I agree it’s a form of moderation and has long term value, there are issues with how voting is applied (quickly and without feedback), leading to questions being buried or closed instead of improved.
As an example: I read a poor question where I can decode what the problem they are having is, but it’s not obvious from the question, so it should be improved.
Since I know a good solution to this problem, I don’t downvote or close-vote because I don’t want to bury or close the question while I answer. Instead I suggest how to improve the question in a comment, and start writing my answer.
When my answer is finished after 10 minutes, the question is -4 and has multiple close votes even though the author has already improved the question, and my answer was accepted. The question isn’t even an hour old, it’s now a perfectly acceptable question with an accepted answer - but it’s now closed and has score -4 with votes applied to a question that looked nothing like it does now.
So the biggest issues are
- too easy to downvote with no moderation effort
- too easy to end up with improved questions still closed
- Fundamental improvements doesn’t reset score (why should asker be encouraged to write a new Q instead of changing old?)
In short the problem is that everything happens at once, and the system is designed around questions being final when in reality the first 24h sees significant change.
So I’d suggest some simple fixes:
- close votes and downvotes removed (by moderation queue confirmation) if a question is sufficiently rewritten in the first (say) hour. It should simply count as a new one.
- downvotes registered but not shown during the first (say) hour. Herd mentality seems to make the threshold for downvoting lower.
- clicking downvote should at least have an optional comment or a selection among issues (like the close votes already have). That the material has an issue is content moderation but it isn’t feedback. I agree with Jon’s goal that the site should aim to make better askers - not just a better Q/A site. It should even be an explicit goal.
The way to get more feedback is to reduce the load of the individual in community moderation.
Try going to a tag... lets say Java. https://stackoverflow.com/questions/tagged/java Now, on each post on that page read it, edit it if it needs editing (this includes things like "i am trying to use a path..." - fix all the grammar (the post I'm looking at is tagged java, c#, android, .net and google maps - https://stackoverflow.com/q/49349444 ). Give a comment if additional clarification is needed.
Then go back to the tag page and see how many more there are. Repeat.
To help the most people possible, and help the people who are answering questions know which are the posts that they can help the most - vote on the questions. Up and down.
If you know it is a good question that you can decode - fix it first. People casting close votes don't see that you're trying to answer - they see an unclear post. Whats more, an unclear question (that might be clear to you) isn't clear to everyone else. You've spent a minute or so decoding it - contribute that time back to community by fixing the post. Don't leave it in a comment - fix the question before writing an answer.
Consider that if a post is getting close votes there and down votes even with the fix, it might not be a good question for Stack Overflow. Not all good questions are good questions for a particular site. One doesn't ask coding questions on Hacker News even though it may be a good question.
That a post is too easy to down vote - by percentage, down votes are a small, small fraction of the total votes cast. Up votes are 8x more common than down votes.
Improved questions many still not be a good fit for Stack Overflow. A recommendation question or opinion poll, or something that is "how do you design Amazon" is still a poor fit, even if it has been improved.
Close votes expire as it is. If its good enough, it will not get any more and there's no issue.
Down votes are a signal to people answering questions that this isn't a good question. Look at a page with 50 questions on it. Decide which one you're going to spend time on - that is likely based partly on which questions have gotten up votes.
Rewritten questions don't necessarily mean that the question is good now.
In the past, when I've down voted and left a comment I've gotten a dozen angry pings from comments from the person and sometimes half a dozen down votes on my other posts (that then got removed for serial voting). I decided not to leave comments any more at that point. I'm not there to argue with a person about if their question is good or not.
The person asking the question has all the time in the world to perfect it before posting. They can see the preview and how it will look. They can run it through a grammar checker. They can notice even things like the red squiggles under misspelled words on any modern browser.
There is no reason why they can't post a good question the first time - or at least one that shows that they have put in at least as much time as they're expecting of the person answering.
But I can’t choose among 100 questions to answer. It’s on the rare occasion that the question I can answer shows up that a do it. That’s the situation for a lot of us. We answer a few niche questions.
And when that question happens to be poorly written - I can’t pick another. I can ignore it or I can answer it.
Some times I’m almost inclined to commenting “give me your email and I’ll sort you out because your Q will be closed even though I see your problem and had the exact same problem”.
I think a great feature would be “protection”. If I adopt a question I can take responsibility for both my answer and fixing the question. If I’m say 5k rep and the asker is low rep I should be trusted to be able to be a “mentor”. The question can even be hidden until I and the asker have worked on both question and answer together.
