TLDR: Split up voting, and only give reputation when question is concluded (answer is accepted). Then award reputation based on activity after, by people who revisit the topic and vote.

DISCLAIMER: You aren't required to read linked questions. I explain what they seem to be about in Analysis.

It has passed some time, and I am not longer butthurt, so I thought I would try again and contribute to ye old issue that seems to repeat itself - the SE/SO Drama. Often when this happens in real life it seems to be due to discussing different levels of topic, or approaching it from different perspective. So I tried to find out root cause and not look at the leaves.

Main hypothesis is: "Evaluation of questions (good questions == good answers) is flawed and gives us ephemeral result that is cause of all prior and future drama".

Implied through the ephemeral concept is that questions are searchable when relevant (via searches), but because evaluation is flimsy, it has a short search engine lifespan. Because of this, newer questions tend to be duplicate or unanswered.

This in turn reflects how the platform fulfills users needs and core mechanics of the platform (keeping the content searchable and maintaining community) which elevates the topic into polarized conversation - Drama.


From the first question about "Evaluating the quality of answers instead of questions" it seems that there is a strong census that there are bad questions (which need to be pruned since they yield bad answers). Thus strictly moderating content standards for quality of questions/answers irregardless of topic is a necessity. So questions should be informed and educated in order to even qualify for evaluation.

From the second question about "Why is SE giving so much attention to the "be nice"-policy?" where the main topic seems to be relationship between users (expanded to moderator/user relationship). The main strife seems to be the question/answer policy, because it seems to produce drama between "parents". Where the topic soon derails into two sides of "content keeping" and "social interaction" - often escalating away from the original topic.

From the third question about "Exit strategies for chameleon questions" where main the topic seems to be returning the topic back to its subject - "how content made by people impacts our own prejudice of "majority" of users" (Chameleon questions, low effort questions, house keeping, etc.). This in turn makes people who are interested in keeping the sanctity of content approach it from a strict point of view, which in turn get told on to "mom" - i.e., Mystical's Answer about DDT that eventually sparks into full on drama.


It seems that every topic returns to the questions, which in turn impact how people interact with content. This also implies a perspective on content, because to some answers are crucial and would require one answer per question, and to others different perspectives are crucial and would require multiple options.

What I've come to notice is that users with long-standing use of this platform tend to divide across spectrum of "Archivists" (content focused users) and "Socialites" (community focused users).

  • The Archivist type of users tend to focus on ease of access, often gained with years of using platform, and clarity of content. In Jon Ericson's answer they could fill the role of Persecutors.
  • The Socialite type of users tend to focus on community, indulging in chat options, badges, reputation and status filling the role of Rescuers.

As extreme representations of spectrum, those are opposing forces that keep this platform running. "Archivists" tend to be ones that make it possible for people to find what users need, while "Socialites" tend to be the face of the platform, driving its content. Those are parents of SO/SE and without both we wouldn't be able to function. Give respect where respect is due.

Dreaded drama triangle

However, in DDT (Dreaded Drama Triangle) we are missing one important corner which is the "Victim" (which is bad naming in general and I will be calling it "Subject")—the one we always forget when Drama gets heated.

  • The Mort type of users which is right at the middle of the spectrum, less knowledgeable than "Archivists" and in lower status than "Socialites" who use SO/SE as their playground, main source of content and action, but lacking in ways to communicate with both Archivists and Socialites - your one and only friendly neighborhood Subject.

Bell Curve of Social distribution

We can safely say that the amount of MORTs is much larger than the amount of Archivists and Socialites combined. In ideal world, MORTs would slowly transition within the spectrum of Socialite or Archivist. However, that seems to yield low efficiency, since MORTs are mostly populating continent that result in drama.

Users who tend to strive to either extreme of the spectrum easily find their place here. Due to their preferred MO, they tend to find those who understand them. Baby Archivists quickly learn how to navigate the search bar, and baby Socialites tend to follow this type of users and learn from them directly. MORTs unfortunately tend to be banned into oblivion quickly, racking up downvotes and losing reputation gaining the Asking Ban very quickly - which starts the cycle.

