-14
$\begingroup$

I recently received the following public comments on how to use down-votes from a senior diamond moderator, Dr. Xander Henderson, who patiently explained me the policy of the site on how to use down-votes (a small part in the second comment that was related to the old post was removed and a typo was fixed):

There is no requirement on this site that downvotes be given for "scientific" reasons. Anyone can downvote (or upvote) any post for any reason. The only restriction is that votes cannot give the appearance of targeting a specific user—votes need to be about content, not people.

There is no "correct" or "incorrect" way to vote. Anyone can vote for anything for any reason, so long as it doesn't create a pattern of voting on the basis of author, rather than content. This isn't a matter of what is "correct" or "incorrect"---rather it is about what is permissible on this site, and what is not.

On the other hand, one can find the following statement in the official page of MSE on down votes:

Downvote answers that are incorrect or don't provide sufficient information $\color{blue}{\text{to be}}$ $\color{blue}{\text{to be useful in}}$ $\color{blue}{\text{answering}}$ $\color{blue}{\text{the question}}$. Some answers may not attempt to answer the question at all, and should be flagged.

The above MSE's official statement clearly shows that the interpretation of term useful in the following Xander's comment cannot be considered as the MSE's official page adds $\color{blue}{\text{in answering}}$ $\color{blue}{\text{the question}}$ after $\color{blue}{\text{to be useful }}$ to clarify it (moreover, recall that low-quality is a relative concept here and different members may have different opinions, it is why a democratic mechanism considered for closing and opening questions).

It may also be worth noting that the hover text on the downvote arrow is "This answer is not useful." The meaning of that phrase is very much subject to the interpretation of the voter. For example, a lot of users believe that answers to low-quality questions are not useful.

I personally think (see below for my reasons)

MSE policy/rule: Answers should not be down-voted based on any type of non-scientific reasons

where non-scientific reasons include targeting a member, putting social pressure on answerers who answered a question that some members think is not good, making an answer more visible compared to the other parallel answers, targeting a question, etc.

My question is: which of the above rules/policies or new possible ones should be considered in MSE?

I need to clarify that I am seeking an upgraded rule/policy that everyone in MSE is required to follow (not some guidance or suggestions), and MSE adopts the policy.

Actually, I suggest changing the current standard of the site, which, as of right now, seems to be based on this old answer (according to a Xander's comment clarifying this below). However, after comparing this old answer with other answers, all given 10 years ago, you may see that they follow different ideas, and provide some suggestions, while here I am seeking a new version of clear restrictions officiary considered by MSE as a rule/policy (not only some suggestions). According to Xander's above comments, the only restriction that is currently considered by MSE is targeting a member. I want this to be upgraded to any other type of voting based on non-scientific reasons.

Reasons supporting the suggested rule/policy

Before joining MSE, when I was using MSE as a only reader, and during these 7 months that I has been an active member, I have been always thinking that the only restriction of using down-votes for answers is voting based on non-scientific reasons (voting for questions is not discussed here questions have a different situation). My reason has been that an answer is an answer, nothing more, and can be evaluated only from a scientific perspective (note that credit-related issues are also scientific-related ones, but it may be better that members reflect them by using flags before judging). A controversial instance of down-voting based on non-scientific reasons is when some members use down votes to signal other members to stop answering those questions that they want to close for their own reasons, see here for a discussion (recall that the procedure of closing and opening questions is a democratic procedure in MSE). Another example is to use down votes to make a specific answer that the voter likes more visible when there are a number of answers.

When a reader who is not aware of the above policy of the site comes across a correct answer (with negative or low rating due to the down votes given to it based on non-scientific reasons), after seeing the negative votes or low ratting, he or she thinks that the answer is scientifically wrong, while the story is different. This can be scientifically misleading to 1) only readers, 2) temporary members, and 3) members who are not active in meta. It also conveys the message that the answerer did not care about his or her duty to write the correct answer, which damages the answerer's academic reputation (while the answerer is a volunteer in MSE to help and has the right to decide which questions are good or bad, considering the democratic procedure of MSE for closing and opening questions).

