’s assessment, directly attack the asker, or even blatant trolling. This is usually the last thing the asker needs in such a situation.

9
  • 1
    To clarify... you'd allow a question (just as an example) "How to raise awareness of discriminating behaviour in a male-only lab" if the asker was asking a hypothetical, or the asker was male, or e.g. a team-leader trying to implement preventive measures, but remove the question "How do I make my all-male colleagues aware of their mildly discriminating behaviour/atmosphere in the lab?" even though any answer to either of those questions would answer the other one? Where would such a question fall if "asked for a friend", or "asking after visiting a different lab for 4 days"? Seems... a bit odd
    – penelope
    Mar 12, 2019 at 14:31
  • 1
    @penelope: I can understand that this seems odd if you think of this as denying somebody the joy of having an HNQ, but this is about protecting people who already went through some pain from more pointless pain. So at the end of the day, what matters is how likeley we consider it to be that the question being hot will hurt the asker. It strongly affects those odds whether the asker is the victim or just a third party (friend, team leader, etc.), so yes, I would take that into account.
    – Wrzlprmft Mod
    Mar 12, 2019 at 16:52
  • 3
    @penelope: Now, I acknowledge that to some extent we are patronising the asker here by giving them a protection they did not necessarily ask for, and in an ideal world I would let the asker make this decision. However, this requires the asker to be present at the right moment, to know what is about to be unleashed unto them, and a moderator to be present to enact their decision, once made. In my experience, this is rarely the case. We may leave a comment along the lines of: “If you do not say otherwise, I will de-HNQ in fifteen minutes.”, but that’s about it.
    – Wrzlprmft Mod
    Mar 12, 2019 at 16:53
  • 2
    @penelope: Finally, that’s a good example why drawing the lines for categories is less important than the rationale (see my penultimate point in the question). If we have are talking about a “mildly discriminating atmosphere” and the asker seems that they wouldn’t be hurt by the HNQ crowd or even has a bring-it-on attitude, I would probably not de-HNQ the question.
    – Wrzlprmft Mod
    Mar 12, 2019 at 16:53
  • 4
    I was just trying to understand what you proposed. I can get behind your rationale much more than your proposed "category". I guess, in a sense, what I am thinking about is that many people at least somewhat familiar with SE purposefully anonymise or generalize their question to better hit the "lasting value" criteria. But, thinking about it, I guess who you are proposing to protect are mostly distressed, first-time contributors.
    – penelope
    Mar 12, 2019 at 17:55
  • 4
    And actually I think I kind of agree with that - I would possibly even "despecialize" it further and say that de-HNQing candidates would be questions where the asker is distressed about an emotional or controversial topic (I'm not fully happy with "emotional" and "controversial" as my descriptors)... And I guess I'd extend the same to your bottom suggestion - e.g. I do not see a reason to de-HNQ a post-doc asking how to best deal with and support his (medically) depressed PhD student, as it is nowhere near as emotionally charged as if the student was asking it.
    – penelope
    Mar 12, 2019 at 17:58
  • 1
    @penelope: I was just trying to understand what you proposed. – No worries, your question did provoke some (hopefully) good exploration of how we have to compromise between fairness, minimal intervention, protecting users, etc. in this case.
    – Wrzlprmft Mod
    Mar 12, 2019 at 19:39
  • 2
    de-HNQing candidates would be questions where the asker is distressed about an emotional or controversial topic – I am not sure whether I agree. For many other topics (e.g., somebody whose paper got rejected, somebody who failed a course), askers are usually not subject to strong criticism or other hurtful comments from the HNQ crowd. Of course there may be topics I haven’t thought of, but for those you can write individual answers.
    – Wrzlprmft Mod
    Mar 12, 2019 at 19:41
  • 1
    @penelope: And I guess I'd extend the same to your bottom suggestion – Please comment on that suggestion then to keep the discussion where it belongs. (Also note that bottom and top may change over time here.)
    – Wrzlprmft Mod
    Mar 12, 2019 at 19:42
17

I think we should prevent and from being featured. While these questions are important for our community, I do not think they make for good advertisements. The answers often get the job done, but we don't have a user base that is expert in the nuances of these difficult questions. Further, they often attract poor answers from the HNQ that garner lots of up votes which makes them move from not good advertisement to poor advertising. Finally, they seem to attract a lot of discussion which eventually degrades into rude/offensive ranting, which becomes really bad for everyone involved.

