Justifying downvotes

Service: Narrative

In my opinion giving a reason to a downvote could solve a lots of issues such as decreasing or making visible at least the unfair or abusiv downvote and increasing the quality of posts as the authors know what to do differently in the future. 

This could be done by forcing the downvoters to select a reason  for downvoting from a dropdown list or writing an other reason. 

How do you see this suggestion?

 

Original Post

Activity Stream

Currently, Narrative requires you to select a reason for down voting. Can you help describe how you would change this moving forward? Are you looking for different reasons?

Thank you for the answer!

I remembered that I downvoted a content and it didn't required a reasoning but I just checked and it seems like I didn't actually downvoted so I didn't noticed that Narrative asks for reasons to downvote.  I apologize for the mistake. 

Although I would like to add that in my opinion it would be beneficial to the authors to see the reasons why a certain content was downvoted by others. In this way authors could learn from their and other's mistakes. I, for example, have a post that doesn't have 100% rating and I have absolutely no clue what could be the reason and what to do in the future to avoid being downvoted. I searched for an option to see the reasons of the downvotes but couldn't find anywhere. 

I also think the transparency behind the downvotes could also help to fight abusive downvoting by making it easier to authors to report someone who downvoted for an unjustified reason. Let's say if someone abusively downvotes my post for being fake news I would be able to notice this and report the user, given that I also provide proof that my article it's not fake news. In this way an author's reputation will not be affected by downvotes intented with unfair purposes.

I think the list of reasons people could choose from should be a little bit more extensive and it should also allow people to write their own justification, if the one they have in mind is not included in the list. 

Thank you.

 

I'm not sure if a suggestion was made on this site before - but there has been a lot of support in community discussions for requiring a typed reason for downvoting.  Just a simple text field in the current downvote popup, where people have to provide a comment.

There are several advantages to this.

First, it would discourage abusive downvotes because leaving a phony or gibberish reason is actually leaving a trace in the system that the downvote was illegitimate.  There will always be the specter of the system detecting this in some way, and applying a penalty for it, so people will think twice.

Second, it leaves the door open do devise ways in which phony or gibberish downvotes can result in a penalty for the downvoter, and/or can be ignored in the computation of quality.

This can happen through moderation and/or automation.

When moderation goes live, the moderator will read a piece before approving it as on topic for the niche.  If any downvotes applied to that piece are forwarded to this same moderator, they are in a good position to determine, with very minimal effort, whether the reason given is in good faith or not.

In terms of automation - a dictionary check with a little bit of smarts for accepting misspelled words, would be able to ignore downvote reasons that consist of "sdc QNocnds sdav r", and even stock non-responses such as 'this sucks', or 'this is crap'.

 As @Szabo points out, people being able to see these reasons would remove the suspicions and doubts about why downvotes are landing on their posts.  And in some instances, their paranoia might reveal itself to be unfounded (in others not).  In any event, this sort of transparency could be very beneficial.

 

Malkazoid posted:

I'm not sure if a suggestion was made on this site before - but there has been a lot of support in community discussions for requiring a typed reason for downvoting.  Just a simple text field in the current downvote popup, where people have to provide a comment.

There are several advantages to this.

First, it would discourage abusive downvotes because leaving a phony or gibberish reason is actually leaving a trace in the system that the downvote was illegitimate.  There will always be the specter of the system detecting this in some way, and applying a penalty for it, so people will think twice.

Second, it leaves the door open do devise ways in which phony or gibberish downvotes can result in a penalty for the downvoter, and/or can be ignored in the computation of quality.

This can happen through moderation and/or automation.

When moderation goes live, the moderator will read a piece before approving it as on topic for the niche.  If any downvotes applied to that piece are forwarded to this same moderator, they are in a good position to determine, with very minimal effort, whether the reason given is in good faith or not.

In terms of automation - a dictionary check with a little bit of smarts for accepting misspelled words, would be able to ignore downvote reasons that consist of "sdc QNocnds sdav r", and even stock non-responses such as 'this sucks', or 'this is crap'.

 As @Szabo points out, people being able to see these reasons would remove the suspicions and doubts about why downvotes are landing on their posts.  And in some instances, their paranoia might reveal itself to be unfounded (in others not).  In any event, this sort of transparency could be very beneficial.

 

This would be really useful for feedback, too!

I'm fine with the current setup of downvotes, and especially with them being TOTALLY anonymous.

I think we put too much emphasis, on the need to know the reasons for downvoting. You should never put pressure on a reader to know why he or she disliked, in the real world as online. Maybe transparency about the selected reason is okay, but no more, to me.

Algorithms can be set to detect many abusive downvoting.

MALAY BANERJEE posted:

Yes if anyone downvotes any post, he/she should make a comment to justify downvote.  Recently two of my post downvoted by someone without any reason.

Look at my post.

https://www.narrative.org/post...-are-replacing-human

https://www.narrative.org/post/story-of-a-refugee

Please make it mandatory to mention a reason in the comments section if someone downvote any post.

Thanks everyone.

 

I think the downvote should remain anonymous.

So if the typed reason goes in the comments, it should be a special entry there where the identity of the downvoter is not displayed.

The other option is to not put the mention in the comments, but for it to be a mouseover overlay when the mouse hovers over the quality score.

I like @Malkazoid's suggestion of a text box where an actual reason has to be typed in, and authors getting these anonymously. Content creators could receive a notification that they received a downvote with the typed out reason why the post was downvoted. They can then, if they wish, make an appeal by clicking on a nested "downvote comments" section on a page. It could look something similar to the Tribunal appeals voting page. Make it all transparent, but anonymous.

Garden Gnome Publications posted:

I like @Malkazoid's suggestion of a text box where an actual reason has to be typed in, and authors getting these anonymously. Content creators could receive a notification that they received a downvote with the typed out reason why the post was downvoted. They can then, if they wish, make an appeal by clicking on a nested "downvote comments" section on a page. It could look something similar to the Tribunal appeals voting page. Make it all transparent, but anonymous.

Yes - "downvote comments" link like for the appeals comments: this is probably the best way.

And that's a great idea for the author to be able to appeal upon review.  It would reduce the number of downvotes needing the attention of a moderator.  Fantastic.

@Vico Biscotti Providing an explanation for downvoting is like giving a productive feedback to the author. Just like in the field of education or businesses, a professor or a manager doesn't simply dislikes an assignment and leaves it like that, but also provides an explanation which improves future productivity and quality. Similarly here, on Narrative, it's in everyone's interest to increase the quality of the contents and to avoid unfair actions which can be achieved. For the authors the identity of the downvoter (and who also would have to provide a productive feedback) could remain anonymous but for moderators though it should be visible in order to have the possibility to take actions against abusive downvoters. 

