RE: Flagging War: What The Hell Do You Have In Your Mind? A Short Term Solution To Fix a Long Term Problem?
The frontend and content discovery is a joke for now. This is a social media platform at the end, is not the frontend the most important side of it??
Couldn't agree more with that!
Stop the voting/downvoting bots completely and force manual curation in this platform.
How? With a digital layer it seems like a futile exercise to even consider the possibility.
You will let the blockchain become a place where people do not want to post for the fear of being downvoted.
Actually I get about 5-10 people a day telling "me flag this flag that", When I ask them to at least flag it themselves they respong they're too scared of reprisals, and don't really consider relying on a whale or some "justice system" to defend them.
I'd like to see more projects like @freezepeach (dead) https://steemit.com/introduceyourself/@freezepeach/freezepeach-the-flag-abuse-neutralizer that make people feel safer that there is logic in the policing being done and that they will be defended with the same logic if someone decide to go rogue on them.
Every flags should be accompanied with a reason if it is not disagreement on reward. And bringing a post below 0 is definitely not a disagreement on reward.
So maybe a downvote button instead of a flag and a strong warning before downvoting anything below 0 could help.
How is it possible that a blockchain supported by millions and dozens of developers, is not able to come up with a good frontend? Really, this is something I cannot live with.
Bots are gonna kill this platform. A way must be found. I see my posts getting 45 likes and having 5 views. This is RIDICOULOUS. We could have a review system for applications using the blockchain, like in many cases outside Steemit. The review could be based on the consensus of the community or the applications should pass a certain verification process, that could be even automated.
I agree
Agree with this as well
Low value bots (spam bot) can be suppressed by bandwidth limiting and usually removing their financial incentive is enough.
I was considering doing something to break the trails, it's easy to detect account that blindly follow some authors, and the fact that it happen with users at high level is concerning, (Ex. @timsaid trailing @gavvet). I don't see it as being so much of a problem though as vote count doesn't affect the user experience. (unfair reward does though)
What would also be great is to have a way to reduce reputation of an account that isn't posting anything. Whatever that would do, it would be in a way disagreeing with someone's voting pattern.
Reducing the reputation due to inactivity sounds like a really interesting idea, never thought about that!
New accounts have too much bandwidth and account creation is still faulty when I find hundreds of new accounts a week being created for free by @steem by those like @dart & @zlo
Which is also the case for a lot of my photography posts - but it's hard to determine if real people like the thumbnail (which could be classed as a "legitimate" upvote, or if it's a bot upvote.
I think I see one solution. Would it be possible to add in a future hard fork that all upvotes and rewards from accounts that don't also generate a view be removed automatically?
I think upvotes are a function of the blockchain itself, whereas views are a function of the Steemit UI. People using other apps to view the blockchain will not cause an increase in the steemit view count.
That is actually a good idea. You have to at least open a post to write a comment, so why not for an upvote?!?!
Because it's trivial to make a post? Also, do you really want to limit consumers that can reward content creators to other content creators?
Trying to throttle the bots is becoming critical. If something does not happen the value of steem will go towards zero. If you do not open a post, are you really a consumer? I am not sure what the answer is here, honestly. I know I had some steem on poloniex quite a while back and that I learned about the platform much later. Looking at the currency evolve and seeing blocktrades offer "power" as an option is kind of hopeful.
I'd love to work on a better frontend for steem but my job takes up all my time :(
@freezepeach is not dead, but it's in beta and is running at a slow pace to work the kinks out. I worked extremely hard on being accurate in that intro post, but even then I knew that I am still a pup on this platform and I don't have it all figured out, so I have proceeded cautiously while the wallet is small and there is little incentive for people to overload it with their spam that got flagged.
I have been working on this quietly in the background, gaining an alliance, asking for perspective, and listening to some of the most respected names on the platform. @danielsaori was also kind enough to make an extremely nifty application for admins of the service to neutralize flags easily, which will make it easier to manage.
Expect a new post from the service announcing clarifications on some topics and the future ahead for @freezepeach within the coming week.
@r0nd0n
Good to know, you can count me in for support.
Could you as well have a look at my solution regarding UserAuthority (UA)? I've written two very detailed and well-recieved articles about it and have begun developing the code as well together with @stoodkev --> https://github.com/realScipio-ua-js
You could also embed it / implement it to freezepeach, for detecting / identifying malicious flagging:
How about my UserAuthority (UA) HF21 proposal?
account_can_downvote (bool) = (ua_account - ua_minimum_threshold > 0) ? true :falseThe UA probability distribution uses all follower data inside the blockchain already (with follow-transaction per block). By dumping all UA results (per block) inside a binary index (ua.db), we can hash the ua.db result and store the hash inside a blockchain block for consensus.
And it's not just an idea, I've been building the code already together with @stoodkev here https://github.com/realScipio/ua-js
Can you calculate already some examples?
What is the UA for @berniesanders? What for @blocktrades?
And what for @cron, @spaminator, @appblizz, @markboss?
I'd be curious about that as well. I'd have to do some work on bots. I use a private bot that was graciously shared with me to fight abuse.
@mack-bot & @quarry are used as markers and probably wouldn't gain a following and won't follow anyone. If they did they'd probably be followed by spammers to watch what they are doing. lol
Not yet, I'm also on a backlog of Utopian posts to be moderated. :P
PS: also keep in my mind, once UA is deployed, new "rules" apply, so different behavior. Once people understand their following influences the UA scores of everybody, they will think better about who (not) to follow...
New rules, new behavior.
I echo the sentiment about the front end. I want to discover more yet I find I have to wade through so much signal to get to the noise.
Trending is too static. New is a lot of stuff - it’s like if you wanted to represent IoT with posting. Tags are a little weird in terms of topic. And all of these contribute to hinder you from getting a quality Feed.
When I think about how something like Spotify gives you outstanding discovery, i hanker for that. I can only see that coming with maturity on the site though, and that might not be quick.
When I first started with @steemcleaners there was a larger portion of the community reporting abuse that commented & flagged posts themselves and reported it for a reward. Today the number that comment and flag themselves is relatively tiny compared to the number of reports.
100% I can do little to help those flagged maliciously. Many don't understand that the members of @steemcleaners cannot use the account directly to upvote. We use our own accounts to try and counter the flags.
As long as I can use a bot to do it I don't have a problem giving a reason. @mack-bot & @spaminator do not comment just to prevent more spam. Every flags should be accompanied with a reason if it is not disagreement on reward. And bringing a post below 0 is definitely not a disagreement on reward.