Determining possible cause

Within this context, I was curious why would people polarize this way exactly. The main issue seems to be with questions so probably the cause lies there.

Questions tend to have a few integrated functionalities:

  1. Voting: Allowing opportunities to express the quality of the question.
  2. Archiving/Saving: Allowing for users to save the most useful questions
  3. Sharing : Allowing the platform to gain more users
  4. Editing : Allowing users to rephrase the question to maximize its quality.
  5. Following: Allowing users to track status and changes to the question
  6. Flagging: Allowing users to distinguish the question for whatever reason.
  7. Commenting: Allowing users to interact with the asker without answering the question + Voting and Flagging

We can split these functionalities into three "brackets":

  • Content or functionalities of Archiving, Editing and Flagging
  • Social or functionalities of Sharing, Following and Commenting
  • Morty or functionality of Voting often used by both. Archivists would use voting to determine quality of the question, and Socialites would use voting to gain/lose status since voting is easiest and direct way to gain/lose reputation.

This means that voting is used to directly express quality of content, but as well as expression of the user—which would make it very misused.

Within context, this could switch the "primal" roles of Archivists and Socialites (as Persecutors and Rescuers respectively) depending how voting is used.

  • If Socialites vote on a question for status, that messes up with the Archivist role, making questions without answers, or closed malformed questions. (Socialites are Persecutors and Archivists are Rescuers)
  • If Archivists vote on question for content, that can result in repercussions of Socialites, because it harms community (i.e., stops noobs from being part of the platform). (The Archivist is a Persecutor, and the Socialist is a Rescuer)

So whenever this issue arises, it is left unresolved, because both sides have valid reasons for doing so and a plethora of history to call back upon as an argument. This removes the subject from DDT and focuses on the extreme sides of the spectrum. These types of issues pile up till it explodes unpredictably into a sitewide drama.

Image separator for ease of reading

Without any resolution, MORTs have no other way to refocus the issue and start abusing other features. They start flagging users that they are angry at, and it rises till mods get banned, fired or reinstated.


So to test out the hypothesis, I am using Voting as the only MORT-like functionality in regard to keywords that are representing a conflict between Socialites and Archivists (in that priority).

I am searching for keywords that indicate conflict or drama (such as "hostile", "be polite" and "nicer to new") and representing them in metric that is quantifier of Social aspect of voting (Up/Down ratio).

Result of SEDE query about mentions of keywords regarding human interaction

The SEDE query above is the result of searching for keywords and then comparing the amount of views to the verdict (the ratio between upvotes and downvotes per question) giving us a count of users per verdict (Vcount) - SEDE link for potential Archivists.

Looking at the graph (from the older question), we can notice a pattern where there is a "peak" of the Drama (escalated/deescalated by external influences). We can notice that at years 2011, 2013, 2016, and 2019.

This means that how we evaluate questions is ephemeral since the same keywords show periodically, meaning repetition of the same problem.

To clarify, this isn't a problem of Reputation nor tied to Archivists or Socialites. This problem is tied to the fact that voting is used for both, and therefore useful to neither. Actions taken by one can be undone by another, which then in turn produces strife. Simply put - evaluating quality of questions via voting doesn't work.


If this hypothesis is correct, then we should be able to observe a rise in drama this year with possible culmination on 2022/2023 - thus why I am writing this now.

Proposition for a solution is pretty simple. Balancing the amount of reputation points per upvote/downvote is just influencing how much Socialites can contribute to this platform. We need them, and thus it always sucks.

Instead, split voting into two - effect and affect.

  • Affect - main representation of voting and it doesn't yield any reputation points. The Archivist type of user/moderator will then have an easier time to comb over questions and answers and make a proper archive.
  • Effect - secondary representation of voting, where reputation points are gained only if question is of a proven value. Quality questions gain Quality answers that are often revisited much later and gain traction steadily - serving as a reputation points farm. Reputation points should be defined similar to the verdict (Upvote/Downvote ratio) and therefore it would give an incentive to users to edit them, interact with the community and keep the archive up to date.