The idea @Malkazoid mentioned (having only a text field to provide reasons) has a major disadvantage: it makes hard to see a bigger picture (statistics / analytics) of the feedbacks. A certain degree of standardization in the form of including a few pre-defined options for reasoning could help the authors to see the impact of certain aspects of his or her content. For example if people could choose from predefined options the platform could provide data like out of all downvotes how many where made due to "difficulty understanding the text" or any other reason. If reasoning would be possible only in the form of a text field you would have to read all the entries and then try to come up with a conclusion that by the way is prone to subjectivity and misinterpretation. 

 

@Garden Gnome Publications in this way how will moderators or anyone will be able to reduce the reputation of abusive downvoters? Cause let's say someone is on a downvoting rampage for no reason, authors will be able to contest the downvote but since the identity of the downvoter is not visible to anyone, no measure can be taken to prevent that phony user to abuse downvoting. 

Szabo posted:

@Garden Gnome Publications in this way how will moderators or anyone will be able to reduce the reputation of abusive downvoters? Cause let's say someone is on a downvoting rampage for no reason, authors will be able to contest the downvote but since the identity of the downvoter is not visible to anyone, no measure can be taken to prevent that phony user to abuse downvoting. 

Two possibilities.  First, knowing who the downvoter is, is not necessary in order to apply a penalty.  In fact, it might be better if the identity is not known, so that the moderator is not tempted to play favourites.

Second - the identity could conceivably be visible by moderators, and/or the Tribunal.  Personally I prefer anonymity.  I don't want the platform to go easy on me if a make an abusive downvote, just because I have a high reputation and have helped the platform since its early days.  The rules should apply to everyone, and anonymity helps that.

Szabo posted:

@Garden Gnome Publications in this way how will moderators or anyone will be able to reduce the reputation of abusive downvoters? Cause let's say someone is on a downvoting rampage for no reason, authors will be able to contest the downvote but since the identity of the downvoter is not visible to anyone, no measure can be taken to prevent that phony user to abuse downvoting. 

I don't understand the technical part of how this might work, but maybe creating a threshold? Imagine someone is abusing with downvoting that threshold will allow whoever is responsible to have access to the user identity?

Szabo posted:

The idea @Malkazoid mentioned (having only a text field to provide reasons) has a major disadvantage: it makes hard to see a bigger picture (statistics / analytics) of the feedbacks. A certain degree of standardization in the form of including a few pre-defined options for reasoning could help the authors to see the impact of certain aspects of his or her content. For example if people could choose from predefined options the platform could provide data like out of all downvotes how many where made due to "difficulty understanding the text" or any other reason. If reasoning would be possible only in the form of a text field you would have to read all the entries and then try to come up with a conclusion that by the way is prone to subjectivity and misinterpretation. 

 

This would be true, but I don't think the idea is to replace the pre-defined options (that are already there, and could even be expanded).  The idea is to have people enter a brief reason after checking the check box.

I suppose this may not have been said explicitly yet because folks are trying to tackle the issue of downvoting abuse primarily... but there is no reason why we should cause analytics to suffer while doing so.  I agree, the more data the system can collate to inform how to improve it, the better.

Hi all  - If the "disagree" button, the "poor quality" button" and the "violation" button were used properly, then a simple anonymous pop up to type in a reason for poor quality could be used - in fact might be the way to make sure the current downvoting set up is used properly.

Might be wrong but I think it's possible for an automated system to track votes to a profile without sacrificing anonymous voting. This information could used by the tribunal to sort out complaints and appeals - not in the sense that tribunal members have unlimited access to it, only that it can be accessed if sorting out a complaint requires it.

Makes better feedback for content contributors.

Also, moderators seeing posts with an unusually low score that doesn't seem warranted, could flag potential misuse.

Being able to type in viewable reason might reduce the number of posts being flagged as violations that can't be dealt with under current legal rules?

And now for the main plug - I think upvotes should at least be divided into high and medium, and the equivalent to the I disagree, added. 

Szabo posted:

@Vico Biscotti Providing an explanation for downvoting is like giving a productive feedback to the author. Just like in the field of education or businesses, a professor or a manager doesn't simply dislikes an assignment and leaves it like that, but also provides an explanation which improves future productivity and quality. Similarly here, on Narrative, it's in everyone's interest to increase the quality of the contents and to avoid unfair actions which can be achieved. For the authors the identity of the downvoter (and who also would have to provide a productive feedback) could remain anonymous but for moderators though it should be visible in order to have the possibility to take actions against abusive downvoters. 

The idea @Malkazoid mentioned (having only a text field to provide reasons) has a major disadvantage: it makes hard to see a bigger picture (statistics / analytics) of the feedbacks. A certain degree of standardization in the form of including a few pre-defined options for reasoning could help the authors to see the impact of certain aspects of his or her content. For example if people could choose from predefined options the platform could provide data like out of all downvotes how many where made due to "difficulty understanding the text" or any other reason. If reasoning would be possible only in the form of a text field you would have to read all the entries and then try to come up with a conclusion that by the way is prone to subjectivity and misinterpretation. 

 

Hi @Szabo I think that the analytic is still there - we still have to select the  "poor quailty button. The reason box is just added on. If predefined reasons can be worked out, this would be good for feedback , too and add some metrics - but none would be lost if just the text box was added, IMO.

@Szabo, I get your intention. Feedback is indeed useful, but I don't agree with forcing it in blogging. When you publish, you're offering your finished product. When they don't like, a generic downvote is already much more feedback than a published author usually gets. It's up to the author, to do his or her best, maybe with the help of an editor. Readers should be put in the most comfortable position.

Of course, it's completely different in a learning environment, at school or in classes. But it's a different context. If Narrative was an "educational" context, everything would be totally different.

I'm concerned about abusive downvotes too, but I think they should be mostly detected by algorithms, and they're not a significant percentage anyway. Abusive upvotes are MUCH more, for certain.

Vico Biscotti posted:

@Szabo, I get your intention. Feedback is indeed useful, but I don't agree with forcing it in blogging. When you publish, you're offering your finished product. When they don't like, a generic downvote is already much more feedback that a published author usually gets. It's up to the author, to do his or her best, maybe with the help of an editor. Readers should be put in the most comfortable position.

Of course, it's completely different in a learning environment, at school or in classes. But it's a different context. If Narrative was an "educational" context, everything would be totally different.

I'm concerned about abusive downvotes too, but I think they should be mostly detected by algorithms, and they're not a significant percentage anyway. Abusive upvotes are MUCH more, for certain.

I agree - abusive upvotes are a much bigger problem for Narrative's system.

