-9
\$\begingroup\$

Quoting the official description of downvoting priveledge:

Use your downvotes whenever you encounter an egregiously sloppy, no-effort-expended post, or an answer that is clearly and perhaps dangerously incorrect.

There is nothing like "Because I have a different opinion" on the list, and all the possible reasons listed would also most likely require some other action, such as trying to improve the answer by commenting and/or editing, or, if it is so bad that it cannot be improved, flagging it so it is sooner or later deleted. In the last case things are obvious, downvoting is a temporary action so that the answer sinks in the bottom in the time before a mod can review it and delete.

In first two cases downvotes will remain even after the answer is improved, and make a negative impression for those who will read it later. As of my experience on other sites with voting system, if you see a vote before you vote yourself, score is very unlikely to change because of some strange form of hive-mind. If people see a negative score, they most likely vote down, often even before reading the post, and vice versa (vote up before reading the post) if the score is overwhelmingly high. This means that only the first couple of votes really tends to affect the overall public reaction to a post.

As I understand SE's policy, it is more important to improve existing answers than to delete them, so a downvote may possibly mean "Improve it". However, I have seen a lot of answers downvoted without any reason specified, and some of my generally positively rated answers downvoted without any negative feedback in the comments, I would assume that everyone around has experienced this kind of stuff. An upvote is a clear reaction of "This is a good way you follow!". A downvote is "You didn't do well!" without actually specifying what's wrong, so it is not really helping the author to improve his post writing skills or his post.


So, here is how I would see a possible new voting system.

  • Votes get hidden from public view, and you can only see vote count after you vote yourself. You can try to guess the approximate score by question's or answer's position, but nothing more. This will make users think individually instead of behaving as a hive. You can always see a vote count for your own posts.

  • New answers are shown at higher positions for some short period of time than they should be based on their score only, just as comments on YouTube or game reviews in Steam Store, so they also have a chance to get public attention. Otherwise, on average, comments already in the top tend to get even more attention and be upvoted by hivemind, and new posts just don't get the same amount of views to hijack high positions. It is possible sometimes that a post is not relevant anymore. For example, the author asked for a rule clarification and the current accepted and/or top answer is based on interpretations and even some assumptions, and a new answer appears after a year when this rule is clarified by game developers or someone finds a clarification elsewhere in official sources. Of course, it is often important to help the asker in-time, when he needs to use a rule in a game session rather than after a year, but I see the primary role of this web-site as an archive of answers that you can read at any time after the question was asked.

  • When you downvote, you are asked if you really want to do it by a popup, and have to confirm your choice. Then, if you do, you are forced to leave a special anonymous comment that describes your reason of downvoting, or vote for an existing one (to avoid "Me too!" comments), you can select and write several reasons. Such downvote comments are only visible if you can see the post's vote count (it is your own post or you have already voted for it) and needed to give the user knowledge of what he did wrongly without breaking the anonimity of comments.

  • There should appear a special priveledge similar to voting for closing questions which would allow declaring a downvote reason irrelevant for the post: if some given amount of users votes against a downvote reason. If a downvote has no more reasons marked as relevant, it is no longer counted. It should be used if a reason is irrelevant by the moment of removing it, such as if the post was edited or the reason was irrelevant from the beginning. For example, if users downvote a post saying that the question is about Vampire: The Masquerade and the answer quotes Vampire: The Requiem rules without even bothering to convert them, and the issue is solved by an edit, for example, by quoting relevant VtM rules, users can vote for this reason to be irrelevant and the downvotes will be removed. If, however, there is still another issue left, like "This question has too much of irrelevant info", those downvotes that have also been associated with it will stay.

I hope that my system isn't too complicated and it is clear what I mean.

This is not a duplicate because I am suggesting a system to fix things rather than just naming what's wrong.