Clarification - ongoing: Once a question is asked, for a period of time, users don't gain or lose any reputation points. They are evaluating the quality of the question till the answer is provided and accepted. Once an answer is provided, the user will gain reputation points based on "verdict" (Up/Down - 1) x BaseValue. Bad questions won't yield a negative reputation, but will fall in relevancy (and deleted), and good questions/answers will gain reputation and rise in relevancy—till they are obsolete or updated. Maintain not suffocate.

What do you all think?

  • 2
    If I get you, you mean a sort of "like" button, but not tied to rep. Is that right? How many votes does anyone get, can they vote both ways on the same question? Would there be other incentives eg. badges? Isn't this going to be seen as a way towards allowing "mindless social fun" by the back door?
    – W.O.
    Oct 11, 2022 at 23:06
  • @W.O. I don't think that any "like" type of button would serve any community any good. To elaborate a bit further. Once question is asked, for a period of time users gain/lose no reputation. They are evaluating quality of the question till answer is provided. Once answer is provided user will gain reputation based on "verdict" (Up/Down - 1) x BaseValue. Bad questions won't yield negative reputation but will fall in relevancy (and deleted), good questions/answers will gain reputation and rise in relevancy - till they are obsolete or updated. Maintain not suffocate.
    – Danilo
    Oct 11, 2022 at 23:14
  • So, questions like this one (6 years old, unanswered) would earn the author no rep because no answer is given? (OK, main meta's a bad example, but you get the idea if good-quality unanswered questions languish).
    – W.O.
    Oct 11, 2022 at 23:18
  • @W.O. Exactly! Questions like my one (if not answered) would earn author no rep, and would also be eventually deleted. What is the point of question if it doesn't produce discussion or solution? It is a clear/subtle signal that answer won't be found and should be searched elsewhere. And regarding your link, yes as well. It is a nice question but no resolution has been found. So it either has no support (Question is too early for its age) or it is trivial.
    – Danilo
    Oct 11, 2022 at 23:20
  • 2
    Why is this tagged [site-recommendation]?
    – bobble
    Oct 11, 2022 at 23:26
  • @bobble For implied use of the "Proposition" functionality. Less clutter less "where should I..." in short of googling. Feel free to remove it if it doesn't follow general consensus.
    – Danilo
    Oct 11, 2022 at 23:29
  • 1
    I don't understand your comment or the "Proposition" section, if either is supposed to be an argument for why this is relevant to a tag for telling people which site they should ask their question on.
    – bobble
    Oct 11, 2022 at 23:44
  • 1
    Based on this comment: "Once question is asked, for a period of time users gain/lose no reputation." Does this answer your question? What you think it would happen if it was a "time delay" between an "up-vote or down-vote" and it's sum in the answer?
    – Rob
    Oct 11, 2022 at 23:49
  • @Rob not even in the slightest. That questions asks about time delay to impact freshness of answers. Please read the question, just because one word is the same it doesn't mean entire question is the same... Because answers and questions aren't the same.
    – Danilo
    Oct 12, 2022 at 0:10
  • 1
    Danilo, that's the canonical (see the banner) for meta.stackexchange.com/q/363452/282094 which covers both questions and answers.
    – Rob
    Oct 12, 2022 at 0:15
  • @Rob this question isn't even related to that one. One you have linked suggest hiding votes due to prejudice, mine doesn't deal with that at all. I don't mention even once such issues, quite opposite I advocate for evaluation of the posts which is objective measure not subjective.
    – Danilo
    Oct 12, 2022 at 0:19
  • Please people, there is clear difference between cause and effect, analysis and speculation. This question isn't "time delay" but "evaluation after conclusion", it isn't "prejudice of virtual point system" but "effect of undefined system". It's not the same.
    – Danilo
    Oct 12, 2022 at 0:24
  • 9
    As much as I want to figure out what you're saying in this post, there is far too much waffle and not enough clarity in the remainder. Whatever you might be suggesting, it's never going to happen, if only for the lack of determination of what exactly it is.
    – Nij
    Oct 12, 2022 at 1:24
  • 11
    Please be clear, not clever. Please be concise, not prolix. Put what is needed & only what is needed in your post & don't expect us to read an entire link & figure out what part has what to do with your post.
    – philipxy
    Oct 12, 2022 at 4:00
  • 9
    @Danilo this is one of those posts that badly needs an executive summary. I added a TLDR, but feel free to adjust it Oct 12, 2022 at 4:26

