Discouraging frivolous close voting by adding a small cost like downvotes on answers

Procedural note

We are here to discuss how a new system should work. Your views and productive discussion are all welcome. Making it personal is not. Do not level accusations. Do not make assumptions about people’s motivations. I really don’t want to have to start moderating here too because we can’t have mutual respect.

6 Likes

Obviously some standards and consistency are good, but of course there will always be a range of interpretation of what the rules mean and how they should be applied in any one instance. There will always be cases in the gray area where people disagree. That kind of diversity of thought is actually a Good Thing. Even if you don’t believe that, punishing people for not having correctly guessed the eventual herd-think result is very wrong.

What I mean is that just because you are so sure that a question got closed wrongly or due to “unintended consequences”, someone else is equally sure that it was obvious the question should have been closed long ago.

We need to respect diversity of opinion, not stifle or punish it.

This is the real crux of the problem. The far more obvious and likely reason is that others simply disagree with your point of view.

A bunch of people are going to have disagreements. That’s fine. Voting will apply the prevailing attitude. However, we must always be respectful of others’ votes, whether we agree with them or not. Punishing people for not agreeing with the eventual outcome is very disrespectful.

Wow, I really have to stifle myself here. The idea that anyone else gets to declare your vote “wrong” is so repugnant to me that I don’t really know how to respond without it sounding like a rant. I guess that if we disagree at this fundamental level, I’m not sure what arguments I can make that you would see as valid.

I do agree that some votes are malicious or vandalous, but of course this is impossible to prove since it requires knowing the intent, whereas we can only observe the outcome. That same outcome could have resulted from other perfectly legitimate reasons. Since we can not prove bad intent, we can’t have the Thought Police punish you just because your vote ended up not agreeing with the herd-think. Rewarding you for agreeing with the eventual herd-think is just as wrong.

I think the best way to make dishonest votes (where you vote other than the way you truly believe is right) less likely is to make all votes public. We still can’t judge any one vote as malicious, but people will think more about their vote if they have to publicly stand behind it. Put another way, if you’re not willing to stand up publicly and be counted, then the weight of your opinion should be less.

Note that none of this ever declares a vote as wrong. It’s still your vote and nobody can make you change it. But your voting pattern will be one thing people judge you buy.

3 Likes

This might be overly complicated, but I have no real objection, in principle, to this scheme. Remember that edited questions could have changed considerably, so more reset of the requirements after an edit might make sense.

Now you’re going too far. Reasonable people can reasonably disagree. Your voting scheme sorts out the result. OK so far. But then punishing or rewarding people for having their opinion agree with the eventual outcome is just plain wrong.

Let’s discuss close and reopen vote mechanics. But, you don’t ever get to judge someone else’s vote as “frivolous” or wrong. And therefore, trying to punish or reward voters on that basis is equally wrong.

3 Likes

Regarding making votes public. I’d agree with close votes being made public when they were not effective. I am not sure about publishing them whilst pending, as I have the concern that it might lead to the thought that some high-experience user voted to close that post and therefore it must be correct, leading to a questionable closure.

For other votes (up and down), I still think they must not be published, however that is not within the scope of this topic.

Otherwise I agree, that close votes that were not effective shouldn’t be penalized, unless someone continously abuses them in bad faith. To that would belong, IMO, intentionally voting to close questions where the community decided to allow that type. However this should be dealt with manually by moderators being able to revoke the privileges for some time.

2 Likes

Either you make all votes (of a particular type, like close/reopen) public or not. Singling out those that didn’t end up agreeing with the collective result doesn’t make any sense. That feels like trying to call them out for being “wrong”.

Yes, people might be influenced by a particular known users voting a particular way. That’s simply information that voter considers in making their decision. It’s up to them to weight that by however much they think it matters. That can be good. There is nothing wrong with the thought process “Hmm. I thought this was bad, but xxx thinks it’s good, and I have come to respect contributions of xxx. Let me go back and look at that post again.”. It’s still ultimately that user’s call how to vote. More information isn’t bad.

3 Likes

Sorry if my post was misunderstandable. I thought that effective (successful) close votes would be shown on the post and in a “revision history”. My proposal was intended for these cases, where this would not happen (close vote rejected by “consensus”/expired).

IIRC moderators on Stack Overflow can see who flagged a question and who voted to close it, from some special timeline. This might be an option for this use case too, however it deprives us from the community flagging these cases.

I would strongly caution against putting too much code into the process of penalizing what some perceive as frivolous close voting.