But abusive downvotes, even though less frequent in my opinion, are also a threat to our community's well being and popularity.

Feedback is not something we have a right to expect, but we should consider making a simple, short note mandatory to remedy the abuse.  Just because something is not expected, does not mean that we could not have it, and benefit from it, if we decide that's what we want here?

If I consider something low quality enough to downvote it, I am willing to provide a few words to say why.  I'd actually really welcome the ability to do so.  I think the people who will be put off by this will mostly be the people who don't have a valid reason for downvoting.  If you do have  valid reason, you tend to want to express it, no?

Could be as simple as:

"There have been too many posts on the same subject in the past two weeks".

or

"Copy-paste"

or

"Superficial and poorly researched"

or

"Disorganized and incomplete"

or

"Unoriginal photo without any context: no story or reason why people should care"

----------------------------------------------

We could even have it so there is an option to say, via checkbox "I don't want to provide a reason".  But in that case, either the downvote has no effect, or perhaps if it is from a higher rep person, it could have a very small effect.  And only the downvotes with a reason given would have full weight.  These could even have a slightly stronger weight than the current downvotes do... since we would know they are MUCH more likely to be well-considered downvotes.  

This would go some distance to create more nuance than just 100% upvote, 100% downvote.  

I do agree with you that people are not owed an explanation, but if people have to give a short one, I do believe it would benefit the platform.

Vico Biscotti posted:

@Szabo, I get your intention. Feedback is indeed useful, but I don't agree with forcing it in blogging. When you publish, you're offering your finished product. When they don't like, a generic downvote is already much more feedback that a published author usually gets. It's up to the author, to do his or her best, maybe with the help of an editor. Readers should be put in the most comfortable position.

Of course, it's completely different in a learning environment, at school or in classes. But it's a different context. If Narrative was an "educational" context, everything would be totally different.

I'm concerned about abusive downvotes too, but I think they should be mostly detected by algorithms, and they're not a significant percentage anyway. Abusive upvotes are MUCH more, for certain.

HI @Vico Biscotti Since there's so much more here than just the standard blog post, I think there's a need for more specifics. Especially since the community is expected to curate to a large degree - and the current DV system - or at least how it's being used -  doesn't seem to be working as it was intended to. And same goes for the upvotes. 

There's also scale, I think, to consider. Right now, if some one gets 20 ups and one down (variable of course due to rep scores of voters) this generally isn't big deal. 

But if a comparable post is only getting 1 up - then gets 1 down - it just doesn't add up the same way. 

As volume grows the potential for large disparities in how posts are rewarded is evident - and I think this is what you see re: upvotes?

I also see it in downvotes, as they stand - maybe not as big an issue, but still a problematical one.

@Malkazoid, I'm not sure that an explicit reason would be a barrier for abusive downvotes. On the contrary, it would be for fair downvotes. It's easy to make up something, and - from what I saw - abusive downvoters are usually illogical actors. Also, they want to damage, and to go through the extra step, much more than fair readers. I got comments from trolls with "boring", "learn English", and so on. No need to write a poem, and they have a lot of that crap in mind. Unless we want to stop them - and algorithms could already do much - they just can't be stopped.

@Colleen Ryer, I'm not saying downvotes are not a problem at all, still they are a tool I wouldn't do without on this platform (remember that no banning or prevention is here...). Downvotes on small number have a weight, I know. But if I get a few upvotes, it's very likely that I have to improve my blogging strategy anyway, that I'm in a position where downvotes are a minor factor. Even if any downvote would automatically be changed into an upvote, that wouldn't make the difference. Having a downvote justified - or even removed - wouldn't change my blogging business at all, nor my growth as an author. Abuses should be fought, I'm just saying I don't see a mandatory comment as a tool for this, and also that it can't help your growth as an author (maybe the contrary!).

With respect does it matter what is a bigger problem?

We have a major problem with spam (upvote abuse) and downvotes. We need a change in the system that rewards people properly in a positive way that deals with both issues.

Vico Biscotti posted:

@Malkazoid, I'm not sure that an explicit reason would be a barrier for abusive downvotes. On the contrary, it would be for fair downvotes. It's easy to make up something, and - from what I saw - abusive downvoters are usually illogical actors. Also, they want to damage, and to go through the extra step, much more than fair readers. I got comments from trolls with "boring", "learn English", and so on. No need to write a poem, and they have a lot of that crap in mind. Unless we want to stop them - and algorithms could already do much - they just can't be stopped.

I think that's what makes this idea useful!

A comment that says "Boring" could be considered an abusive downvote - in fact it could be one of the responses detected automatically by the system, and given zero, or very little weight.  It has no value because a post of extremely high quality on astrophysics, might be boring to the average person - it remains an extremely high quality post!  The comment is how we can identify a downvote as abusive, and act accordingly so as to preserve the quality of the post, and penalize the abusive downvoter.

"Learn English" is an easy one for a moderator to determine.  If the post is completely in another language, then that would not be an abusive downvote (even though it would be inconsiderately worded, it would reflect on a reality that the English version of Narrative cannot have posts in other language rise to the top). 

However if the post is in English by a non-native speaker, the moderator could halve the weight of the downvote... to honor the fact that low language skills do in fact lower the reading experience for most people, but also honor that the post is of some value still, and can be understood with a bit of effort.  If it is felt that the non-native speaking aspect isn't really affecting the readability at all, with just a strange expression here or there, then the downvote could be considered abusive.  This is something very easy to determine by the moderator who originally approved the post.

 

@Vico Biscotti was looking more at economic ramifications for voting. If the commenting box is valuable as feedback, see it as a bonus, rather than the reason for being. Too, when people put spammy reasons in the box, then it's confirmed the DV is bogus, at least.

But if it won't stop abusive voting, then there's no point in it. Preset reasons for downvoting might be more useful - it's a matter of figuring out the best reasons to put in the list. 

Colleen Ryer posted:

But if it won't stop abusive voting, then there's no point in it. Preset reasons for downvoting might be more useful - it's a matter of figuring out the best reasons to put in the list. 

There is no reason to suspect it wouldn't deter abusive downvoting if there are penalties involved.

Penalties could ramp up from a small rep hit, to losing a percentage of earnings that month and blocked from posting and voting for increasing periods of time.

Malkazoid posted:
Colleen Ryer posted:

But if it won't stop abusive voting, then there's no point in it. Preset reasons for downvoting might be more useful - it's a matter of figuring out the best reasons to put in the list. 

There is no reason to suspect it wouldn't deter abusive downvoting if there are penalties involved.

Penalties could ramp up from a small rep hit, to losing a percentage of earnings that month and blocked from posting and voting for increasing periods of time.