On answers

I kindly request answerers only focus on the policy/rule they think should be followed here by all members. Please clearly state your policy/rule, or the ones you want to compare (and then clarify why you think the policy needs to be followed by all members).

Already, we have following two options (the current one according to Xander's comment and the one suggested above):

Option A: MSE policy/rule: The only restriction is that votes cannot give the appearance of targeting a specific user—votes need to be about content, not people.

Option B: MSE policy/rule: Answers should not be down-voted based on any type of non-scientific reasons.

On policy enforcement and implementation

Policies set by administrators for a human system are not always fully traceable, especially those related to the values of system. Value-related policies are placed in order to properly orient the members of that system. The system states them in order to preserve its own values.

All policies of a site like MSE with highly academic members should not be enforced by moderators as the members follow them even if they don't agree with them, and there is no need to pursue their behavior. Currently, due to the site's lack of explicit disclosure of its policy, many members, as reflected in moderator comments, are now openly using down-votes to evaluate answers based on non-scientific purposes and are not afraid to say so, which is very painful for a scientific site.

To use the output of discussion, which is a policy on how to use down-votes for answers, there is no need to change the business model or the software (I don't know why some members want to induce it on me). After agreeing on a policy on how to use downvotes, the new policy is only required to be clearly stated in the MSE's site for the members (which is clearly within the MSE's authority). For example, if Option B is considered as the best policy, the current statement in down votes needs to be updated as follows to clarify everything for MSE users:

MSE guideline on how to use downvotes for answers: Downvote answers that are incorrect or don't provide sufficient information to be useful in answering the question. Do not downvote answers based on any type of non-scientific reasons. Some answers may not attempt to answer the question at all, and should be flagged.

$\endgroup$
20
  • $\begingroup$ Your question seems to be answered below the question you cite. Look at "the ugly" downvotes. math.meta.stackexchange.com/a/15132 $\endgroup$
    – Xander Henderson Mod
    May 14 at 15:12
  • $\begingroup$ Thanks for providing the existing question. The scope of my question differs from the old question: Is it appropriate to downvote answers for the sake of deleting a question?, which clearly focuses on a specific action of deleting a post about 10 years ago. The question above wants to clarify any possible restrictions such as targeting a member (not ethical appropriateness) on the use of down-votes for answers. $\endgroup$
    – Amir
    May 14 at 15:43
  • 1
    $\begingroup$ Hope you let the question be opened to get answers from the current active members on this topic that may result in the update of the only current restriction of targeting a member to any non-scientific reasons. $\endgroup$
    – Amir
    May 14 at 15:43
  • 1
    $\begingroup$ Read the answer to which I linked. While the question may not be an exact duplicate, the answer addresses your question. That answer is, as of right now, what I understand the consensus standard on this site to be. If you want to change that standard, you need specifically address that answer. $\endgroup$
    – Xander Henderson Mod
    May 14 at 15:47
  • 7
    $\begingroup$ It may also be worth noting that the hover text on the downvote arrow is "This answer is not useful." The meaning of that phrase is very much subject to the interpretation of the voter. For example, a lot of users believe that answers to low-quality questions are not useful. $\endgroup$
    – Xander Henderson Mod
    May 14 at 17:31
  • 1
    $\begingroup$ I tried to improve the question following your comments. Also, I invite you to kindly help to personaly improve those parts in the question that need more clarifications. $\endgroup$
    – Amir
    May 14 at 19:08
  • 4
    $\begingroup$ It isn't a matter of the question needing improvement. It is a matter of the question already having been answered, and your dissatisfaction with that answer. $\endgroup$
    – Xander Henderson Mod
    May 14 at 19:36
  • 2
    $\begingroup$ I find this handling of a policy question as duplicate not very productive. The linked question was asked almost ten years ago, and naturally, the user base has significantly changed since then (half of the accounts who posted in this thread have been deleted). Moreover, it's not like a community consensus had been reached on this question - the two answers with the highest vote totals (+35 and +40) have opposing standpoints on "ugly downvotes". The community opinion on this proposal by the OP seems pretty clear (current vote total of the question: -5), but I see no reason not to discuss it. $\endgroup$
    – MaoWao
    May 15 at 7:40
  • 1
    $\begingroup$ If you want your question to be re-opened, please edit your post to state more clearly what question you want answered. As far as I can tell, the question is "which of the above criteria or new possible ones should be considered in MSE?" This is not clear. Considered by who? Are you seeking a rule/policy that everyone is required to follow? Are you asking for guidance or suggestions for yourself? What do you mean by "should be considered"? $\endgroup$
    – D.W.
    May 15 at 8:14
  • $\begingroup$ @D.W. Thanks for your feedback! I tried to improve my questions following your comments. Please let me know if more clarifications are required. $\endgroup$
    – Amir
    May 15 at 8:40
  • 4
    $\begingroup$ How do you plan to enforce this? $\endgroup$
    – Xander Henderson Mod
    May 15 at 11:20
  • 6
    $\begingroup$ I don't think that you understand how this site works at all. Votes are anonymous. Moderators cannot see how anyone voted. If you want to change that, you are going to have to go to Meta Stack Exchange and see if you can get the company to change anything. $\endgroup$
    – Xander Henderson Mod
    May 15 at 13:26
  • 5
    $\begingroup$ @MaoWao This question has gone through a number of revisions. Earlier revisions were answered by the linked question. I could reopen this version, I suppose, but I would immediately reclose it as being "not at all possible". The reason that there is no strict policy on voting is that voting is anonymous. It is impossible to control how anonymous voters vote. Even at the level of targeted voting, moderators can only see if a large number of votes from one user are sent towards another user (and not what posts those votes were on). $\endgroup$
    – Xander Henderson Mod
    May 15 at 13:28
  • 2
    $\begingroup$ For useful answers in low quality questions, my hypothesis is that there are two conflicting views (or more) that lead to conflicting voting behaviours: 1) The question-answer pair is not useful, hence downvote; 2) The answer is useful given / as long as the question exists, hence upvote. $\endgroup$
    – peterwhy
    May 16 at 20:10
  • 2
    $\begingroup$ "All policies of a site like MSE with highly academic members should not be enforced by moderators" - oh, hell no. That assumes some level of good behaviour from all the users - and that is a very faulty assumption $\endgroup$
    – Rory Alsop
    May 17 at 7:25