\$\endgroup\$
15
  • 6
    \$\begingroup\$ In my experience, mass upvoting of just the best answers happens because one gets tired of reading answers after a while, so position is a worse offender than the actual number of upvotes. \$\endgroup\$
    – Zachiel
    Aug 4, 2016 at 21:31
  • 1
    \$\begingroup\$ @Zachiel Right you are. So here goes my "New answers are shown at higher positions for some short period of time" suggestion. The problem is especially relevant for questions where answers are exceptionally long. \$\endgroup\$ Aug 4, 2016 at 21:36
  • \$\begingroup\$ Now this question makes a perfect example for itself -- two downvotes without any reasoning given. That's why my suggestions go. \$\endgroup\$ Aug 4, 2016 at 21:47
  • 9
    \$\begingroup\$ Please read the existing discussion on this subject. You're stepping into a seven-year-old topic with a lot of complex nuance relating to fundamental network goals. You'll find that many of your suggestions have been discussed on meta.se, and that there's a lot of practical experience from trying different variants of the mechanic over the years. \$\endgroup\$
    – BESW
    Aug 4, 2016 at 22:00
  • 3
    \$\begingroup\$ Possible duplicate of Downvoting behaviour does not enable me to improve my post \$\endgroup\$
    – BESW
    Aug 4, 2016 at 22:12
  • 1
    \$\begingroup\$ If it's not a duplicate of that one, try this, this, or this. You may also find our help page about meta useful, especially the "Voting is different on meta" section. \$\endgroup\$
    – BESW
    Aug 4, 2016 at 22:13
  • \$\begingroup\$ @BESW I think that I read most of info (it's just 18 posts) in meta.SE by tags you provided, and didn't really find a complex system that I suggested, posts are mostly pure complaints about how bad things are, some are really old and address things how they were years ago. Your words "possible duplicate" and "If it's not a duplicate of that one" make me think that you didn't even read the posts yourself and just want to show off and/or discredit my suggestions, I hope that I am mistaken. \$\endgroup\$ Aug 4, 2016 at 22:25
  • \$\begingroup\$ The answers to your post would be a collection of previous answers, b/c most/all your suggestions aren't new; the new bit is you're using several of them in tandem, but you don't attempt to rebut (or even acknowledge) the reasons they've been rejected separately. "Gets the same answers" is a common Stack rule of thumb for identifying duplicates. \$\endgroup\$
    – BESW
    Aug 4, 2016 at 22:56
  • \$\begingroup\$ Your collected suggestions also seem at odds w/years of Stack experience learned about how to best gather the wisdom of crowds, so I'm not seeing evidence of familiarity w/those parts of the discussions. Your comment about downvotes implies you haven't read the existing discussions on RPG.SE either. \$\endgroup\$
    – BESW
    Aug 4, 2016 at 22:56
  • 7
    \$\begingroup\$ So no, I'm not being disingenuous or hasty. If you want to be taken seriously on meta.se, you need to show your work and build on the history of the subject, and I'm linking you to resources which are vital to build on if you're going to get anywhere with this (eg, one reason requiring comments explaining downvotes gets recieved poorly is that they're sadly often an opening for argument and personal attacks). You will also get a better response if you have specific linkable examples of the problems you're trying to solve. \$\endgroup\$
    – BESW
    Aug 4, 2016 at 22:58
  • 6
    \$\begingroup\$ Please don't accuse people of egoism for taking the time to dig up resources for you; you can be gracious even if you already have that information. Would you rather that I'd not participated at all? because that's kinda what I'm wishing I'd done now. \$\endgroup\$
    – BESW
    Aug 4, 2016 at 22:58
  • \$\begingroup\$ If I'm wrong, tell me I'm wrong. I'm wrong a lot. But try to assume incompetence before malice, eh? \$\endgroup\$
    – BESW
    Aug 4, 2016 at 23:17
  • \$\begingroup\$ @BESW Sorry for misunderstanding then, and thanks for possible help. Will try my luck at META.SE. \$\endgroup\$ Aug 4, 2016 at 23:17
  • 3
    \$\begingroup\$ Good luck on meta.se. When you are there I suggest you assume any who engage with you are (a) well-intentioned, (b) listening, and (c) intelligent and thoughtful citizens. They may not all be, but you'll not be far off assuming that of most around these parts. \$\endgroup\$
    – nitsua60 Mod
    Aug 4, 2016 at 23:49
  • 1
    \$\begingroup\$ You may also want to visit a few other stacks, where the very same software and system is used by a different set people and role models. You may have a very different experience based on their use of the system. \$\endgroup\$
    – nvoigt
    Aug 5, 2016 at 11:42

2 Answers 2

11
\$\begingroup\$

Some of these ideas have already come up in the same or similar forms in Stack Exchange's history. Many have gained no traction or been firmly rejected; there's a reason they aren't part of our system yet. I suggest you do some research into the topic so far on Meta Stack Exchange, and I'll be posting relevant areas here.