Have thought all along that if there is downvoting, a way to monitor and correct abuse should be in place, whether there's a requirement to explain, or just tick boxes. Seems there's only a penalty for flagging AUP/TOS that isn't- and this because someone has to manually check.

I think to make the "poor quality" vote work it needs to be more specific.

@Malkazoid, the content of the comment could help automatic detection, but at the current level I don't see detection at all, not even in patterns. I wouldn't go at language detection level, if no previous level is there.

From the point of view of moderators, the moderator should moderate independently of down-comments. A downvote is warning enough to draw the attention to a post, in my opinion. The moderator will judge if another downvote, or not, is needed.

Nrve posted:

The problem with the reason to downvote suggestion is people will eloquently bs enough to justify their bad behavior. 

This may indeed happen. but eloquent BS:

a) requires work, which helps balance the symmetry of the face off between Narrative and bad actors

b) can usually be quite transparent to a moderator, especially if the system has flagged a user a being susceptible to downvoting certain people, due to a niche appeal, for instance

Do you see the work some trolls are willing to put in? It's more than honest users.

I hope so about b but I worry it will slip through the cracks and cause more fights.

Nrve posted:

Do you see the work some trolls are willing to put in? It's more than honest users.

True, some trolls are dedicated.  Other revenge downvoters, or even simply people downvoting for the wrong reasons, will be less willing to put in a lot of work inventing a BS reason, once they realise the real reason isn't acceptable.

I hope so about b but I worry it will slip through the cracks and cause more fights.

If we take the route proposed by @Garden Gnome Publications, and allow authors to appeal downvotes when they feel the reason given is not legit, then it would be up to the author to not let anything fall through the cracks...

The system could work and is not a bad idea but I gues I feel it's unnecessary. I know Vico and others do not like the clap system but it's simple and it works. An upvote tied to rep is all we need. Or a way to vote that doesn't give rewards. This site would be a lot more active and have a lot more traffic in the long run. The rep system was suppose to be our mechanism for upvote abuse. And we can add a flag for upvote abuse as well, I'm cool with that.

I'm not convinced we have a downvote problem. Inherent in this discussion is the assumption that one is entitled to upvotes. No one is entitled to upvotes. And the people I see complaining about downvotes are people with the most NRVE earned on the platform, which tells me they aren't getting a lot of downvotes. They're mostly upset because they've got a few downvotes and don't understand why. Therefore, they assume, there is someone abusing the downvote button. And that may not be the case.

I agree with @Vico Biscotti that upvote abuse is much more common. We've got people upvoting everything. And I've gone to thin content more often to see almost double-digit votes with the content at 100% much more than I've gone to a really good post and seen it downvoted.

I think the bigger problem here is that people don't understand how a posts quality score is arrived at. It's based on an aggregate of upvotes and downvotes with weight based on rep of the voter. If anything, the algorithm could be tweaked to change some of the weightedness of the votes. Or maybe we need to move away from the all-or-nothing upvote/downvote system toward a more granular approach where each voter assigns a quality rating from 0 to 100 with a text box for comments and/or a list of "issues" they can choose from to justify their rating. 

Personally, I trust readers to give their best judgment on quality. The aggregation of votes is supposed to act as a check and balance against downvote abuse. A more granular system could be more effective, especially as the platform grows and more people vote on the posts.

It's nothing to do with rewards but user experience. I mentioned before our activity rate halved when that troll was around. I don't care about the downvotes or what some random person thinks about my post I care that we have a system that is pleasant for users. People do not need negative tools on the internet, this has been proven on every site.

@Nrve, actually, the clapping system is not that bad in general. But the context here is different. On Medium you have a limited pool of rewards that you distribute among the authors you read. And you pay. And members can be blocked/banned.

Weighing users votes by their number (so, letting each user have a limited contribution) AND excluding low reps from contribution AND having patterns detection or reporting in place, I would be fine even with the absence of the downvote at all.

cc @Colleen Ryer

@Garden Gnome Publications I wonder that because the voting system isn't transparent, it's generating mistrust and misunderstanding? Because we can't really pinpoint if downvoting is abusive or not, doesn't help. And although we can see that upvoting is really over done in places - we don't dare tackle it - too many backs would be up in a flash. 

If the whole voting system was more transparent, with some more flexibilty as you suggest -rating system rather than what it is now, I think it would serve better over time - easier to understand because it's easier to see how actions connect with outcomes. Less trusting required, more objective evidence. 

Yes @Colleen Ryer, there is an inherent mistrust in human relationships. This is why there is so much talk in the blockchain world about trustless, permissionless transactions. Decentralization is supposed to solve the problems inherent in centralized systems where the central authority is biased. I'm skeptical, but I understand the arguments.

What we have here is a system that is working the way it was intended, but because upvotes and downvotes are anonymous, when people get downvotes and don't understand why, they assume someone is abusing the system. Maybe they are, maybe they're not. We can't say because we aren't collecting any data that can help us figure that out.

So what's the solution? Collect the data. It's not like it isn't possible. Narrative Company has the tools to see who is voting for whom, and to detect voting patterns. If we implement some transparent measures that give the entire community visibility on why people are voting the way they are voting, then we can make a determination as to whether abusive downvotes are happening or not. And if they are, then we have the data to make the tweaks necessary to fix the problem, or encourage people to be more honest and transparent in their voting habits. But without the data to back it up, all we're doing is making assumptions.

I'm making a vote, and you can quote me on this, at the end of the day I do not care why ppl are using the downvote I care about the user experience and it would be a lot more pleasant without a downvote option. Replace it with a flag for problem behavior and a system that encourages people to make better content with carrots instead of sticks.

I do think we have a small abuse problem now and it will be a lot bigger in the future. But even if everyone is using it properly (which I can say with 100% certainly they will not) we are still better off without it.

Vico Biscotti posted:

@Nrve, actually, the clapping system is not that bad in general. But the context here is different. On Medium you have a limited pool of rewards that you distribute among the authors you read. And you pay. And members can be blocked/banned.

Weighing users votes by their number (so, letting each user have a limited contribution) AND excluding low reps from contribution AND having patterns detection or reporting in place, I would be fine even with the absence of the downvote at all.

cc @Colleen Ryer

@Vico Biscotti how would this work for curation? -or maybe I just need lunch to see it 

Yes it will be abused as well but a flag is a more serious offense of something you should not be doing on this site like spamming or stealing work. And when a user abuses them it will really make them look bad when questioned. But a downvote is a way for people to get away with making people feel bad about their work for whatever reason and then say well that's my opinion or it's your fault, even if that person who voted doesn't know what they are talking about. And even if they do it is elitest to say you are not good enough for this site.

To put it another way:

Flags are objective complaints which we need

Downvotes are subjective complaints which we do not need.

Everyone has an opinion on what quality is but not everyone wants to hear it.

Garden Gnome Publications posted:

Yes @Colleen Ryer, there is an inherent mistrust in human relationships. This is why there is so much talk in the blockchain world about trustless, permissionless transactions. Decentralization is supposed to solve the problems inherent in centralized systems where the central authority is biased. I'm skeptical, but I understand the arguments.

What we have here is a system that is working the way it was intended, but because upvotes and downvotes are anonymous, when people get downvotes and don't understand why, they assume someone is abusing the system. Maybe they are, maybe they're not. We can't say because we aren't collecting any data that can help us figure that out.

So what's the solution? Collect the data. It's not like it isn't possible. Narrative Company has the tools to see who is voting for whom, and to detect voting patterns. If we implement some transparent measures that give the entire community visibility on why people are voting the way they are voting, then we can make a determination as to whether abusive downvotes are happening or not. And if they are, then we have the data to make the tweaks necessary to fix the problem, or encourage people to be more honest and transparent in their voting habits. But without the data to back it up, all we're doing is making assumptions.

I too am skeptical re: decentralization. 

Not sure what to think where obviously plagiarized work is on display and voting system isn't taking care of it - unless there are glitches in process of being fixed.

If collecting the data and more transparency will level things out, this is probably easier than changing the current system - and agree assumptions aren't enough to come to conclusions about what is and what isn't the matter ... 

Vico Biscotti posted:

Hmm... @Colleen Ryer, I'm not sure to have understood...

 @Vico Biscotti Weighing users votes by their number (so, letting each user have a limited contribution) AND excluding low reps from contribution AND having patterns detection or reporting in place, I would be fine even with the absence of the downvote at all.

The more a person curates, the lower the weight of their vote?

oh, and Low reps are already limited in voting power.

If I may add @Colleen Ryer I think we would be better off with a system that doesn't need to be curated.  A site where good content goes to the top and people read and like what they like because they like it. Look how bad steemit is because of a need for curation. 

Yes this is a fantasy of sorts and we will need flags or something but I dont want a system where people feel they have to clean the site. That's what algorithms are for.

 

Colleen Ryer posted:
 
 

I too am skeptical re: decentralization. 

Not sure what to think where obviously plagiarized work is on display and voting system isn't taking care of it - unless there are glitches in process of being fixed.

If collecting the data and more transparency will level things out, this is probably easier than changing the current system - and agree assumptions aren't enough to come to conclusions about what is and what isn't the matter ... 

I'm in full agreement.

It may make more sense to try improving the current system with more data collection and transparency, than to strike it and run with something else immediately.  There's a good chance the measures proposed in this thread would take the wind out of the sails of people who complain about downvotes, as well as reducing their abuse.  And I think everyone here agrees that algorithmic measures and automation should be there too - downvote comments are not meant to be a replacement for those.

I am seriously skeptical of decentralisation as a primary goal, and very critical of sacrificing quality and efficiency upon its altar... at the end of the day - it is a minority of people who care more about decentralisation than they do about quality, so the Team should be really clear what the right choice is whenever those two goals go head to head.  I say, decentralise as much as possible without sacrificing any quality, but no further.

@Malkazoid

Well said.

However my opinion is that people care most about a positive supportive experience. Then quality. Or you can put them equal.

We are focused too much on what readers want and are not focusing on what content creators want which are the most important group is at this time.

I'm not saying saying strike the system down, just add a double vote and change the downvote to a flag. It's the same system just changed to make for a better experience.

Nrve posted:

@Malkazoid

Well said.

However my opinion is that people care most about a positive supportive experience. Then quality. Or you can put them equal.

We are focused too much on what readers want and are not focusing on what content creators want which are the most important group is at this time.

I'm not saying saying strike the system down, just add a double vote and change the downvote to a flag. It's the same system just changed to make for a better experience.

I think readers are just as important as content creators.  The more readers, the more interaction with content, and so the more fulfilled content creators feel.  The more engagement content gets, the more content creators will want to keep posting here, and the more content creators will want to join.

I would like to agree with removing the downvote: I know it would solve some user satisfaction issues, for some people.

I think my resistance is based on two things.

1) As a matter of general principle, I like to help people keep their feet on the ground.  In the real world, some people don't like your stuff.  Every creative has to learn to cope with that fact.  If people are too precious about this, I recommend they only share their creations with their close friends.  Narrative does not have to be for everyone.

2) When someone is faced with the possibility of only upvoting, doing nothing, or flagging content as a violation of some sort, I think we will have less contrasted results separating out our content.  Those who think the quality is actually bad, but falls short of a violation, will not have a voice. They will not vote.   This will invariably lead to bad quality content not being separated as much from average or high quality content.

 

Good points but at the same time:

1.) This isn't the real-world but a site we are asking people to spend their free time on and give up their work on when there are other options.

2.) I think people who get downvoted and feel it's unfair will just post low quick quality to offset loses or cut and paste their work as they feel they are going to get downvoted anyways. There is no incentive to work hard. People need reasons to work hard and be better. The low quality downvote makes them work just hard enough to get by and get rewards. This is exactly what you see in steemit for two years.

Nrve posted:

Good points but at the same time:

1.) This isn't the real-world but a site we are asking people to spend their free time on and give up their work on when there are other options.

I view this as part of the real world - it isn't a MMORPG.  It is a place for content creators to earn real money.  I feel like the same rules should apply here as the rules that apply anywhere you're getting paid for your work.

2.) I think people who get downvoted and feel it's unfair will just post low quick quality to offset loses or cut and paste their work as they feel they are going to get downvoted anyways. There is no incentive to work hard. People need reasons to work hard and be better. The low quality downvote makes them work just hard enough to get by and get rewards. This is exactly what you see in steemit for two years.

Maybe.  But I think Steemit simply never had a paradigm encouraging quality.  In fact its very architecture choices and policies meant it was always going to be more profitable to invest in upvote bots than it was going to be to invest time and skill in adding a little more quality.

I don't think many quality contributors will suddenly become something else, because of a few downvotes that do nothing to reduce the value proposition of posting on Narrative.  

Cut and pastes and quick low quality posts should, if the system is working, end up being less profitable than higher quality posts.  That's enough incentive for people capable of higher quality, to deliver it.  If the system is not working in such a way as to make that happen, then I think that's what we have to focus on fixing...

1.) This is a side point that I will not try to convince you here but I think the MMORPG is an apt metaphor as crypto lets us earn in new ways and we should embrace that not shun it. It's going to be the future and what people want. But regardless. The point is we are NOT paying people until people buy the coin. We cannot treat this as a paying publication until nrve is worth a steady dollar or two. And I feel like we are treating it like it is. We need to get people and traffic first. Anybody can print a coin nowadays it doesn't make it valuable until you give it worth.

2.) You are not wrong about steemit lacking a quality algorithm but that's one point of concern. I'm saying we can make this both a positive place and encourage quality. And I disagree about the downvotes. A lot of people (especially new ones who have yet to make money) will be really turned off by getting downvoted. And another big issue with steemit is people posting a long quality article on the site, getting no upvotes and leaving. Now imagine a new person working hard and getting a downvote.

Nrve posted:

1.) This is a side point that I will not try to convince you here but I think the MMORPG is an apt metaphor as crypto lets us earn in new ways and we should embrace that not shun it. It's going to be the future and what people want. But regardless. The point is we are NOT paying people until people buy the coin. We cannot treat this as a paying publication until nrve is worth a steady dollar or two. And I feel like we are treating it like it is. We need to get people and traffic first. Anybody can print a coin nowadays it doesn't make it valuable until you give it worth.

To me, it doesn't matter how much NRVE is worth, it matters what the USD equivalent of your earnings is.  And that equivalent represents a good value proposition already.

The fact that the value of NRVE is low only adds an investment dimension to that offer.

Either people are convinced by this value proposition or they aren't...  If they are, a downvote causing their post to have 98% quality instead of 100% won't change much.  If they aren't, they won't be here in the first place...

I agree downvotes are unpleasant, but every work experience has some negatives.

2.) You are not wrong about steemit lacking a quality algorithm but that's one point of concern. I'm saying we can make this both a positive place and encourage quality. And I disagree about the downvotes. A lot of people (especially new ones who have yet to make money) will be really turned off by getting downvoted. And another big issue with steemit is people posting a long quality article on the site, getting no upvotes and leaving. Now imagine a new person working hard and getting a downvote.

I completely agree it would be a more positive experience on the surface, with no downvote.  But if that causes low and high quality to be less contrastable, it will result in payouts where low quality earns more, and high quality earns proportionately less.  That seems like a bad trade off to me.  If I'm creating good quality content, I'd rather get a downvote but earn a few bucks more, then never get any downvotes, but earn less.

BTW - this might feel like I'm super opposed to removing the downvote.  I'm not - not really.  These are just the facts as I see them.  If tomorrow the team removed the downvote and replaced it with something else, I'd just keep on keeping on.

I agree about the USD equivalent but that can only be achieved by a high price coin or a popular site. We cannot expect people to keep buying niches and increasing the reward pool without this site being popular. However the monthly minted coin reward pool is consistent and if that is worth a million or ten million USD that can go a long way for the content creators and the Narrative team.

I'd rather more people use this site (by having a more positive experience) and bring traffic on which will lead to a higher value coin. And I think a double vote and rep would be better at making content less contrastable without the down side of the downvote. It looks like we reached an impasse on point 2.

Nrve posted:

I agree about the USD equivalent but that can only be achieved by a high price coin or a popular site. We cannot expect people to keep buying niches and increasing the reward pool without this site being popular. However the monthly minted coin reward pool is consistent and if that is worth a million or ten million USD that can go a long way for the content creators and the Narrative team.

I'd rather more people use this site (by having a more positive experience) and bring traffic on which will lead to a higher value coin. And I think a double vote and rep would be better at making content less contrastable without the down side of the downvote. It looks like we reached an impasse on point 2.

Maybe the main question is: how much are downvotes really affecting people?

The team probably have the data they need to figure this out.  So I'd invite them to analyse the data and only switch paradigms if they find people are actually leaving or posting worse quality after receiving downvotes...

If we leave it at that, then you and I don't have to worry about it!

I'm seriously having trouble understanding your way of thinking on several fronts @Nrve. I've been writing for over 30 years. Sometimes I got published, sometimes I didn't. Sometimes I got paid, sometimes I didn't. Sometimes I got a paid a lot and sometimes just a little. I've never measured my worth as writer by my experience. I've always tried to find ways to learn how to get better at it. 

Improving oneself is no guarantee of success, but one sure way to guarantee no success is to place one's happiness and self-worth on others. 

This is a character issue. To claim that people will post low quality content in retaliation for downvotes is simply silly. Some will. But those people deserve more downvotes. The ones who deserve upvotes are the ones who don't whine about the downvotes and just put the bullet to the bone. Writers write. If they're good, they'll get paid. 

Vico Biscotti posted:

Exactly what I'm saying, @Colleen Ryer. Crazy clicking having no more total weight than a single click. According reputation, of course. Maybe a refined system would be a total weight like the weight table used for voting in niche suggestions.

@Vico Biscotti so no curating then. I wasn't talking about crazy clicking.

Fair enough. I care about how many people have left because of it. I feel like a lot of people are not expressing a downvote problem cause they left and those that stayed care more about quality. I could be wrong but I really noticed the drop the week after the troll. Maybe it's a coincidence but I think that negative experience turned off a lot of people.

Malkazoid posted:
Nrve posted:

1.) This is a side point that I will not try to convince you here but I think the MMORPG is an apt metaphor as crypto lets us earn in new ways and we should embrace that not shun it. It's going to be the future and what people want. But regardless. The point is we are NOT paying people until people buy the coin. We cannot treat this as a paying publication until nrve is worth a steady dollar or two. And I feel like we are treating it like it is. We need to get people and traffic first. Anybody can print a coin nowadays it doesn't make it valuable until you give it worth.

To me, it doesn't matter how much NRVE is worth, it matters what the USD equivalent of your earnings is.  And that equivalent represents a good value proposition already.

The fact that the value of NRVE is low only adds an investment dimension to that offer.

Either people are convinced by this value proposition or they aren't...  If they are, a downvote causing their post to have 98% quality instead of 100% won't change much.  If they aren't, they won't be here in the first place...

I agree downvotes are unpleasant, but every work experience has some negatives.

2.) You are not wrong about steemit lacking a quality algorithm but that's one point of concern. I'm saying we can make this both a positive place and encourage quality. And I disagree about the downvotes. A lot of people (especially new ones who have yet to make money) will be really turned off by getting downvoted. And another big issue with steemit is people posting a long quality article on the site, getting no upvotes and leaving. Now imagine a new person working hard and getting a downvote.

I completely agree it would be a more positive experience on the surface, with no downvote.  But if that causes low and high quality to be less contrastable, it will result in payouts where low quality earns more, and high quality earns proportionately less.  That seems like a bad trade off to me.  If I'm creating good quality content, I'd rather get a downvote but earn a few bucks more, then never get any downvotes, but earn less.

BTW - this might feel like I'm super opposed to removing the downvote.  I'm not - not really.  These are just the facts as I see them.  If tomorrow the team removed the downvote and replaced it with something else, I'd just keep on keeping on.

personally, I don't see downvotes as a minus - if I know what they're for and they're legit. But I can't tell at this point. And I think a graduated vote would work just as well - if it's transparent, as well. So which ever way works out the best, Since the present system is yet to be made transparent - can't really say it should be replaced. 

But this thread is about downvotes - and the graduated system I think would correct certain things that are obviously out of whack in the upvote department -without using downvotes that might really upset people, the way things are set up right now.  And I'm getting dizzy switching between the two main issues/suggestions 

@Garden Gnome Publications

I think our disagreements come from how we see the site. To me it's not a place where writers get published or improve themselves (the publishing world is already that). To me it's a place where writers and other content creators can make a living wage while working on their next project. 

We are treating it like the New Yorker when the world has that. What the writing world doesn't have (or other types of content) is a place to make a living wage that is still in their field. I want content creators to work here instead of a part time job while they try to make it in their field.

Content is a feast or fathom world and we can fix that.

 

Colleen Ryer posted:
Vico Biscotti posted:

Exactly what I'm saying, @Colleen Ryer. Crazy clicking having no more total weight than a single click. According reputation, of course. Maybe a refined system would be a total weight like the weight table used for voting in niche suggestions.

@Vico Biscotti so no curating then. I wasn't talking about crazy clicking.

No, no. Misunderstanding here. I mentioned "crazy clicking" just to explain the same amount of total distributed "appreciation" weight independently of the number of upvotes/claps, related to rep.

Colleen Ryer posted:

And I think a graduated vote would work just as well - if it's transparent, as well. So which ever way works out the best, Since the present system is yet to be made transparent - can't really say it should be replaced .

Not sure it would work just as well, as explained in a previous post.

If all the choices people have are degrees of upvoting, the intent of assigning low quality value to something is lost.  Those people who want to say - this is actually bad, will have no option but to not vote.  That's a loss of representation that will be missing from the results, resulting in less gradations in the data, and NRVE being assigned less precisely as a result.

Yes, the graduated upvote will create more gradation in the upper tiers but that cannot replace the loss of gradation in the lower tiers.

But this thread is about downvotes - and the graduated system I think would correct certain things that are obviously out of whack in the upvote department -without using downvotes that might really upset people, the way things are set up right now.  And I'm getting dizzy switching between the two main issues/suggestions 

I can't blame you - I'm getting dizzy too.

Okay @Nrve, that makes sense. I see where you're coming from now. 

I don't think we'll fix the feast or famine problem for creatives. Narrative, if successful, will increase the competition. Increased competition means creatives will have to stand out more in order to earn. That's not bad. It's actually good. Competition forces us to better ourselves. Some will survive and some won't. That's the way the world works, and always has.

I don't see any reason it can't be both The New Yorker and a place to practice. For a writer like me, I might be able to finagle a byline in The New Yorker. And I'd gain good coin if I did. But I'd have to chase another byline. And then another. And then another. And so on, until I die. What Narrative offers me is the hope of earning full-time income writing about what I want to write about without chasing endless bylines. And because the economics are different, that possibility is predicated on my earnings potentially increasing AFTER I've earned them. It's not just trading bylines and hours for dollars. It's also a retirement program.

That may be wishful thinking, but it's a dream that is wrapped up in opportunity. And there's no reason that other writers can't use it as a place to play around while they pursue other projects. They might even find that it turns into their full-time gig. Wouldn't that be nice?

Nrve posted:

If I may add @Colleen Ryer I think we would be better off with a system that doesn't need to be curated.  A site where good content goes to the top and people read and like what they like because they like it. Look how bad steemit is because of a need for curation. 

Yes this is a fantasy of sorts and we will need flags or something but I dont want a system where people feel they have to clean the site. That's what algorithms are for.

 

Who or what decides which content gets to the top? automation can only go so far - and leaves so much to be desired where it's used already.

@Garden Gnome Publications

It would be great. We are not far apart in what we want for this site. 

My only argument comes from how we get there. I agree fully about the competition idea. And I see it as people will compete with each other, get better and quality content will rise to the top. But in order to get there we need a lot of users and a lot of content creators who are having fun here before the competiton ramps up. And I know you and others don't want to lose quality as we get there I just think we can get there in a way that doesn't include the downvote. It will cost us too many potentially great content creators who might be shy or thin-skinned but are very talented and could use a positive push. 

Malkazoid posted:
Colleen Ryer posted:

And I think a graduated vote would work just as well - if it's transparent, as well. So which ever way works out the best, Since the present system is yet to be made transparent - can't really say it should be replaced .

Not sure it would work just as well, as explained in a previous post.

If all the choices people have are degrees of upvoting, the intent of assigning low quality value to something is lost.  Those people who want to say - this is actually bad, will have no option but to not vote.  That's a loss of representation that will be missing from the results, resulting in less gradations in the data, and NRVE being assigned less precisely as a result.

Yes, the graduated upvote will create more gradation in the upper tiers but that cannot replace the loss of gradation in the lower tiers.

But this thread is about downvotes - and the graduated system I think would correct certain things that are obviously out of whack in the upvote department -without using downvotes that might really upset people, the way things are set up right now.  And I'm getting dizzy switching between the two main issues/suggestions 

I can't blame you - I'm getting dizzy too.

@malkazoid guess we're coming at it from different angles. right now there are times I want to vote - but I don't vote at all  because I'm pretty sure it's going to have a very negative effect -

If I could make it a vote for a reduced -or even no - reward - but it's sent as an upvote - I think people take it better - especially if the reasons for the vote are preset - easy to understand

And there are times I don't upvote when i want to - and if I could use a reduced value vote - I would. A post has accumulated a lot of what i suspect are loyalty votes - I appreciate the post, but I don't agree with the number of votes it already has, and don't want to add to the problem.

A tiered upvote adds options I don't have now - 

Nrve posted:

If I may add @Colleen Ryer I think we would be better off with a system that doesn't need to be curated.  A site where good content goes to the top and people read and like what they like because they like it. Look how bad steemit is because of a need for curation. 

Yes this is a fantasy of sorts and we will need flags or something but I dont want a system where people feel they have to clean the site. That's what algorithms are for.

 

Who or what decides which content gets to the top? automation can only go so far - and leaves so much to be desired where it's used already.