6
  • 2
    Honestly, any *-misconduct question.
    – Azor Ahai -him-
    Mar 11, 2019 at 23:45
  • 5
    I think this is too much and would propose a more nuanced approach.
    – Wrzlprmft Mod
    Mar 12, 2019 at 8:16
  • @Wrzlprmft I agree it shouldn't be a blanket ban on the tags and should be something like what you propose where we look at the question and use our best moderator judgment to decide if the popularity is helping or hurting.
    – StrongBad
    Mar 12, 2019 at 13:25
  • 2
    "they often attract poor answers [...] that garner lots of up votes" - Could you provide a few examples?
    – user
    Mar 14, 2019 at 7:53
  • 2
    This would be such a huge improvement to the quality of the site.
    – Noah Snyder
    Apr 4, 2019 at 1:29
  • Any chance we could actually make this happen? It got a lot of upvotes!
    – Noah Snyder
    Jun 17, 2021 at 21:37
4

Why not "all of them"?

I'm not sure myself that this is the perfect solution, but take this answer as a way to think about the benefits of HNQ.

What good comes to our community from a question being in the HNQ, usually? Most of the time, all that happens when a question enters it is that we get a bunch of votes and answers from people that are not very knowledgeable about academia. They tend to skew the votes, so that they do not reflect the opinion of university people anymore.

Also, the HNQ tends to promote click-bait questions and controversial issues, which (in my opinion) are not an effective way to advertise our site to new potential users. I'd see more value in a curated list of the best questions and answers, than in a contest on who can get the most clicks by attracting the attention of random users.

The HNQ benefits Stack Exchange more than it benefits us as a community, I believe.

8
  • 8
    Well, if it weren't for the HNQ, I wouldn't be here. Whether this is good or bad, I'll leave it to the community! ;-)
    – Massimo Ortolano
    Mar 14, 2019 at 16:02
  • @MassimoOrtolano Are you sure? Maybe you'd have found the site anyway, since you seem genuinely interested in discussing these topics. For instance, by asking yourself "hey, maybe there is a SE site for academia, too". Maybe if we had a "curated list of the best questions and answers" instead of the clickbait ones, you'd have arrived here anyway, or even sooner.
    – Federico Poloni
    Mar 14, 2019 at 16:54
  • 1
    Also, I'd tend to distrust anecdotal examples, because there is an inherent bias in this kind of comments: we can find examples of user who are here because of how things work now, but not examples of users that are not here but might have been if we had a different system instead.
    – Federico Poloni
    Mar 14, 2019 at 16:56
  • This is now a possibility, but in order for it to be done we would need to have a separate meta question about it. Do you want to ask it?
    – StrongBad
    Mar 14, 2019 at 18:03
  • @StrongBad Let's use this answer to test the waters; looking at its score, for now it seems like it's not going to be a popular idea.
    – Federico Poloni
    Mar 14, 2019 at 18:55
  • @FedericoPoloni Yep! Even though I'm genuinely interested in these topics, I wouldn't actively searched for the site. At the time, I just used to search TeX.SE to solve LaTeX issues (but at the time I wouldn't have asked any question) and I've noticed an Academia question on the HNQ list. Actually, I joined all the communities in which I'm active now through the HNQ.
    – Massimo Ortolano
    Mar 14, 2019 at 20:11
  • 4
    @Massimo Ortolano Opposite anecdote: I found some of the HNQ-popularized and politicized threads in another stack so off-putting, that I've actually deleted my account there.
    – henning
    Mar 30, 2019 at 8:57
  • 1
    @henning What generally keeps me away from a community are not specific threads but the general attitude of the community and the type of moderation. I'm a practical person and participating to SE is a hobby: if the attrition passes a certain threshold, I do something else.
    – Massimo Ortolano
    Mar 31, 2019 at 16:51
2

Please, don't remove questions from the HNQ list at all

I think that the HNQ is a non-issue and we should not exclude anything, just leave the algorithms do their work (whether good or not). So, I add this as an answer because I think that we should give the community also the possibility to choose this option.

