I saw this solution with -56 votes , and I thought it was a bit harsh. Once the answer is in negative territory, it's quite obvious the community doesn't support the answer... what is the driver for being able to continue to downvote an answer?

Continued points loss due to downvoting an answer:

  • discourages future contributions
  • encourages deleting the answer

... neither of which is beneficial to the community.

I imagine it's quite discouraging to continue to lose points due to downvotes. It'd make me think twice before future contributions. Deleting an answer should be discouraged - the answer may be correct, just not the best fit for mainstream users.

Are there alternative ways we could positively incentivise a better system? For example:

  1. To give opportunity to improve a question/answer, should downvoters be required to provide justification and/or provide constructive feedback? (to discourage "drive by" voters)

  2. To discourage deleting answers, should there be a lower limit on downvotes or points lost as a result of downvotes? (and leave deleting to moderators)

    1. some answers don't appeal to the majority, but may have some value to a minority
    2. to provide an example of "what not to do"

UPDATE: linked answer has been removed

  • 2
    If an answer is accepted, I don't think there needs to be a lower limit of down votes. A bad answer must always be kept away from later viewers. – iBug is disappointed in SE Apr 30 '17 at 0:55
  • 10
    If the author wanted to avoid additional downvotes they could have improved their answer and address the concerns of the downvoters. The reason it's getting downvotes is obviously due to the author's suggested approach to the solution. In other words, the author could, improve the answer. – Ramhound Apr 30 '17 at 1:06
  • Should upvoting (or points gained) also have an upper limit? – Cai Apr 30 '17 at 8:21
  • Did some quick and dirty numbers. Not counting rep-caps, and purely off scores, capping upvote-reputation on 50 only makes me lose about 7300 points of reputation on my account. I'm a little too lazy to do it further (and I just was wondering the same thing). I'd oddly be more happy about capping absolute "loss/gain" per question over hiding the spread in situations like thi. That's almost a different question though. – Journeyman Geek Apr 30 '17 at 8:53
  • @Jason-C / et al ... and how about downvotes on questions? (this one in particular). I'm not particularly incentivised to keep it at the current rate. Do downvotes mean people don't want this feature - or does it mean this question is of low quality and the author should improve it? (or is it sweet irony?) – Nick Grealy May 1 '17 at 7:45
  • Just thought I'd ask: why is this question getting downvotes? @Ramhound – Nick Grealy Nov 15 '18 at 21:37
  • 1
    @NickGrealy - You will have to ask somebody else who downvoted your question. My original comment stands, the answer you are asking about should have been improved, but it wasn't improved so it was deleted. – Ramhound Nov 15 '18 at 21:47
  • 1
    @NickGrealy Since you seem to be confused about the votes here: voting on Meta.SE is a bit... borked. For questions like yours, tagged 'feature-request', people will use downvotes to say they disagree with implementing such a feature, instead of 'yay, well-written feature request'. See also the 17 upvotes on the answer saying that no, such a cap is not needed, which is basically the explanation of those downvotes in this case. Hope this helps you understand the votes a bit better :) – Tinkeringbell Jul 1 '19 at 9:53
  • @Tinkeringbell - ahh.. that makes much more sense, thanks for taking the time to explain. I'll strip the feature request (/solution) angle... my intention was purely to raise the question. – Nick Grealy Jul 1 '19 at 14:09

Note that that post is +43/-79 so, rep cap and ordering aside, that's actually a theoretical gain of (43 * 10) - (79 * 2) = +272 points. It takes 5 downvotes to offset 1 upvote for rep, which essentially solves the net "point loss" premise here.

In fact, there really are not a lot of answers with a significant net point loss. On SO, for example, there are only 1,970 non-deleted answers where the user lost more than 10 points. There's only 231 where the user lost more than 20. And that's out of 21,646,847 answers. This isn't really a major cause for concern. Here's a chart to put that in perspective, if you like charts, of # of posts that lost more than the indicated threshold of points (or log scale), you can see the falloff is pretty quick.

As for putting a lower bound on the displayed score, I don't think you want to do that, there is value in that information.

Also note, previous points aside, you couldn't put a hard limit, it'd have to be a function of number of upvotes or something; posts with heavy voting activity would warrant a higher threshold to keep the score meaningful.

So, no, I don't think so. I can't really think of a straightforward way to make this happen and, more importantly, I don't really see any concrete evidence that this overly discourages users to the point that the site suffers as a whole (and to be honest I think it is important to encourage users to post good answers or improve their existing ones, and discourage poor answers, but that's just me). Users with delete privileges can vote to delete answers, which is an option, but of course, there's also value in knowing what answers not to follow, so you could argue for keeping them, too. That'd be case by case and up to the personal opinions of folks with delete privileges, a bit outside the scope of this post.

It'd make me think twice before future contributions.

This sounds like a good thing, doesn't it? You should think twice before posting a contribution, I think.

  • It makes perfect sense now that I think about it but I was seriously surprised at those numbers – Cai Apr 30 '17 at 8:19

You must log in to answer this question.

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