First, consider how much of the “out there” misconceptions and misunderstandings of how the (relatively) simple process of Stack Exchange works. Look at how few people read and understand the existing documentation on Stack Exchange. Making a system that is more complicated than this leads to the very real possibility of people misunderstanding this system and complaining about it. Understanding a close question process is important.

Next, the more that the system has baked into code rather than community means that it is that much harder to change things. If it is later realized that the close process needs to be changed in some way - the more (for lack of a better word) dials that are locked into the way things work, the harder it is for either the community to change around it or for the code to change.

Closing questions is likely a place where the communities that form around different subject areas will need flexibility. Having the code dictate that it works such a way that is complex and not something that can be changed except through code makes it that much harder for the communities.


I would also take a moment to take exception to the word “frivolous”. Flippant, glib, foolish, joking… Close votes are never these things. In my time on Stack Exchange, I’ve never seen anyone go through and cast close votes on everything just because they could. There have been times were people didn’t understand the reasoning behind the why (“the curation of this question and its answers is taking more time than the community is willing to spend on it” isn’t frivolous).

It is important that the people who are invested in the site and work to maintain a particular vision of what it should be have the tools to do so and are not penalized for using those tools in pursuit of that vision of the site. This implies a certain amount of investing in that vision. Stack Exchange has used reputation as a proxy for counting that investment - that has good parts and bad parts to it. Maybe the way of counting investment in the site needs to be re-examined… but that’s an entire rabbit hole of its own to go down.

However, please do not label certain moderation actions frivolous until some significant investiture of time is made into the success of the site and the community around it.

6 Likes

I think people are comparing close voting to up or downvotes, when a far better comparison would be to flagging.

When you flag a post or comment, you are asking a moderator to take the time to look at it. If enough of your flags get declined because the moderator disagrees with the validity of your flags, then you get a flag ban. (Raised more than 10 flags in a week and over 25% of those were declined)

On Stackoverflow, if you are reviewing and fail an audit, you get a review ban.

When you vote to close, you are asking at least 3 other users to take a look at a post and up to 9 if it gets closed, edited and reopened.

There have been times on Outdoors.SE where I counted, and the number of posts that entered the queue and got closed was something like 10%.

People are free to upvote/downvote however they want, but if people are going to ask other users to look at something, they should at least put some thought into it.

6 Likes

Hmm, this gives me an idea for a modification: only apply any rewards if the outcome is unanimous, or penalties if the reviewer was the only dissenter, so that meaningful gray areas are more easily accommodated.

Ah, there we go.

I don’t care about intent. (Well, I do, but only a little.)

What I was concerned about is the phenomenon of poorly-informed/irrational close votes cast for reasons that just do not apply to the question. That is, it’s not that someone deliberately tried to do a lousy job of voting. (Who would do that?) They just didn’t have good reasons for voting how they did. In extreme cases, you see things like a “proofreading” custom reason used to (try to) close things that aren’t asking for anything like proofreading at all. The problem is not that somebody has a different opinion. The problem is that the community agreed on a sensible guideline to require of askers, and someone is applying it wrong.

I can’t always tell why this is done, of course; it may be a question of judgment in a gray area in the policy, in which case no harm done (although it would be nice to sort out the best approach); it may be a straightforward disagreement with established policy (but as mentioned, such a protest vote is very dubious); it may be ignorance of the policy, or carelessness when considering how to vote (but then these should be corrected). And I’m not just making up random possibilities here: these are all errors I myself have either made repeatedly, or been at serious risk of making.

Yeah, that is difficult. I don’t see close voting as an expression of free speech (?), but as an accurate, knowledgeable judgment based on guidelines and experience. The fact that it’s a vote is an implementation detail.

3 Likes

You have neither the information nor the right to judge this.

1 Like

That is likely more due to the low participation in the review queues than wide disagreement that the question should not be closed.

9 posts were split to a new topic: What should be done about answers posted as comments?

PSA: Please don’t discuss here, how comments containing answers should be handled. Use that thread for this discussion:

1 Like

Questions ageing out of the close vote queue on Outdoors.SE, rarely if ever happened. People were really pretty good about clearing the queues out quickly. Of the ones I counted, all of them got the full review treatment, and didn’t fail to get closed simply because of a lack of prompt reviewers.

That is a problem on some of the bigger sites, but it seems that having more close votes than the reviewers can go through is an argument for making the close votes more accurate, not less.

1 Like

I have seen several times discussion in meta that show the opposite.

Q: “why was my questions closed?”
A: “I was among the close voters, it was a mistake, sorry”

Frivolous close votes exist.

Such an example for the case of a question, which received votes to close in this way, is here

At some times people have even admitted that they simply voted just because they already saw other votes (that was in the case of downvotes on an answer).

