This question appears on the HNQ list right now (I've posted a screenshot here, with the quetion I mentioned, circled in yellow, in case you wanted to see it).

What surprised me was that the question was 21 days old. This is something that I thought was impossible. In July 2020 I asked here if anyone's ever seen an HNQ more than 4 days old and never go a reply.

I understand that the exact HNQ criteria are not publicly available, and that only an approximation was given to us (and this totally makes sense, because otherwise people could cheat the system). However, what I'm seeing today seems to either be a bug or quite a major change to the HNQ requirements, which would be the relaxing of the "maximum 3-days" criterion. Knowing about this change wouldn't make it any easier for people to cheat to get their questions in HNQ as far as I can see, so I think it's appropriate to ask about it.

For the record, I'm not saying that this change is a good thing or a bad thing, I'm just curious if I'm seeing a bug or a fundamental change to how new a question needs to be in order to enter HNQ!

Before asking this, I did look for similar questions, for example by looking at the latest questions in the tag, but couldn't seem to find anything about this asked before.

  • 1
    Unable to reproduce. I don't see the question on the full list of HNQs. Commented Dec 19, 2021 at 23:04
  • I, likewise, just checked and it's not in the current full HNQ's list. Commented Dec 19, 2021 at 23:06
  • 2
    It became a HNQ on the same day it was asked, November 28 (and I vaguely remember seeing it in the list a while ago). Is it possible that you were looking at a page you loaded a while ago, or your browser loaded a cached version? I'll bet the other questions there are similarly old; the one I checked was.
    – Laurel
    Commented Dec 19, 2021 at 23:13
  • @Catija perhaps it would be more appropriate to close this as "not reproducible" rather than as a duplicate? Also, my question itself already has a link to the question you re-directed it to in the dupe, and it's not really correct that "this question already has answers" in the other question. Commented Dec 19, 2021 at 23:25
  • 2
    @Laurel here is a screenshot taken right now, which shows that I edited this question 32 minutes ago, which you can verify yourself, and the HNQ is right there in the same screenshot! Commented Dec 19, 2021 at 23:26
  • 2
    Your question is about how a question that is 23 days old could be on the HNQ list and you seem to think the triggers are secret. The duplicate explains that those statements are incorrect and explains exactly how the list works.
    – Catija
    Commented Dec 19, 2021 at 23:27
  • 3
    @user1271772 I suppose the cache of hot network questions could have been out of date even while the rest of the site was fine. Even still, it does seem like a bug that you got a list of the HNQs from several weeks ago (all the ones I checked became HNQs November 30th or earlier).
    – Tyberius
    Commented Dec 20, 2021 at 3:52
  • @Tyberius what is the likelihood of the cache being ~20 days out of date? Commented Dec 20, 2021 at 20:19
  • 72 hour hard limit: meta.stackexchange.com/q/325060/282094 - Formula: meta.stackexchange.com/a/61343/282094
    – Rob
    Commented Dec 20, 2021 at 21:21
  • @Rob can you explain why the formula didn't work in these cases? meta.stackexchange.com/a/358924/391772 Commented Dec 20, 2021 at 21:33
  • 1
    user1271772, That is a different bug, for which you are not satisfied with the comments replying to your answer. This question (above), which is on-topic in this comment section, I voted to reopen. --- Unless I know it is not a bug then I can't explain it, since I agree that the duplicate doesn't apply, that it's a new bug or that the dupe is incorrect, about all I can do from the review queue is ReOpen; someone else might answer.
    – Rob
    Commented Dec 20, 2021 at 23:08
  • @Rob Thanks. My question was closed in only 7 minutes on a Sunday and by only one vote, which I think is highly unusual for a question like this (I've see way worse take far longer to get closed). I've been told that the dupe "explains exactly how the list works" but the output of the given formula does not match the actual hotness score. Commented Dec 20, 2021 at 23:35
  • 1
    Apparently the complaint started years ago (that the formula doesn't work), resulting in a different complaint being filled. The above is "how did it exceed 72 hours", I agree (as far as we know) that shouldn't happen. Thus, the reopen vote (and comment pointing this out). Note that exits are not annotated.
    – Rob
    Commented Dec 20, 2021 at 23:54


Browse other questions tagged .