So the context is
- asker: can’t write a good Q without help
- answerer: wants to help make better asker perhaps MORE than I want to make a better Q/A site, can’t use time to answer another Q instead
"Protecting" a question in the way you described would likely result in people claiming a question and then preventing others from answering it in a timely manner when it doesn't need any help (there's a lot driven by gamification). Just picture the fastest gun in the west problem and the "now I can click and prevent everyone else from answering?"
If its poorly written - you can do something. You can edit it. You can close vote it. You can down vote it. Those are signals to the author that it needs to be worked on.
Look at https://meta.stackexchange.com/q/216683 - and that the best way to drive away a user is to do noting. To ignore them. Closing their questions, down voting their questions, commenting, editing... and yes, up voting and answering - those all help.
Editing a question is much better for new users than voting on their question.
> "Protecting" a question in the way you described would likely result in people claiming a question and then preventing others from answering it in a timely manner
I was being unclear: I didn't mean protecting it from other people to answer - mainly from downvotes and close votes.
That is: I want to say "I think I know what this asker is trying to ask. I'm take responsibility for editing the question to be up to standard and/or mentoring the asker. Regardless - it WILL be up to standard within half an hour so just HOLD OFF with the ninja downvoting".
Answer it if you want. Kill it after I edit it if you want. But don't close it as it stands.
While that may be making that user feel better, it will frustrate every one who is looking for a good question to answer and sees something without any down votes... and sees it to be a poor question.
Close votes only impact the ability to answer a question. If the question is improved while it is closed, it goes back in the reopen review queue and people can then take time to reopen.
A question that is closed before it gets the "try {some code}" answers that are wrong comes out ahead if the person asking the question is able to improve it... at which point it can get reopened through the existing processes.
Poor questions go in the triage queue first. Questions that are deemed that can be improved then go to the help and improvement queue. Those questions are hidden from the general site for some time.
That functionality exists already in those queues. You just have to use it. When those queues fill up, poor questions can indeed spill over into the main site. Go to https://stackoverflow.com/review/triage/stats and review (remember that "needs editing" means that you are capable of editing it into a good question) and https://stackoverflow.com/review/helper/stats
> That functionality exists already in those queues. You just have to use it.
Thanks, there definitely seems to be more to it than I was aware of - and I still thought I was a pretty seasoned user (8k). I always found the entire review system rather opaque and hard to understand even though I processed a lot...
I obviously need to dig deeper, but at the same I think a lot could be done to make the processes more understandable. The queues and question state-machine really could be illustrated more clearly for example (and I’m sure there are endless meta-posts about that too).
> Behind the scenes, a "quality score" is calculated for each question based on an automated analysis of the content. Those that score well are sent immediately to the homepage; those that score poorly will now be sent to Triage. From there, they'll go to one of three places based on human input:
> 1. The homepage, where they can be answered
> 2. The close or moderator flag queue where they can be reviewed and eventually deleted
> 3. A new "Help and Improvement" queue where they can be edited
As an example: I read a poor question where I can decode what the problem they are having is, but it’s not obvious from the question, so it should be improved.
Since I know a good solution to this problem, I don’t downvote or close-vote because I don’t want to bury or close the question while I answer. Instead I suggest how to improve the question in a comment, and start writing my answer.
When my answer is finished after 10 minutes, the question is -4 and has multiple close votes even though the author has already improved the question, and my answer was accepted. The question isn’t even an hour old, it’s now a perfectly acceptable question with an accepted answer - but it’s now closed and has score -4 with votes applied to a question that looked nothing like it does now.
So the biggest issues are
- too easy to downvote with no moderation effort
- too easy to end up with improved questions still closed
- Fundamental improvements doesn’t reset score (why should asker be encouraged to write a new Q instead of changing old?)
In short the problem is that everything happens at once, and the system is designed around questions being final when in reality the first 24h sees significant change.
So I’d suggest some simple fixes:
- close votes and downvotes removed (by moderation queue confirmation) if a question is sufficiently rewritten in the first (say) hour. It should simply count as a new one.
- downvotes registered but not shown during the first (say) hour. Herd mentality seems to make the threshold for downvoting lower.
- clicking downvote should at least have an optional comment or a selection among issues (like the close votes already have). That the material has an issue is content moderation but it isn’t feedback. I agree with Jon’s goal that the site should aim to make better askers - not just a better Q/A site. It should even be an explicit goal.