  • The issues mentioned in your question and comments about how earlier answers tend to gather more votes is known as the Fastest Gun in the West Problem and even has its own tag on meta.
  • Hiding vote counts for a while has gained no traction: How about hiding answer score for first few hours?. Hiding them forever until voting is something I haven't seen.
  • Randomly reordering stuff temporarily has received little support: Randomly reorder all answers posted within ten minutes of each other
  • I'm going to get back to the hiding votes & randomly ordering stuff at the end of this answer.
  • Requiring comments for downvotes comes up a lot, including on our own site and on meta stack exchange, and is eternally and forever declined. The staff added a notice that says "please consider adding a comment if you think this post can be improved" and that's about it. They are never going to require explanation. Requiring a formal reason picked via a dialog is something I can't search up as pre-existing, but my gut tells me it's probably been discussed.
  • The system of invalidating other peoples' downvotes because you consider them invalid isn't something I've come across or can search up, but I expect it to go poorly. It's our vote to cast, and we don't cast votes just for kicks.

Overall you will want to pay attention to the voting and (as already mentioned) fastest-gun tags on meta, and especially feature requests & discussions.


Please be aware that hiding votes and randomly ordering answers can actually be (and probably are) just really bad. People visiting our site are looking for the best solutions to their problem. Our scores and answer ordering exist almost solely to guide people toward those best solutions. You removing them means nobody has any idea which answer is any good, so it's almost certainly just an awful idea.

Doing that in the early stages of the question just means a person has to wait hours or days for their answer, when they'd really often rather have one quite soon. Remember that Stack Exchange is built first and foremost for Stack Overflow, Server Fault, and Super User, which are generally for people who have actual jobs to get on with, and time is of the essence. The fact it's also used for fun for RPGs is coincidental — Stack Overflow, the company, primarily exists to serve software developers. We're kinda just hitching a ride on their tangential effort to improve things in other fields with a good Q&A engine. This also means that the Fastest Gun in the West "problem" is also regarded by some as a positive effect because it encourages people to answer fast.

\$\endgroup\$
5
  • \$\begingroup\$ Thank you very much for the info provided! I only disagree with the reordering problem. As users are discouraged from asking duplicate questions, one of the purpose of stackexchange is a database of answers. Even though it is really important to get the answer quickly no matter if you need it for your job or for the next game session, there is a good chance that someone will show up after a year and write an answer outshadowing all the existing ones. If it is the second answer provided, things are OK, both are likely to be read. If it is the 15th answer to a hot topic... You get me. \$\endgroup\$ Aug 5, 2016 at 11:55
  • \$\begingroup\$ Plus, it is not about randomly reordering. I suggest ordering them based on several variables: time and score. Accepted answer always stays at the top. A viewer can always select to sort answers by time, score or both. Thanks again for the info. \$\endgroup\$ Aug 5, 2016 at 11:57
  • 3
    \$\begingroup\$ It might be worth noting, to strengthen your "people visiting our site are looking for the best solution..." statement, the stat I saw on Meta that something like 99% of page views are direct-from-Google non-registered-users. In other words, the vast majority of people who see a page aren't citizens, they're tourists. Having the accepted and top-ranked posts immediately visible to tourists is an indispensable provision of our core functionality: expert answers to well-posed questions. This, in turn, is how we get a reputation as the place... \$\endgroup\$
    – nitsua60 Mod
    Aug 5, 2016 at 13:44
  • 1
    \$\begingroup\$ ...to come for excellent content, is how we eventually convert tourists to citizens, is how we maintain the citizenry. \$\endgroup\$
    – nitsua60 Mod
    Aug 5, 2016 at 13:44
  • 3
    \$\begingroup\$ This MSE question contains a lot of useful context, links, and thoughts from the staff regarding the “top answer is wrong, what should we do” issue: Let's move some negatively scored answers from the top spot. May be worth mining for this answer. This MSE question is a proposal to use a different default sorting that would give new answers more chance to be seen: Sort Answers by Projected Score Instead? \$\endgroup\$ Aug 5, 2016 at 16:16
8
\$\begingroup\$

As part of the Stack Exchange network, RPG.SE's site mechanics are the same as every other Stack site: nobody on RPG.SE can change the voting system, we're just users of the system. Requests for changes to fundamental site mechanics need to be made on the general Meta for the entire Stack Exchange network.

First, though, I suggest you take a look at the [down-vote] tag on META.SE and familiarise yourself with the history of the issues you've recognised.

\$\endgroup\$
1

You must log in to answer this question.

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