2 Answers 2


I used to play RPGs and specifically a certain table top game called BattleTech. Folks were obsessive over details, and a wise old fella would sometimes point out 'it’s an abstraction'.

Reputation and voting is an abstraction - voting on its own is already a reflection of value. The thing is SE's always been deeply gamified. We have high score tables and folks who play the game for different reasons. I'd even say its almost an MMORPG for problem solving.

My main hypothesis is that way how platform evaluates quality of content is ephemeral. Due to this, searchability seems to be tied to engagement of a question/answer and when engagement falls, new duplicate arises (often phrased differently or misnamed). This in turn activates "content keepers" that act impartially, which results in hurt feelings (because reputation is necessary for this platform) that eventually pile up into drama.

I've often talked about the 'motivation' for people to be here, and to me the idea of different types of users is ... simplistic. I'm a moderator on three sites, and the tools, approaches and requirements of each community I work with are different.

I've often felt that SE shouldn't bill itself as a place to ask questions. Its a place to find answers. In a sense we're a tool for knowledge management and linking subject matter experts with topics that interest them.

Questions tend to have few integrated functionalities

We care about two things really, that the question has all the information needed to solve the issue, and that its useful and findable for someone with the same issue. everything else revolves around that.

As extreme representations of spectrum, those are opposing forces that keep this platform running. "Archivists" tend to be ones that make it possible for people to find questions they need (and probably often flagging dupe questions), while "Socialites" tend to be face of the platform answering questions, editing and guiding users in right direction etc. Those are parents of SO/SE and without both we wouldn't be able to function. Give respect where respect is due.

Answering, editing and guidance are curation tasks as much as flagging and closures.

It’s also worth considering, unless you are, metaphorically speaking, playing for a high score, past a point reputation doesn’t matter.

If we really want to do this right, let’s pull out the D&D alignment charts

enter image description here

and with a bit of work

enter image description here

Well, we also need a "I'm only here to play The Settlers of Catan" box for folks who assume this is a forum, but I'm lazy.

To me "good" is engaged and passionate about the site, neutral is folks who are potentially engaged, and Evil is people who actively hurt the site.

Lawful is how 'seriously' folks take the letter of the rules - and focus on essentially ensuring curation tasks are done, neutral is people who are not particularly engaged, and chaotic good here is folks focused on the 'spirit' of the rules rather than the letter of the rules.

As I see it your 'Morts' are neutral. To me, the goal is to move them to the left (and get them engaged on the site) and up. Very few people go from neutral to evil, and when SE messes up, we have people going from the top row, down. We've had moderators go rogue, so that's to be avoided too.

Strangely enough, to me, rep mainly matters when you're neutral good - your lawful good or archivist/curators often find the act of curation to be its own reward. I certainly don't get an real reward from what I do here that's extrinsic - and I vote based on the utility the posts have. My shaping of these communities is purely from the work I do with the community via meta and chat. I'm not focused on my very impressive stats any more, and to borrow another RPG metaphor - I'm focused on the story now that I'm basically maxed out on all my stats (except maybe spelling). If everyone was like me, the boring stuff would take longer to do - the queues, and close voting are important too, but if we were too focused on just curation it would be soulless and very boring.

Now looking at the core of your proposal

Clarification: Once question is asked, for a period of time users gain/lose no reputation. They are evaluating quality of the question till answer is provided. Once answer is provided user will gain reputation based on "verdict" (Up/Down - 1) x BaseValue. Bad questions won't yield negative reputation but will fall in relevancy (and deleted), good questions/answers will gain reputation and rise in relevancy - till they are obsolete or updated. Maintain not suffocate.

