11
$\begingroup$

On occasions, some questions (usually posted by less experienced users) get massively downvoted - say, score of -5 or less.

Given the reputation cannot go below 1, is there a practical site consequence of such massive downvotes?

Certainly some questions completely deserve to be downvoted, but where is the line between being dissatisfied with a specific question and downright discouraging new users?

$\endgroup$
  • 3
    $\begingroup$ Yes, many downvoted/closed/deleted questions can lead to a question ban. There is the same ruleset for answers. Furthermore, people fighting effectively to avoid the Q/A ban, but still contributes LQ posts, can get suspension. $\endgroup$ – peterh says reinstate Monica Oct 11 '17 at 14:36
  • 4
    $\begingroup$ @peterh Do you get banned "faster" if you score -10 rather than -3 on your first question? $\endgroup$ – ZeroTheHero Oct 11 '17 at 14:38
  • 1
    $\begingroup$ I am not sure, but as I can remember, 3 things are considered: 1) question score is negative, 2) question is closed 3) question is deleted. Thus, a downvoted, closed and deleted question counts with 3 black points, while an only downvoted with only 1. The SE doesn't say the exact details of the bans. There are rumors. The most important things: 1) Q-banned user has to undelete the questions he deleted (so they will count only with -2), and working on to edit them 2) if somehow (s)he can collect a few upvotes for them, maybe only 2-3 is enough, then the Q-ban will be lifted. The trick $\endgroup$ – peterh says reinstate Monica Oct 11 '17 at 14:43
  • $\begingroup$ @peterh I'm reasonably sure Zero's point is right there. For the areas you mention, the criterion tends to be "the post has negative score" without distinguishing the -1s from the -200s. Without any evidence, it's all speculation there. $\endgroup$ – Emilio Pisanty Oct 11 '17 at 14:44
  • $\begingroup$ @peterh is here that it is much easier to lift the Q-ban as it seems. Going to the meta, asking them how to fix, and following their advices probably enough. 3) there is also a timeout, maybe a half year or after a year, the q-ban is lifted. | I think you get banned faster with -10 as with -3. This downvoted/closed/deleted calculation is only one of what the system does. $\endgroup$ – peterh says reinstate Monica Oct 11 '17 at 14:45
  • 1
    $\begingroup$ @EmilioPisanty As the SE doesn't disclose the details, there are only speculations. And there I am here, watching the SE carefully and collecting all small information part since years. Furthermore, I think I have an experience, how such big complex IT/human systems work and it makes for me possible to give useful estimations. For example, I predicted much earlier that there is also a timeout rule in the Q/A-ban lifting, that an SE insider admitted it in a since then deleted comment. $\endgroup$ – peterh says reinstate Monica Oct 11 '17 at 14:47
  • 1
    $\begingroup$ @peterh So long as you brand all speculation as such, sure. $\endgroup$ – Emilio Pisanty Oct 11 '17 at 14:48
  • 1
    $\begingroup$ @EmilioPisanty Exactly, these are my speculations, and these speculations are often just so useful as the long rulebooks of the meta sites, despite that I have mostly few to support them. $\endgroup$ – peterh says reinstate Monica Oct 11 '17 at 14:49
  • $\begingroup$ Questions with a score below a threshold (-3 I think?) no longer show up on the main page as new questions. That's a reason at least. $\endgroup$ – JMac Oct 11 '17 at 15:08
  • 3
    $\begingroup$ @peterh There's no need to be all conspiratorial about SE "insiders" "admitting" there's a timeout to the question ban: The help center page on question bans straight up tells you that the ban is lifted after 6 months. $\endgroup$ – ACuriousMind Oct 11 '17 at 15:16
  • 1
    $\begingroup$ @ACuriousMind Tyvm - I didn't follow the changes of the help pages very carefully, only the relevant meta SE posts and the general dynamics of the communities and the applied rules. Although I estimated only 2-4 months for the automatic lift time, at the time even that wasn't known that there is a timeout. My other conjecture ;-) is that only a few (maybe only 2-3) upvotes to the edited, negative-scored posts are enough to lift the ban. $\endgroup$ – peterh says reinstate Monica Oct 11 '17 at 15:21
  • $\begingroup$ If it's a really bad question, in the sense that the homework rules have been blatantly ignored, or it's near, (or actually) trolling, I will d/v, other I will try a comment. But I think the real reason is that it's the annoyance of users and the opportunity to vent about bad questions in general. $\endgroup$ – user171879 Oct 11 '17 at 19:54
  • $\begingroup$ @Countto10 yeah no mercy for the real turkeys, we all agree on this. $\endgroup$ – ZeroTheHero Oct 11 '17 at 19:58
  • $\begingroup$ "where is the line between being dissatisfied with a specific question and downright discouraging new users?" - The first question I answered on a stack exchange site (the EE site, not the physics site) got massively downvoted and eventually deleted and, to be sure, the downvotes were deserved; I didn't do my homework on what is expected of an answer on the stack exchange network. The point I'm attempting to make is that the answer to "where is the line" is "according to whom?". $\endgroup$ – Alfred Centauri Oct 15 '17 at 22:58
17
$\begingroup$

There are some effects of highly negative score not related to reputation:

  1. A question below -4 will be pushed of the front page (i.e. what visitors see at `physics.stackexchange.com? ), cf. this meta answer.