13
  • Agree -- not clear to me why certain questions would be worse than others (regardless of the question, the potential problems are the same: answers/voting by unqualified people, spam, etc.), and I don't think we want to opt out of HNQ entirely (lots of SO users find us that way). I'll follow this discussion, maybe someone can give an example and I'll see that I'm missing something.
    – cag51 Mod
    Mar 11, 2019 at 22:01
  • 3
    My guess is that a large percentage of our huge comment chains are on featured questions (maybe we can pull data to see if that is true). There seems to be a class of HNQs that attracts garbage (see my answer).
    – StrongBad
    Mar 11, 2019 at 23:06
  • I do not think that this pre-emptive blanket answer is needed or a good idea. If you or the community disagree with every single category proposed for a general ban, that’s fine, but it does not need this answer to reach that conclusion. On the contrary, if this answer gets a high number of upvotes due to being posted early and later somebody suggests a category that meets broad agreement, how shall we divine the community consensus?
    – Wrzlprmft Mod
    Mar 12, 2019 at 8:14
  • As for per-case decisions: If in the future we moderators come to the conclusion that there is a pressing and urgend need to remove some individual question from the HNQs due to issues that are very individual and not foreseeable, should we refrain from doing so because the community decided we should not use this tool at a time where nobody had thought of this individual problem?
    – Wrzlprmft Mod
    Mar 12, 2019 at 8:14
  • @Wrzlprmft and StrongBad The problem is upstream: too many people post on the internet troublesome issues without thinking at the consequences, for them or for the others. Whether something goes or not on the HNQ list is totally irrelevant, the post is already "out there", and those who are interested in the case usually do their best to find it (I tell you this by experience). Even dissociating a post from an account is largely irrelevant, especially if you take into account the latency time that these kind of actions require.
    – Massimo Ortolano
    Mar 13, 2019 at 20:35
  • @StrongBad and Wrzlprmft: If you want to reduce the consequences of problematic post, you should work with SE to limit the problem upstream, by educating people to not write posts too hastily and by having tools that allow to act in very short time. So, yes, even if this answer is not going to be popular, I think it should stay.
    – Massimo Ortolano
    Mar 13, 2019 at 20:39
  • 2
    Users and moderators across the network have been working with SE to limit the problem. The best solution they can come up with right now is to let us remove questions (IMO, this is way better than protecting, closing, and /or down voting questions to minimize the impact). It doesn't solve the underlying problem, but you are saying don't use the tool. I think the tool can still be useful in a lot of cases (but that is what this question is about).
    – StrongBad
    Mar 13, 2019 at 20:44
  • @StrongBad "I think the tool can still be useful in a lot of cases": I think not, that's my point, or it's usefulness is just illusory, in the sense that it doesn't prevent the real damage.
    – Massimo Ortolano
    Mar 13, 2019 at 20:56
  • 1
    @MassimoOrtolano: too many people post on the internet troublesome issues without thinking at the consequences, for them or for the others. – This not a case I have been considering when writing the question or any of the answers. I did not suggest that removing something from the HNQs solves problems of this kind, and neither did anybody else as far as I can tell.
    – Wrzlprmft Mod
    Mar 13, 2019 at 22:37
  • @Wrzlprmft Then I don't understand what kind of damage you're referring to in the OP.
    – Massimo Ortolano
    Mar 13, 2019 at 23:25
  • 3
    @MassimoOrtolano: For example, see this this answer of mine: We deleted a lot of inappropriate comments and answer directed towards vicitims of sexual abuse or discrimination in the past coming from users who almost certainly found the question through the HNQs. Also problems with HNQs do not necessarily have to be about the asker at all.
    – Wrzlprmft Mod
    Mar 14, 2019 at 6:26
  • 4
    I disagree with this answer. "the post is already "out there", and those who are interested in the case usually do their best to find it" - maybe, but numbers matter. A sharp influx of visitors with reduced investment in this site and the civility of its discourse (let alone familiarity with academia) is an excellent way to increase the proportion of trolls and other less-than-helpful visitors. It's a given that deHNQing a question isn't a silver bullet that will protect it from every troll on the internet, but that doesn't mean it doesn't help.
    – E.P.
    Mar 17, 2019 at 1:04
  • 1
    I'll also point out that trolls don't pick their targets uniformly. I feel like I draw considerably more obnoxiousness than other people who express similar opinions but who have an identifiably male username. Fewer low-investment trolls is much better from my perspective.
    – Elizabeth Henning
    Jan 31, 2020 at 23:33
-4