1 Answer 1

5
$\begingroup$

As the question has been reopened, I will provide an answer: this is not going to happen.

Enforcement

From a technical standpoint, users may, with very few exceptions, vote (up or down) for any reason. Users are not required to justify their votes, and voting is anonymous anyway. Only a small number of people—all of them StackExchange employees—can only see individual votes. Per site moderators are only permitted to see a very limited amount of aggregate data, which is intended to help us to identify patterns of targeted voting. In general, the tools used on the StackExchange network for investigating and reversing up/downvotes are designed to deal with patterns of voting, which appear to be based on something other than the content or quality of the posts being voted on.

As such, any kind of policy which discriminates between "good" voting and "ugly" voting (in the language of Jyrki Lahtonen's excellent answer) is completely unenforceable. There just isn't any way that the community or the moderators can possibly enforce a policy which requires votes to be justified or explained or cast on "good" grounds (rather than "ugly" grounds).

Site Culture

From a social point of view, there are a number of places on the site where some soft guidance is given about what voting "should" be used for. In the Help Center, it is suggested that

[w]henever you encounter a question, answer or comment that you feel is especially useful, vote it up!

while it is advised that

[d]ownvotes should be used to indicate issues with quality, effort, or accuracy of a post:

  • Downvote questions that don't show any research effort or don't contain enough information to be clear and answerable. These questions may also need to be closed.
  • Downvote answers that are incorrect or don't provide sufficient information to be useful in answering the question. Some answers may not attempt to answer the question at all, and should be flagged. You have a limited number of votes per day, and answer downvotes cost you a tiny bit of reputation on top of that; use them wisely.

In addition, when hovering over the upvote or downvote buttons on an answer, a tool tip pops up, reading either "This answer is useful" (for upvotes) or "This answer is not useful" (for downvotes).

Other advice can be found on the meta portion of the site, e.g. Jyrki Lahtonen's above-cited answer.

