14

Stemming partially from this meta thread: Why are we challenging the premise rather than answering the question (question on potential sexist remarks)?

And especially this comment by @Wrzlprmft:

"@Fomite: For whatever it’s worth, most of the problematic comments and answers come from users attracted to the question via the hot network questions and not directly from our own community."

Does the Academia community actually benefit from being in that listing? As far as I can see, the primary benefit is a sudden influx of users, but do we have evidence that they stay? Growth from that listing is only useful growth if the visitors go on to continue to be members of the community. I'd be happy to hear from individual users here who found us via Hot Network Questions.

As far as I can tell, the primary detriment is the questions that end up on the Hot Network Questions listing tend to drop fairly dramatically in quality, become more controversial, etc. I can say, as a fairly active user and contributor to the site that the questions and answers that have made me consider throwing in the towel have all been Hot Network Questions.

For the sake of "Meta-votes are Agree/Disagree" clairity, I'd suggest an Upvote is "HNQs are helpful" and a Downvote is "HNQs are unhelpful."

5
  • 14
    I can't remember exactly, but I suspect I discovered Academia.SE through the HNQ. Whether I improved the community or not, that's another story ;-) Dec 5, 2016 at 23:34
  • 3
    "I'd be happy to hear from individual users here who found us via Hot Network Questions." I definitely did that; though I agree that I can't really be considered as a frequent user, I like to read the questions here, and I was happy to receive good answers to my (so far unique) question. I may very well post other questions (and perhaps even answers) in the future.
    – Arnaud D.
    Dec 7, 2016 at 18:14
  • 1
    The by Shog9 admittedly just for his (and other StackOverflow user's) amusement created hot list is mostly a nuisance, in particular for serious research-level academic sites. Unfortunately, not even MathOverflow was able to opt out from this SE feature, see meta.mathoverflow.net/questions/1305/… and related discussion ...
    – Dilaton
    Dec 8, 2016 at 11:25
  • 4
    I wrote the same question last week after seeing a wave of users with no history on this site, and visibly no connection to academia whatsoever, creating accounts just to write confrontational comments. Then I ditched it because I thought I attracted enough junk with my last meta post. So, now I want to upvote because I like the question, but I voted down because I think HNQs are clearly detrimental to the quality of our site.
    – Cape Code
    Dec 8, 2016 at 12:08
  • 11
    The problem is not necessarily HNQ, it's the fact that the visitors from other sites are likely to acquire association bonus and have site privileges without being site users. There are various proposals on Meta.SE to address this root cause (including mine - shameless plug)
    – DVK
    Dec 10, 2016 at 1:28

5 Answers 5

15

From what I've seen: At best, they've been slightly helpful. At worst, they're a temporary nuisance.

  • Moderately helpful – As Massimo pointed out, he came to this site through the network questions. This isn't so terribly unusual; I would guess that for every fifty network questions, we gain one user who didn't know we existed. We've been on there a lot lately, so over time that sort of thing does prove useful to the community. It also gives people an exposure to what Academia is like, although sometimes the questions are slightly (often?) more inflammatory than they need to be.

  • Temporary nuisance – Network questions always bring a lot of people, who leave all sorts of often bizarre comments and answers. These people rarely, if ever, return, and don't really add much to the community, or even that question. They frequently cause all sorts of automatic flags to get set off (tons of comments, tons of answers, low quality new user posts) which are a stupid and a pain. These questions also often end up with a bunch of cruft (old useless comments, bad answers), which leads to all sorts of flags being raised on these questions months or years later.

Someone posted the following to Meta a while back: What is the Goal of "Hot Network Questions"? This was one of those questions where the question had a lot more upvotes than any of the answers. My theory is that this place is more like Facebook and Twitter than the admins care to admit, and the only goal of Hot Network Questions is to keep people entertained, thereby keeping them on the site, thereby showing them more ads, thereby making a profit. Whatever the reason, they're there, they're probably not going away anytime soon, and while pontificating is fun it's probably not going to change anything.

6
  • 2
    I'd suggest that, with the goal of SE answers being useful for more than the OP, that the HNQ rendering the occasional question in a very poor state is more than a "temporary" nuisance. We'll also never have numbers, but I'd be curious at the people who found us via HNQ and were turned off by what they found. The hypothetical anti-Massimo.
    – Fomite
    Dec 6, 2016 at 21:19
  • @Fomite - *shrug* You may be right, but we'll never really know.
    – eykanal
    Dec 6, 2016 at 21:37
  • 8
    It's been my experience that poorer quality questions tend to rise to HNQ list in general (NOT on Academia alone, but on other sites as well). A question that ignites controversy seems to feed into "hot" algorithm.
    – DVK
    Dec 10, 2016 at 1:30
  • 2
    "keep people entertained" - very likely, on the other hand, is it a bad thing to keep people entertained with somewhat productive/educative questions/answers?
    – Nobody
    Dec 14, 2016 at 17:33
  • You probably won't like to heard that Academia is "Today's Featured Site". :D ;-)
    – UTF-8
    Dec 17, 2016 at 14:04
  • Like Massimo, I probably came here via HNQ as well. I admit to using the HNQ as a source of entertainment, although my entertainment primarily comes from the "somewhat productive/educative questions/answers" of math, matheducators, academia, and workplace, with a smattering of money and travel, rather than movies, scifi, politics, or skeptics (not that those aren't entertaining in their own right).
    – shoover
    Dec 19, 2016 at 16:22
7