Who does this benefit? Our Evil folks are just going to create another account. Our Good folks know how the game is played. Let’s focus on the folks in the middle.

Folks who just want an answer ... are not going to care about the metagame - so our real focus should be on converting and giving a good experience to random internet folks so they keep coming back and helping new users find their feet. Preventing reputation loss until a question is answered basically means you can ask as many 'poor' questions as you want unless they get answered unless you rejig the question ban system, which results in a worse experience. It’s also a lot more complicated.

  • First of all, I was very giggly all the way while reading your post. Your humor is very resonating with me. Second, I am not discussing motivation as that is as undefined as it gets, and it would force me into speculation. I am discussing how users use platform, where it is basically how users adapt to what is presented. Thirdly, even though I am a DnD player I will refrain from using it, due to alignment charts are based on interaction which is subjective. Fourth, way users use platform is not meaning that it doesn't have a role in moderation.
    – Danilo
    Oct 12, 2022 at 13:07
  • I agree that social networks are gamified became that is how they keep users on the platform. And yeah, most social networks (SE included) uses MMO style of game, but main difference is longevity of the game. Just because a concept is abstract it doesn't mean it has only abstractive repercussions. Going by same example of MMO, those games don't last long and tend to hold their players hostages due to investment amount (money or time). MMOs are also known about resulting toxic behavior, which are in game design well known and have direct cause/effect correlation.
    – Danilo
    Oct 12, 2022 at 13:12
  • Same principle I've described in the question can be translated to MMO just substituting core concepts . So removing gamification will be harmful to the platform, I am aware of it - so I am substituting it. Instead of toxic based mechanics I am proposing implementation of economic based mechanic. In market terms, verdict is equal to volatility and base amount as price while reputation is based on dividend. With that toxicity should be greatly diminished in favor of expertise.
    – Danilo
    Oct 12, 2022 at 13:17
  • Which is also an answer to "Whom this benefits?" - everyone. Mods have less exposure to aggravating content due to volatility where moderating Archivists and Socialites can determine base amount or price /value of evaluation. Verdict or volatility makes process less harmful (evil) because there is no person who can "invest" more than 1 vote without too much hassle for the yield. Reputation gained is a qualifier of maintained topic (in order to stay relevant) and thus it helps remove clutter thus making content more searchable. Its a game still, but one producing less Drama.
    – Danilo
    Oct 12, 2022 at 13:21
  • We then only have to "worry" about probabilistic of interaction (Game Theory) which is easier to handle socially than broad, diverse and unidentified subject. To put it simply, because of the Affect/Effect system, asking a question for those ambitious will be an investment - driving more effort in high quality content. Those who constantly fail aren't punished but have more incitive to search and ask questions - allowing for learning opportunities and removing language barrier. Thus sublty directing MORTs into type of use that they are most comfortable.
    – Danilo
    Oct 12, 2022 at 13:25

Why is my answer better just because the person who asked the question returns and accepts it? Have I not helped anyone who upvotes my answer unless that happens too?

The question asker is one of the least likely people to know that an answer is a good answer. If they were familiar with the subject matter they generally would not have asked the question. So why should they get veto power over every other voter?

What you'd encourage is answerers pestering question askers to accept their answer so they can start to get reputation. We get enough of that already without encouraging lots more of it. Beware of unintended consequences.

  • Its not, but that is not the point after all. Once question is accepted it is left to the test of time. That is the main measure of quality and information. Try to think of it in long term, if any answer is accepted it doesn't mean that it will result in reputation. People who come to the platform will upvote answer that helped them. Sure a posy might come and upvote to gain reputation, but if question/answer is helpful those numbers will be meager even if it is botnet. If, for a chance, there is a corruption it would be easier to notice based on other top ranking answers.
    – Danilo
    Oct 12, 2022 at 12:29

You must log in to answer this question.

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