Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 153 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

It's great that we now have measures that will limit questions from users with a history of bad questions. Can we do the same sort of thing for users with a history of bad answers? I.e., if a user has lots of very low score, unaccepted answers, could we lock them out of answering until they clean up their mess?

This is partly inspired by this question, which asks about spammy users. The kind of solution I am suggesting would also help limit the impact of users who are only interested in promoting their own product.

Minor clarification: I'm not griping about zero score answers (I have tons of them myself!). One way to do this would be to sum up the net votes of all your answers. If the total is less than a certain threshold (say, -5), then you're a bad answerer.

share|improve this question
How can they "clean up their mess"? – NullUserException อ_อ Sep 30 '10 at 15:16
@Null - by deleting crap answers, or editing them to make them better. – user27414 Sep 30 '10 at 15:18
Or by doing nothing. They could just go away if they are only interested in spamming. – user27414 Sep 30 '10 at 15:19

I like this idea, because I strongly feel that answers aren't self-deleted often enough.

The rule wouldn't even have to be onerous: set a threshold for total answers, and if 50% or more are not accepted and scored <= 0 then the user has to remedy that before posting more. Since answers can almost always be self-deleted (unlike questions), this offers a clear and easy path to grace for users looking to redeem themselves.

Obviously if they wish to edit their answers and wait for up-votes... or just leave... that's fine as well.

Ok, here are some stats:

Users with > 30 answers with a threshold score of 0

User                          % Below Threshold Total Answers 
----------------------------- ----------------- ------------- 
klox                          100               34            
Bharanikumar                  100               35            
jeeva                         97.222222222222   36            
Rajesh Jadhav                 94.594594594594   37            
Deepak                        94.594594594594   37            
sanders                       94.444444444444   36            
Phil Jackson                  93.75             32            
Patrick                       91.304347826086   46            
marc-andre menard             91.228070175438   57            
4thSpace                      91.176470588235   34            
combi001                      90                40            

First thing that jumps out: these are users using answers when they should be using comments, and / or doing a lot of self-answering. I do think a cut-off or at least a reminder would be useful here, but if someone wants to slog through these and start flagging they could probably keep the moderators busy for a while... It's noise, but more than malicious it just seems kinda sad.

Users with > 30 answers with a threshold score of -1

User                          % Below Threshold Total Answers 
----------------------------- ----------------- ------------- 
Grumpy                        29.787234042553   47            
Sir Psycho                    25.396825396825   63            
Janie                         21.739130434782   46 

Users with > 10 answers with a threshold score of -1

User                     % Below Threshold Total Answers 
------------------------ ----------------- ------------- 
Arunabha Dutta Choudhury 66.666666666666   12            
Joan Venge               50                14            
unknown                  46.666666666666   15            
XcoderMi2                41.666666666666   12            

-1 seems to be the more popular threshold, but frankly I don't think it's worth bothering. There are users with a lot of answers, and users with mostly negative-scoring answers, but the number of users with a significant amount of mostly-negative answers is tiny.

share|improve this answer
I'd like to see some stats and maybe a couple of data dump queries. – John Saunders Sep 30 '10 at 18:45
@John: good thinking. After badgering SEDE for a while, I'm less enamored with the idea; it highlights some users who probably deserve a bit of extra scrutiny and/or help, but doesn't seem terribly useful as a way to drive out spam. Which I guess means we're either doing a pretty good job, or a really lousy job of identifying spam users... – Shog9 Sep 30 '10 at 23:18

The problem is I often answer a questions by a new user that does not get many views, a lot of new users don’t up vote any answers.

I don’t wish to have to look at some’s voting history before deciding if it is worth the risk of answering their question.

If there is a limit on bad answers then it should only take into account answers with negative votes (and not on meta!)

share|improve this answer
I agree with the bit about negative votes. I think a zero score answer is just fine. – user27414 Sep 30 '10 at 16:46

Bad idea. There are enough 0 votes answers out there, which are good, but no-one has read them or understood them. Hell, there are downvoted ones, which are correct. Some people state their opinion of the right solution, even if their peers disagree. (Hint: Their peers could be wrong.)

We have downvotes for bad answers. Use them! Propose solutions against pity upvoters and smash Jeff's head till he has implemented them. (Hint: Increasing the downvote weight.)

Let people deal with bad answers, bad answerers and their history. Not obscure algorithms.

share|improve this answer

You must log in to answer this question.