The Situation
Voting behavior and content curation have reached a tipping point. For the good of the platform, I think this needs some serious scrutiny.
See what you think of these screenshots.
In each of the above posts, 30+ votes were accumulated within two minutes.
Below is a spreadsheet showing duplicated names in red.
There are almost no unique voters between the two posts. A post is chosen, and all of the same accounts instantly pile in like lemmings.
Now a screenshot of what the “new” posts tab looked like a few minutes later.
Most new posts have a few or zero votes.
Interspersed are posts that have instantly accumulated 30+ votes before it’s even possible for someone to have properly looked at the content to determine whether it’s adding value to the site.
Scrolling down the list over the past two hours of content prior to this writing, it looked the same all the way down.
Because the pile-on accounts have no Steem Power, and are voting so often, this behavior isn’t actually causing most of the affected posts to do well and achieve payouts. What it does do, however, is make it impossible to tell what content people actually like, versus what random bots and no-look clickers just piled into by the hordes.
Personally, I feel this is incredibly counter to the spirit of the curation process outlined in the Steem White Paper.
The White Paper speaks
The key point is on page 17, under Voting Collusion:
"While cooperation to distribute funds to the best work is the desired goal, collusion that undermines this objective should be minimized. There are two kinds of collusion, the most straightforward is when one user simply buys a larger stake than others, and the other involves coordinating a large number of smaller stakeholders to work together.”
What we’re seeing in this pile-on voting behavior is collusive activity that undermines the goal of the platform. But the platform itself is incentivizing this behavior by rewarding voters based on how rewarded the post they vote for becomes. Under this circular system, how can people be encouraged to design bots that find and upvote actual good content? And how can people ever vote for what they like instead of what they think will end up with the most votes?
Unless people decide they don’t care at all about rewards, they have to follow the herd — and if they don’t care about rewards, there’s only one thing to differentiate Steemit from the other platforms people already use for no reward: that idea that the content will be curated for quality by objective voters.
Unfortunately, objective curation is impossible under a collusive vote-herding system that encourages pile-ons into content that can’t possibly have been judged properly at the time the voting occurs.
According to the White Paper, the concept of negative voting i.e. flagging is intended to counter abusive collusion. However, unlike with positive voting, there’s no direct reward for negative voting. The only incentive to properly curate and down-vote bad content is the vague idea that you’re helping the platform — which you are! But who wants to toil away without recognition in the hopes of defeating the bot and spam hordes, when it’s easier to just pile on and benefit directly instead?
But abuse is okay! …
Some will probably notice that the White Paper also says this:
“Eliminating “abuse” is not possible and shouldn’t be the goal. Even those who are attempting to “abuse” the system are still doing work.”
This sounds like, “Eh, whatever. Abusers? No sweat.” But the important clarification is just a little further down in the middle of White Paper page 18.
“All that is necessary is to ensure that abuse isn’t so rampant that it undermines the incentive to do real work in support of the community and its currency.”
What caused me to write this post was my personal belief that we’ve crossed this line and that the abuse is discouraging actual participants that can drive the health of the platform, and who can ensure proper growth and a friendly environment for non-abusive users to join.
Feelings
One currently trending post by @chitty is trying to make the case that getting lots of votes proves your content is well liked, even if you aren’t being rewarded for it immediately, which should be enough to encourage people to keep trying.
However, based on all the information I posted above about the landslide pile-on nature of the current voting system, I don’t feel the “just keep trying” line of thought presents enough of the picture — and that the flaw is not just with the users who feel discouraged. As this popular post by @dev001000000 shows, there are many others who feel this way.
Also, as I say this, please keep in mind who I am: Someone you’ve never heard of. I’m just another minnow. None of my contributions have been rewarded, and that hasn’t bothered me. I haven’t made a huge number of posts, but the ones I did make were thoughtful and fully original attempts to add value to the user experience here. I also have both of my computers mining Steem in the background as witnesses, and in the last week or so I’ve found about 30 POWs in the course of my support for the blockchain. All of this is to say that unlike many of the power users, I have hardly any financial stake at all in the success or failure of this platform, yet I’ve been participating regardless. I only want to see it do well because I want to see it do well.
I love the concept of Steemit. I want to support the site and see it revolutionize social media. But right now, the only thing we’re revolutionizing is bot voting — while witnessing the fascinating phenomenon of humans beginning to behave like bots instead of the other way around.
This thoughtless “pile-on” behavior, where we can’t even differentiate the insta-voting bots from the humans who are emulating them, can’t possibly lead to long-term prosperity for the platform, or to a user environment that is friendly for the average users like me — users who represent the vast majority of potential content contributors and consumers on the site.
I don’t have a brilliant idea to solve this. I can’t even state in straightforward way what it is that’s “wrong” and needs fixing. But if, like me, you want to see Steemit prosper as a platform for original content that can be rewarded for its quality — or just as importantly, not rewarded when lacking quality — this can’t continue.
If you have a suggestion, I’d love to hear it. If you know an influential user that feels the same, please ask them to raise the issue so that it will be heard by someone in a position to do something. If you know @dantheman or @ned or anyone else that might be able to respond, please consider talking to them about this, and #raiseyourvoice to support proper content curation.
Thanks for your time.
I think this could be interpreted as a sorting problem. This does not necessarily have to change the economic policies. It might be a site interface thing.
Yeah, maybe the only change we need to make is to remove the ability to see the number of accumulated votes since that metric is now insignificant. Just reduce this clutter. Small changes like this are harmless, but improve the look, feel, flow, and overall Steeming experience.
I had noticed this myself and wonders show posts with 30+ votes could have no payout. This makes sense now. I think the goal of these bots is to systematically mine steem by being early upvoters of popular content. But it seems they just randomly select posts and hope for the best, as much as a bot can hope. ;-)
I wrote two related articles:
The new curation system penalises those you upvote quickly. In the first thirty minutes, the curation award is shared with the article author. The quicker the upvote, the more goes to the author, starting at 99%.
The problem I pointed out is that bots could collude. Likewise, others have pointed out that it's not that hard to get the bots to more time. Of course, if others vote in the meantime, the bots will be awarded less of the curation reward. Perhaps bot authors are just voting quickly and hoping to win by covering as many posts as possible.
I upvoted You