I don't believe an objective answer to this question is possible. The system does not track information about hotness points, questions on the HNQ list, or the use of the HNQ sidebar: Add an audit log to record when particular question enters and leaves hot list

A number of times, one of our question has been featured on the HNQ list that I thought was neither representative of our community nor a particularly good question. This has happened enough that I proposed a feature be to prevent question from being added to the HNQ list: Allow mods or gold tag badge holders to prevent question from being on hot network questions list

Without the data, I am not sure how to quantify the affects on our community of these "bad" questions being featured on the HNQ list. I think that there have been "good" questions featured and that publicizing our good side is a good thing. What I would like to see is more of our good questions make the HNQ list.

2
  • Most of our meta-questions don't really have objective answers. And yes, while what you'd like to see is probably the ideal case, it's not what's currently happening.
    – Fomite
    Dec 6, 2016 at 0:17
  • 2
    @Fomite it wasn't a complaint. I would have loved to be able to show (or not show) a spike in new users who eventual became high rep or a prolonged bump in traffic or new questions or ... The data just isn't there.
    – StrongBad
    Dec 6, 2016 at 0:18
6

On meta.SE I proposed a mechanism to opt-out specific questions, either by smarter algorithms that watch out for polemical questions or by moderation tools. (I will link to this question there.)

The "hot" questions algorithm should use logic to avoid controversial questions

I am upvoting your question to reflect my view that there are problems with HNQ but I would rather work toward salvaging the feature.

Truthfully though - I think there are bigger "respect" problems on SE and I do not think the umbrella community can escape developing better tools to deal with these problems. HNQ woes are a bit more on the "symptom" side of the scale. I am brooding putting together a larger post on this topic but naturally such a thing must be done very carefully :) but for now you may keep this sentiment in mind.

1
  • If indeed there was a more aggressive way to filter, I think that would indeed be favorable.
    – Fomite
    Dec 5, 2016 at 23:35
2

Obviously personal anecdotes aren't a good way to make rules, but I can say without a doubt that I only found Academia SE through the hot network questions sidebar.

I was only ever engaged on Stack Overflow, which I think is a fairly typical gateway to Stack Exchange in general. In fact, it seems to me computer science, and fields that make a lot of use of computer science, are very heavily represented on Academia in proportion to other fields. This observation would fit that hypothesis.

So, I feel like it's a pretty solid positive despite the increased moderation necessary on the questions that become sacrificial lambs by ending up there. After all, closing the community off from the influence of "outsiders" doesn't sound like a desirable thing to do. Maybe the question can be addressed by improving moderation tools on protected questions?

4
  • Are there stats on what % of new users on Ac.SE were attracted by HNQ?
    – smci
    Dec 11, 2016 at 3:10
  • @smci How do you propose to define and measure "attracted by HNQ"? Dec 15, 2016 at 8:41
  • @FedericoPoloni: the people who brought us HNQ can measure its impact. For them they know the referring URL, it should be trivial. And they have the capability to selectively disable it on those sites where it's causing grief.
    – smci
    Dec 16, 2016 at 7:16
  • So the definition is "an user was attracted by HNQ if, in the session in which they registered their account, they reached this site via a HNQ"? That is measurable, but it might not tell the whole story. Dec 16, 2016 at 7:32
-1

While I agree with the existing answers on and in particular that there should be some mechanism to avoid problematic HNQs, I would like to mention another advantage of HNQs:

They allow good posts (questions as well as answers) to get more attention than they normally would, which in turn results in upvotes and badges. And at the end of the day, this is one of the central mechanisms of Stack Exchange: You get to know how many people found your contributions helpful and are rewarded (with reputation, badges, and sometimes hats). When a post of mine gets massively upvoted due to being a HNQ or an answer to an HNQ, this does not only mean reputation and badges but also (well, at least most of the time) that a large audience learnt something from my contribution, be it some hard information or a way to look at things. And that’s extremely rewarding and motivating.

2
  • 4
    I actually see this as a disadvantage. It means a post can get massive upvotes not so much because it is particularly good relative to other posts on the site, but just because HNQ gave it disproportionate visibility. For instance, my top-voted answer is one that basically just says "No." It actually makes me feel less motivated when simple things like this get tons of votes, while posts that involved much more effort get less attention. Dec 11, 2016 at 16:37
  • 1
    I agree with @NateEldredge - I'd rather a post get high upvotes based on quality rather than external volume. And it incentivizes the wrong sort of behavior.
    – Fomite
    Dec 13, 2016 at 5:46

You must log in to answer this question.

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