Anything that's a soft interpersonal or sociopolitical issue that doesn't require any academic expertise to answer.

As a Stack Overflow user who is neither an academic nor a user of Academia Stack Exchange, I find Academia's presence in HNQ consistently frustrating. Pretty much all from here that currently reaches HNQ is soft questions about interpersonal interactions that require no particular expertise to opine on. Yet when we users of other sites click through to them, we usually find ourselves pre-emptively silenced by the question being "Protected".

It's bad for us, because we suffer the frustration of not being permitted to answer with differing perspectives; on these broad questions about how to decently interact with other human beings, only the (disproportionately left-leaning and otherwise atypical) views of actual users of this site are permitted to be voiced in the answer section. That leaves us with only comments as a permitted way to engage with the issues, but those comments get nuked seemingly capriciously by the mods.

And it's bad for you, because it channels users into precisely the interactions you don't want. You end up with a horde of users who are just as qualified as you to answer the questions they're reading, but who you only permit to use comments to do so, turning every such squishy interpersonal question into a brawl in the comments section that your mods need to clean up.

Assuming that you're unwilling to simply stop using question protection on these questions, the other way you can stop this dysfunction by simply not letting these questions be in HNQ. If there's no particular expertise that someone who works in academia can bring to a question compared to a random member of the public, nuke it from the sidebar. Then you guys get to have the controlled discussion of social issues amongst yourselves that you seem to repeatedly want to have, free from disruption by the rest of us, and we get to go about our lives in blissful ignorance and with slightly lower blood pressures.

6
  • 2
    If your blood pressure is affected by what goes in the HNQ list, you may want to rethink at your priorities in life ;-)
    – Massimo Ortolano
    Mar 21, 2019 at 17:03
  • @MassimoOrtolano That may well be true.
    – Mark Amery
    Mar 21, 2019 at 17:13
  • 6
    I disagree with many of your premises: 1) The point of HNQs is not to solicit input from people all over the network. 2) Answers from HNQ visitors that completely ignore the intricacies of academic culture (or just troll) are a common thing. Protecting questions reduces this problem. 3) Answers in comments from users that cannot answer a protected question are not a huge issue. Most answers in comments come from users who are not affected by protection. Most comments by users that are affected by protection would not be valid answers (often they are not valid comments either).
    – Wrzlprmft Mod
    Mar 21, 2019 at 18:57
  • 6
    4) Whether we like it or not, academic culture and relations are special (e.g., a supervisor–supervisee relation differs from an employer–employee relation in many aspects). This inevitably permeates almost all of our questions. Sure, sometimes you arrive at the same result when ignoring the academic context, but that does not mean that the way is the same. Only a posteriori one can decide that the context does not matter.
    – Wrzlprmft Mod
    Mar 21, 2019 at 18:58
  • 1
    @Wrzlprmft The point of HNQ may not be to solicit input from outsiders, in theory, but in practice that is one of its main effects. And this "input" comes not only in the form of answers (where it is mitigated by protecting), but also in the form of votes (where it is not).
    – Federico Poloni
    Mar 31, 2019 at 17:53
  • @FedericoPoloni: but also in the form of votes – Well, this answer’s argument is particularly about answers, not about votes.
    – Wrzlprmft Mod
    Mar 31, 2019 at 18:21

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .