The blockchain was upgraded earlier today. You may experience trouble posting and transacting while the new bandwidth system stabilizes. Read more here.

Your FEEDBACK matters. Downvote/Flag improvement - A proposal.

in steemit •  2 years ago

A proposal of feature improvement:
A FLAG /DOWNVOTE should contain a justification.


Downvot

Motivation:
Provide a valuable feedback about the downvoted /flagged content both to it's author and the rest of the Steemit community.

Backstory

I found myself in a strange and difficult position lately: I wanted to do something between downvoting and flagging a comment. Details do not matter much at this stage of the story. It is not about being right or wrong here, but the sole act of using the flag button. It seems to be easy: one click and it is done, right? Okay, two. You have to confirm it. Still, sounds like an easy thing to do. Maybe, even too easy?

After several years in the IT industry, code interviews, interviews, discussions with colleagues, I think I have understood quite well the power of feedback. Flagging gives no information, so I left a comment to the author together with a flag. I wrote that I don't want to discuss it in details in the comment section (not to start an off-topic), but if he felt treated unfair, I would like him to catch me on the Steemit Chat so we can talk about it. It didn't go exactly as I would imagine it at first, but we discussed it with @jtstreetman and solved. Overall, the story ended happily, no harm was done. I would even say there was a positive outcome from it. This situation got me, however, thinking about the real problem.

The issue

Steemit seems to struggle a bit with the confusion over the small flag button in the top right corner. There are no rules how to use it. It is both a downvote (a subjective vote) and a flag (reporting a form of abuse). What is more important, while such vote has significant consequences, the platform provides poor to none support to put a meaningful feedback together with it.

IT CAN BE IMPROVED!

A bit about voting

The voting system consist of three options. Let's summarize them briefly:

Upvoting

An upvote is a fairly easy one. It represents a match between user's voting criteria and 'tested' content. What kind of criteria? There are many, like an interest in a topic, sympathy to the author, identification with the presented thesis or idea, a high probability of a reward for voting, etc. Overall, a justification for this choice is that an upvoter and/or the platform benefit(s) from the vote.

A lack of a vote

I assume, it can be described by two simple negations:

  • I am NOT aware this content is posted.
  • I do NOT want to pay for it.

Both conditions can have multiple reasons, but it is not the subject of this post. It should be, however, an important observation for users, that the lack of votes can be sometimes even a stronger indicator of the issue than downvotes.

Downvote/Flag

It is a difficult one, especially because two completely different features are mashed into one button.
A flag means a very low value of the content for the platform, statements in the conflict with fundamental rules provided by platform creators, e.g. plagiarism or spam in case of Steemit. A downvote reflects personal preferences and can be biased by user's emotions.

Searching Steemit back and forward in order to get more insight about thoughts and opinions of users proved me right about the existence of the problem. It is a serious one. The acceptance of @blakemiles84's suggestion could partially solve it. If you do not like it, just ignore it. It would also match with a suggestion found in a short discussion on Github.

Flagging is for:

  • Fraud or Plagiarism
  • Hate Speech or Internet Trolling
  • Intentional miss-categorized content or Spam

For a difference of opinion or offensive material please use the comment section and seek support from your peers.

Unfortunately, it seems not to represent the reality. I have found posts talking about whales using flags to downvote the content which they don't like, completely muting smaller users. It is the result of the power of reputation. A flag not only can remove a post or a comment from the public view, it affects the account and future work of the author. I was totally surprised when I saw that even Dan does it - He can make users vanish. On the bright side, I have seen cases where whales actually saved flagged users. Overall, a flag is used as a downvote, no doubts. What caught my attention the most, was a lack of the crucial part - a feedback.

People benefit from feedback, not flags.

During several years in the industry I have observed how a feedback or its lack affects people, products, companies, etc. It is not always easy to give a valuable feedback, but it definitely helps people to understand motives and improve.

There is a clear reason why all social media platforms allow users to like, promote, share, +1, upvote a content - positive emotions. If a user can see 30, 64, 232 upvotes, it makes him/her at least a bit happy. It boosts one's self-esteem and potentially influences further actions positively. Even in such cases there is a lot of room for a feedback, but it usually appears (Steemit users do actually a fairly good job in this area, I have to admit it).

On the other hand it had been already seen in the past that the idea of a "dislike" can create a lot of controversy. Such feature may be not the best idea at the early stage of any project, before users show an adequate level of maturity (you don't give kids knifes to play with). While a silence and a lack of interest can be a good punishment already, a flag without feedback can make one's day really bad. Downvoting seems to make sense on platforms like StackOverflow where posts have a Q&A format. Even there people from different social and ethnic group can feel offended by negative votes, but rules are quite clear and limit abuses. The story is less colorful e.g. on Reddit. Despite 'reddiquette', users often downvote posts on the basis of loosely defined terms. I believe Steemit should step up and not follow Reddit in this approach. It is not only about one's ego anymore but real profits from the reputation and written content.

How to address the problem?

There are simple options available already: comments, posts, DM on a chat. However, the best one would be a modification of the feature itself. Below, several possible options.

Flag Button

This button is used used to downvote or flag the content. It is the approach used currently. Both reasons result in the same consequences. I have to assume it is done this way on purpose.
Pros:
  • Requires no development.
Cons:
  • No distinction between flagging and downvoting.
  • Makes downvoting an easy action done under the influence of emotions, without thinking twice about it.
  • Does not show the reason of the action, which not only provides no feedback to author but also supports like downvoting a specific content or author for personal profits.
Use flagging and leave a comment

It is also possible now, but unfortunately, it is used that way rarely.
Pros:
  • Requires no development.
  • Gives feedback.
Cons:
  • It can be very difficult to find a comment related to a downvote among others.
  • It looks odd if/after the commented/downvoted content was edited.
  • It may lead to rewarding (upvoting) or bashing (downvoting) downvoters and have strange results: it is a key to the hell of weird human emotions, flame wars and can bring new wave of bots.
Change the flag confirmation (simplified)

Make the confirmation box contain checkbox with several predefined categories like: Fraud, Plagiarism, Hate Speech, Internet Trolling, Intentional miss-categorized content, Spam, Downvote. Make it mandatory. If the reason is a biased downvote, call it that way.

Pros:

  • A simple mechanism to provide a simplified form of feedback, visible to the author and other users.
  • It can be tracked easily, which provides a way to monitor all kinds of abuses and expose inappropriate behaviors of users.
Cons:
  • Requres development both on the front- and back- end.
  • May occur an important category is missing, like *false statements*, useful e.g. for scientific articles.
  • May not be enough, even if we skip a 'downvote' without a comment, cases like a potential plagiarism would benefit from a justification (for example a link to an original source attached).
Redesign the feature or change the flag confirmation (suggested solution)

Proposed way is to add both:
  • a mandatory text field to provide a justification
  • a mandatory checkbox (like e.g. in the previously discussed option) but with additional check "Other"
Make it visible to users on Steemit (not only in the blockchain).
Pros:
  • Gives feedback fully understandable to the author and others.
  • It is clean. A flag like and upvote can be removed if the voter decides that way. It does not start unnecessary discussions in the comment section.
  • It can be tracked easily, which provides a way to monitor all kinds of abuses and expose inappropriate behaviors.
Cons:
  • Requres development both on the front- and back- end.

Promote people's work, give them feedback, let them improve.

Prioritization of a backlog is crucial for every product. I would like to raise a question: what is more important for Steemit? A great engine, or a great community? An average user can see a fairly simple platform to post and vote not technical things inside. Despite my love of computer science, I would vote for a better user experience over engine improvements.

The discussed feature would greatly improve the quality of a user's life. It may be a step towards making a real difference in the way people interact with each other on social media. Let us all both improve our work and provide feedback to others, not only when everything is 'great', but also in difficult cases, without unnecessary emotions or making off-topics in a comment section.

Finally, the proposed approach would open a way to limit and expose all kinds of abuses of power, vindictive behavior or strategic or political moves. It would make Steemit the first platform which not only says it promotes higher ideals, but which actually pushes forward to achieve them.

I would like to put it on the list of issues on Github, but as in many cases I am biased by our own perspective.
You know what I think, I would like to know your opinion.


Your feedback is welcome.

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

You flagged my post yesterday and caused me to lose money. But let´s not dwell in the past. Here is an upvote from me instead.

·

I guess our private chat makes us more friends than enemies now :)

·
·

Flagging system was updated yesterday, did you notice?

·

I read the release notes, but this feature improvement is about something different. It allow users to provide a justification together with a downvote/flag. Don't you see any difference?

·

do you have a link to the post?

I Like it
👍❤️✅🆙
@radoslaw don't make a habit of downvote ,
I don't like that negative stuffs and I am sure others would agree with me too.

·

I would like to give people a chance to get feedback. If you think about flagging it does not say much about the reason. It makes people react emotionally. With a downvote - a subjective vote, without any comment it is even worse. I do not want to make a pattern from downvoting, I want to make the existing feature valuable :)

·
·

I had danman downvote me once for no reason, I still can't recover my reputation ladder until today.
He apologies, well , from that day I hate those that do downvotes.

·
·
·

To be fair, I downvoted/flagged @fyrstikken's post about the deal with Heidi yesterday and left a comment. We talked about this whole issue today, actually for several hours and we are more than good now I think. I would take my flag off if not the payout window :( Feedback is in my opinion a great way to start talking and improve as a community.

·

@bullionstackers have the same opinion here - @radoslaw - feedback is a better option - or if they really want to keep that flag at least give a person a warning and not be trigger happy or unreasonable in flagging.

·
·

I was going to use this at my #refer-it channel R
Now I am not, after looking at it.
Too much BS here , that I uncovered.
Only good contents stuffs will be in there.

·
·
·

Could you please elaborate a bit more, what do you mean by

Now I am not, after looking at it.
Too much BS here , that I uncovered.

?

·
·
·
·

I always, look at the history sources.
Yours proposal is good. 👍
You will need to convince me more , for me to Support this. In the Future.

·
·
·

I cannot respond to your response, could you clearly say what do you have in mind?
I do not expect any kind of promotion from you, but I would like to understand your motives.

I always, look at the history sources.
Yours proposal is good. 👍
You will need to convince me more , for me to Support this. In the Future.

Could you please write what exactly happened?

·
·
·
·

At this point, I will remain Neutral .

·
·

@englishtchrivy The problem with a flag is that it is actually a mandatory feature to limit plagiarism, spam, etc. but without a feedback no one except the one who pushed the button can respond to the question 'why flagging was used'.

I would even say that despite the whole 'negativity' of flagging, it should be used at some point in other cases like e.g. false statement - if someone ignores you on chat, your comment is one of 100 and you know that the statement is false (not because you think so, but e.g. 2+2 is not 146) flagging with a reason 'false statement' and feedback what exactly is wrong and why may be a good thing. It requires however more changes - payout windows should be longer to give an author time to edit such an article and a voter to remove a flag.

As we observe increasing number of flags already, adding a checkbox with categories and a field to put justification seems to be a must.

·
·
·

@radoslaw if it's just to control plagiarism and spam I think cheetah is already doing a great job. Can't we just refer any post that we confirm as spam and plagiarism to cheetah? I saw @onyx did that, or is it just for @onyx? exclusively? then perhaps it should be directed to him. And if cheetah confirms the article's stolen then flag our hearts out .
I saw some pips here abusing the flag and use it just to commit revenge or just because they want to flag the other person just cause they don't like them. I wish there's a system that goes like this - if the post is just not tasty - why don't first have people say "flag" because ....reason ... then if there's at least 5 or ten people who agreed that it should be flagged but those ten should have at least specifically given the reason for doing so then - the flag takes its effect. Some people just flagged something just because they hate the author - or they think the post is a trash . The truth is there's so many post that makes you cringe because they are either empty, no substance , not adding value to people's lives and obscene but those don't get flagged. The flag got abused for hate mostly for revenge which I don't get (rolling eyes) some even said - I'm flagging this because he doesn't want to see the post, happened to read the post and it actually helped me understand Steemit stuff more and I wasn't the only one it helped cause I went scrolling down to the comment thread but one just flagged the post just because they don't like the author - I guess we could also control our feed right? We could just mute the person, too if we don't want to see them or unfollow - I just don't get why flag? It's just one of those pointless misuse of the flag. - Let's make Steemit a positive kind of social media, there's plenty of hate in the world already. World peace people!

The impact a "flag" has with "no reason" for the author is HUGE.
Triggers a lot of questions of the kind "what did I do wrong?"

A feedback would be an improvement to the network as a whole. Sometimes even resulting on the "flagger" taking the flag out, if the flaw is taken care of properly before the 36 hour timer expires.

Thank you @radoslaw and @renzoarg for helping my out of a horrible situation. The lack of feedback is .. Please think about it .. Unconscionable and Unacceptable in ANYTHING in life. You are basically saying philosophically, psychologically and physically that the term "For every action there is an equal and opposite reaction" has NO value and is a moot point . There is nothing without feedback, life can't exist without feedback.
The point is, there is NO growth, understanding or construction without feedback. The creators of steemit are for the sake of conversation, 'programmers'. You can't code without feedback, how do you expect to 'code' a social network without a feedback loop to improve your code?

The upvote and flag just causes a dummy loop that not only prevents growth but detracts from the growth. So much 'Intelligent Design' on the back end, but chaos on the front end ?
I wish I were half as articulate as @radoslaw, perhaps I am, hence his post compared to my attempt at a comment ..

Please don't demand citations, this is just my opinion. I truly hope the concept is understood even though the wording and structure might not be.

Calling the master himself: @dan / @dantheman, could you please share your thoughts on this proposal? I have seen discussions on Github, e.g. Witness Police of Abusive Voting, but I think the presented is a complementary solution to all other actions. I've asked you in this thread where to put it, but I guess you missed it. Any feedback, please?