6 Likes

That’s clearly not a frivolous vote. According to the voter himself:

Believing the question was posed poorly and the title not reflecting the question are hardly frivolous reasons. This user then even left a comment about this.

He did say later that in hindsight he might have been too quick to close, but that only illustrates this is a judgement call, and for that user, it was right on the edge.

There was nothing wrong here, other than you writing a question that was bad enough that it got at least into the gray area of several users. Don’t blame the messenger. If you had taken a little more care in writing the question, properly proof-read it, and picked a better title, the question would probably not have been closed in the first place.

2 Likes

Questions should ideally be well written. But believing that a question was posed poorly is not enough and is a very gray area. One should know it. (if a person is not understanding a question, then this may have two reasons, only one of them is that the question is unclear, the other reason occasionally occurs as well)

The crux with that linked question was that voters may vote something for being off-topic while it is not. The problem may not always be in the question but it can also be in the person(s) voting for the question.

It is not like voters never make mistakes. It becomes a semantic discussion whether a mistake like in the link, where somebody was convinced about the vote, is to be called frivolous or not.

The particular question that I linked to is and was very clear (it even explicitly stated the issue that they have “why is P(Y=k|X=15,B) = P(Y=k|B)?” ), or at least clear enough, and there have been reasonable answers (one of them is mine) that show it was not too difficult to provide an answer.


If these type of questions (a person studying a book and encountering a remarkable feature and wondering how that works so he places it on a Q&A site) are not possible then we can just as well close the website.

Note that in this particular question the person asking the question may have been unclear or vague but that is because the person was confused about the topic. The point of the question is to take away that confusion. This follows the saying “there is no such thing as a stupid question” (on SE/SO many people believe this is not so).

I could answer the question because I understand the purpose of the exercise that the OP made an inquiry about (some introduction to beta-binomial distribution by having a mixture of two binomial distributions), and at the same time understand why the OP could be confused about it (the exercise is a contrived example, and may easily confuse a person that is overthinking the reasons for it).

The most important thing that the question needed fixing was a better title and improvement of the image (conversion to text). But that is far from a good reason to down vote or close vote. Such reception is making people feel very unwelcome (and it is unbelievable that SO/SE company is fighting their status as being unwelcome, while having these unwelcoming tools in place).

I’d say we could introduce an option to mark questions as ‘requiring updates’ or ‘need improvement’ and make ‘close voting’ something that is reserved for questions that are really problematic and should not be answered (because of the risk that answers will not make sense).

1 Like

is one of the dumbest myths propagated on the internet. Of course there are stupid questions.

This is merely semantics. That’s not what the word believe means in this context. He believed it to be posed poorly. Which is another way of saying in his judgement, it was posed poorly. That is a perfectly valid reason for voting to close. Of course others may disagree with that judgement, but it is his judgement to make when casting his vote. Others arrive at different judgements and vote differently. That’s the system working correctly.

It’s a perfectly valid and legitimate reason. The question needed to be improved. It was put on hold while that happened. It was then re-opened. That’s exactly how the system is supposed to work.

You are taking this phrase much too literal and black and white. The point of that saying is that behind every question is a cry for help. Obviously we can differentiate between how much leeway we want to give in providing help to those who ask for it. However, I believe that one should always handle this with care and remain polite. This contrasts with some of the behaviour and methods on SE/SO (which are even promoted by the system) which can be at times very obnoxious and toxic.

The issue here is not that people are placing close votes while they don’t believe a question to be posed poorly. The matter is whether people are making their judgements too easily (and that is also what the person stated “I may have reacted too quickly”).

There is no ‘vote differently’. It is only the votes of the close-voters that count. The system made an arbitrary boundary that five people can close the question. That is a weak democracy (a sort of opposite from a concensus system).

That is not automatically a system working correctly. Especially, when it leads to questions being unnecessarily closed or downvoted (the one that I linked to is a clear example) which leads to an unsympathetic environment.

1 Like

Hmm, are you talking about downvotes or closevotes? I don’t see an explanation of frivolous close votes in there, and I don’t have any way to find the relevant CV review URL to see what people voted for.

Ironically, my stance toward downvotes (and upvotes) much more closely resembles Olin’s toward closevotes. I don’t mind giving general guidelines, but there’s no sense trying to second-guess any specific cases save the most brain-dead of misclicks or the most blatant of voting rings (e.g. upvoting pharma spam). In the linked example, “question was posed poorly” and “title not reflecting the question” are both perfectly plausible, if somewhat modest, reasons to downvote. Neither is adequate reason to vote to close without a good bit more specificity in exactly what was wrong.