Nrve posted:

@Colleen Ryer

That's why I suggested tier voting. The more tier upvotes the higher the post goes. And the low stuff naturally goes down 

I would like it too, and the smiley no reward but thanks for the post option you mentioned below. But I agree with @Garden Gnome Publications that the present system isn't fully tested - adding transparency and collecting data to improve it might work too. So in order of preference, I prefer what you've suggested. In order of what's likely most practical for the company, fixing up the present system is probably more doable. MY WAG anyway.

@Colleen Ryer

Fair enough. Im just adding one vote as a content creator that I am not enjoying my experience here (outside of rewards and some cool people here). If it wasn't for the rewards and potential I would not be here posting work with my time and I would rather do freelance or deal with publications. And they can be quite unpleasant

Colleen Ryer posted:

@malkazoid guess we're coming at it from different angles. right now there are times I want to vote - but I don't vote at all  because I'm pretty sure it's going to have a very negative effect -

If I could make it a vote for a reduced -or even no - reward - but it's sent as an upvote - I think people take it better - especially if the reasons for the vote are preset - easy to understand

And there are times I don't upvote when i want to - and if I could use a reduced value vote - I would. A post has accumulated a lot of what i suspect are loyalty votes - I appreciate the post, but I don't agree with the number of votes it already has, and don't want to add to the problem.

A tiered upvote adds options I don't have now - 

Aha!  I'm seeing things through your eyes now.

That makes sense.

I think you've presented a good argument for adding a graduated upvote, but not really for removing the downvote.

It would be weird to have people voting specifically to say they don't want a post rewarded... it could be even more negative than saying you feel something is low quality.  There's a more personal quality to saying: "I don't want the writer to receive rewards for this", than to saying "I found this to be low quality", even if the latter then entails the former.

Malkazoid posted:
Colleen Ryer posted:

@malkazoid guess we're coming at it from different angles. right now there are times I want to vote - but I don't vote at all  because I'm pretty sure it's going to have a very negative effect -

If I could make it a vote for a reduced -or even no - reward - but it's sent as an upvote - I think people take it better - especially if the reasons for the vote are preset - easy to understand

And there are times I don't upvote when i want to - and if I could use a reduced value vote - I would. A post has accumulated a lot of what i suspect are loyalty votes - I appreciate the post, but I don't agree with the number of votes it already has, and don't want to add to the problem.

A tiered upvote adds options I don't have now - 

Aha!  I'm seeing things through your eyes now.

That makes sense.

I think you've presented a good argument for adding a graduated upvote, but not really for removing the downvote.

It would be weird to have people voting specifically to say they don't want a post rewarded... it could be even more negative than saying you feel something is low quality.  There's a more personal quality to saying: "I don't want the writer to receive rewards for this", than to saying "I found this to be low quality", even if the latter then entails the former.

 @Malkazoid the wording does matter - I was just sayin' off the cuff. Needs improvement, or practice makes perfect .... sounds a lot better than poor quality  -even if they all mean low or no rewards... and too, can't please everyone all the time ... 

plus I think tiered also would leave the "no vote" as an more of an indicator that the post wasn't seen ...

@Malkazoid

Maybe it just comes down to the wording and optics. We wouldn't word it as nice try but not good for rewards but more so thanks for posting on the site. If you keep working hard you can get this super upvote and make money.  Someone says my post was low quality is an insult I might take badly (and yes maybe the person is too sensitive but some ppl are). But being told thanks for posting and knowing that if I do better I can make money will motivate me to work harder.

 

Hi @Vico Biscotti nesting isn't working properly for me - I'm having a hard time understanding what you mean by

"No, no. Misunderstanding here. I mentioned "crazy clicking" just to explain the same amount of total distributed "appreciation" weight independently of the number of upvotes/claps, related to rep."

does this water down the upvotes as number of votes goes up?  how does it relate to rep?

 

I can see where some of wording and symbology is a little sour. I see it in particular with punishments for certain actions. Some of them are harsh. Perhaps we can reward content creators with pie. If I like your post a little bit, I give it a piece of pie. If I like it a lot, I give it half a pie. I really totally love it, I give it a whole pie. If you violate AUP or plagiarize, then I just throw the pie in your face. 

This is very important since it could save the community of a lot of stress that could be avoided. All some users are in for is to downvote a content. This will enable appeal for downvotes to be made as well.

Okay, @Colleen Ryer, I try to recap.

First off, I LIKE downvotes. But I've problems with the lack of graduality in upvotes, because cheap content is excessively rewarded. AND people seem to take the downvoting badly. AND voting circles are there. So, I mention one possibility - similar to Medium clapping - in which I would be favorable to the removing of downvotes, having at the same time a couple of advantages.

Imagine giving a user a certain number of "points" per month. These points are proportional to reputation (amplifying high reps, and disempowering low reps). So, let's say I have 100 points per month.

Now, let's say that for each post you can give 0 to 10 "ups".

If at the end of the month I gave a total of 1,000 ups, each of those ups will add 0.1 to the "points" of the post. If I gave only 100 ups, each of those ups will add 1. If I gave 1,000,000 ups, each up is worth 0,0001.

All the rewards will be distributed weighted by the earned points.

Advantages:

  • You can express a different level of appreciation of different posts, thus rewarding them differently and letting perceived quality emerge.
  • Upvoting everything makes sense only up to a point. The more you vote, the less your vote is worth.
  • You can't have more influence than your peers. The crazy upvoter has the same weight of who takes the time to read a few long posts.
  • No or low ups would be equivalent to the downvote, not in the meaning but at least in the effects (cc @Malkazoid).
  • No more talking of downvotes.

BUT, this should be coupled with the detection of hacking. By itself, the system is still weak (even if less than the actual voting system). In particular:

  • Blatant multiaccounts should be detected and penalized.
  • Low earned points, for a post, should equal to zero. In the given example, let's say a threshold of 200 points could exclude the impact of a single or a couple of undetected multiaccounts or things under the radar.
  • Patterns should be detected. A member voting only one other member, for example, should have a limited impact. Even if some patterns are not 100% evil, prudence should draw the line.
  • Reporting of violations should remain available.

That's all!

PS: I guess a suggestion or a post could be worth, here...

PS2: the suggestion is now there: https://community.narrative.or...native-voting-system

PS3: the post is also there: https://www.narrative.org/post...native-voting-system

MALAY BANERJEE posted:

Yes if anyone downvotes any post, he/she should make a comment to justify downvote.  Recently two of my post downvoted by someone without any reason.

Look at my post.

https://www.narrative.org/post...-are-replacing-human

https://www.narrative.org/post/story-of-a-refugee

Please make it mandatory to mention a reason in the comments section if someone downvote any post.

Thanks everyone.

 

I checked out these links and promptly up-voted both. They are solid posts and I'm glad you're contributing on this platform.

Add Reply

×
×
×
×