  2. Users with more than 20k reputation can vote to delete questions of score -3 or lower immediately after they are closed without the usual 2 day period in which users can't vote to delete closed questions.

  3. The 10k tools (a set of statistics available to users with more than 10k reptuation) contain categories like "most downvoted question/answer" in a certain timeframe.

Let me remark that in general, you shouldn't ever hesitate to cast a vote because of the effect (reputation-wise or psychologically) it might have on the user. If the post deserves to be downvoted, downvote it. If it doesn't, don't downvote it. The author and its current score should not factor significantly in your decision to vote on a post.

$\endgroup$
  • $\begingroup$ Additional meta link on point 1: meta.stackexchange.com/q/62729. (You'd think these would be easier to find.) $\endgroup$ – Emilio Pisanty Oct 11 '17 at 15:11
  • 3
    $\begingroup$ About the last paragraph, some communities had meta discussions suggesting to avoid to excessively downvote new user's questions, unless really bad, to avoid discouraging participation (I generally agree with this nice suggestion). $\endgroup$ – Massimo Ortolano Oct 11 '17 at 15:43
  • $\begingroup$ @MassimoOrtolano I think discussing whether we should do that would be a suitable topic for a new meta post. $\endgroup$ – David Z Oct 11 '17 at 18:37
  • 1
    $\begingroup$ @MassimoOrtolano I'm not sure if I agree with that philosophy, mostly because I think that's a slippery slope. Changing voting habits because of other votes cast creates a more subjective voting technique. It is no longer about the content; but about how the community has already dealt with the content. If you think a question is bad, it shouldn't matter if 5 other people think it's bad. Then we create a system with feedback loops; are people voting up because there are too many downvotes; are people voting down because they don't think the Q/A is that good, etc. $\endgroup$ – JMac Oct 11 '17 at 18:39
  • 1
    $\begingroup$ @DavidZ I'll try to write a meta post about that over the weekend. $\endgroup$ – Massimo Ortolano Oct 11 '17 at 20:21
  • 7
    $\begingroup$ This is too absolute "If the post deserves to be downvoted, downvote it. If it doesn't, don't downvote it" and there should be an exception for young new users. I always check the age of the poster and most often "stupid" questions and answers are due to the below 18 yrs age of the poster. One should try to be gentle in explaining and try to educate the young ( which is one reason I often answer questions with close votes, hedging my bets by putting in a place holder answer which can be edited after closure comes. $\endgroup$ – anna v Oct 14 '17 at 6:03
  • $\begingroup$ @JMac Although I generally agree with you, one can wonder if it is constructive to cast the $n+1$ downvote (with $n$ suitably large) if this leads to no improvement of the site. Granted there are some turkeys that deserve no mercy, but those are not all posts that are massively downvotes. I think anna v’s comment indicates there is room for nuance (although for me age is not something I would consider.) $\endgroup$ – ZeroTheHero Oct 15 '17 at 16:50
  • 3
    $\begingroup$ @annav I strongly disagree. Not to get into an extended discussion about it here, but this site is meant to be a resource for readers more than for askers. (This ties into why we view asking good questions as contributing to the site, rather than gaining a benefit from the site.) If a young person asks a bad question, that's just as bad for us as if an older person does. If we don't give them the same response, what benefit do we get in exchange for taking the "hit" on a bad question asked by a young person? $\endgroup$ – David Z Oct 16 '17 at 9:21
  • 3
    $\begingroup$ @DavidZ Let us disagree, so be it. I have no investment in the site but to the physics and the people themselves asking or answering. The voting system takes care of the value of the question, which is fine. I just think that a 14 year old interested in physics should be shown where his/her misundertandings lie, so that they might appreciate physics and keep on studying. $\endgroup$ – anna v Oct 16 '17 at 10:40
  • $\begingroup$ @annav a 14yr old or anyone at any age... $\endgroup$ – ZeroTheHero Oct 17 '17 at 18:21
  • $\begingroup$ I previously pointed out that an already existing low score may actually be a deterrent for some voters, though I argue one shouldn't let that get in the way as well. $\endgroup$ – Kyle Kanos Oct 18 '17 at 9:45

You must log in to answer this question.

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