However, it is important to note that this is only guidance about how the designers of the site thought about voting more than a decade ago, when the site was established (the Help Center advice and the tooltips have changed very little over time). This is advice, not binding policy.

The take home message is that upvotes are meant to be used for "useful" content, and downvotes are meant to be used for "not useful" content. But what is "useful" or "not useful" is highly subjective, and it is the right / privilege of each user on this site to decide for themselves what, exactly, this means.

TL;DR

What you want simply isn't possible, and isn't going to happen.

$\endgroup$
8
  • $\begingroup$ Thanks for your answer! My question in the OP is Which of the given rules/policies or new possible ones should be considered in MSE? Could you clarify your answer to this specific question? How should I interpret your final statement What you want simply isn't possible, and isn't going to happen in this regard? Do you mean 1) no policy should be considered, 2) no policy is allowed to be considered in MSE 3) the current policy given in Option A, extracted from your earlier comments, should be only considered?, or something else ? $\endgroup$
    – Amir
    May 16 at 5:12
  • 1
    $\begingroup$ @Amir the statement means what it says, and none of your 'interpretations' pay any attention to that. It simply isn't possible -- the software that runs the site does not permit what you want to happen. It isn't going to happen -- StackExchange, the company that owns this site (and others) is not, at any point in time, going to change that behaviour as it is how they intend the sites to run and it is part of their business model. $\endgroup$
    – postmortes
    May 16 at 6:09
  • 1
    $\begingroup$ @postmortes Indeed, the MSE can clearly state any value-related policy it wants to guide the community and asks its members to follow them, and there is no need to change the business model, I never talked about it. All policies should not be mechanically monitored by IT tools, as most of the active members of a scientific site like MSE have academic principles and most of them follow the policies/rules even if they do not like them. Please see section "On policies" in the OP. $\endgroup$
    – Amir
    May 16 at 8:06
  • $\begingroup$ @Amir you write as though you are under the impression that MSE is separate from StackExchange. It isn't; it is one of many sites that are hosted and run by that company. When you propose 'policies' that require software changes (which is what you have repetitively done) that needs to go to StackExchange. And it may require a change of business model to achieve what you want, which is at their discretion, not MSEs. Perhaps you might spend some time learning about the StackExchange Network before continuing your crusade? $\endgroup$
    – postmortes
    May 16 at 10:28
  • 1
    $\begingroup$ @postmortes Please don't use sensitive words like crusader here, I am not looking for any war with senior members here (I just want to help while I know some members like you may consider it as a war). Again and again please note that: I'm also not looking to change the business model or the software (I don't know why you and Xander want to induce it on me). I only want clear explanations are added on the MSE's site for the members on how to use down-votes for answers (which is clearly within the MSE's authority). $\endgroup$
    – Amir
    May 16 at 11:08
  • $\begingroup$ The final output is very simple, just updating the current MSE's guideline in this page vote-down as follows (just as an example): "Downvote answers that are incorrect or don't provide sufficient information to be useful in answering the question. Some answers may not attempt to answer the question at all, and should be flagged. $\color{blue}{\text{Answers should not}}$ $\color{blue}{\text{be down-voted based on}}$ $\color{blue}{\text{any type of non-scientific reasons}}$. " $\endgroup$
    – Amir
    May 16 at 11:10
  • 1
    $\begingroup$ @Amir if that is really what you wanted then you should have created a simple, short question on meta saying exactly that: "I propose to change the guidelines on MSE on downvoting to include the following text..." and then posted two answers, one for agreement and one for disagreement. This is well established practise and you have plenty of reputation so there's clearly no reason for you not to have to done this if it's what you wanted. $\endgroup$
    – postmortes
    May 16 at 11:19
  • 1
    $\begingroup$ Although I have "plenty of reputation" as you say, I've only been a member of MSE for about 7 months, and I don't have enough experience on the meta. Also, the idea given in a comment above is the result of the feedbacks that I have received so far in interaction with other members. I hope the question will be reopened to let other members express their opinions so that we can get to clearer understanding of what needs to change on the site. Posting a poll-like question without discussion doesn't seem useful at this point since the subject hasn't been discussed in meta for long time. $\endgroup$
    – Amir
    May 16 at 17:04

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