This was successful but with the rise of more short-form content on the platform (content that can be read or viewed in less than 30 minutes), the community and the witnesses have come to a consensus that the 30-minute rule is taking curation rewards away from human voters who are actively consuming content and voting on material they like. For this reason, HF20 will reduce this window from 30 to 15 minutes.
I find it difficult to imagine that no one but me sees the obvious problem with this solution.
What human being spends 24 hours a day sitting on Steemit waiting for content? Does not the usual use case suggest that people have lives that they go about, and occasionally engage with one to two hour sessions – if that – of using the platform actively? Doesn't that suggest that a shorter curation reward window actually rewards bots far more than human beings, who are much more likely to discover the existence of the post or comment well outside of a 30 minute window. Possibly, given the amount of time it takes to recharge Voting Power, on the order of a couple of days between major sessions of engagement with the platform.
This rule seems specifically and explicitly designed to privilege bot monitoring of feed streams much more than human engagement. Even for short-form content, the discovery time is going to be well beyond 15 minutes.
I'm not even suggesting that this idea is wrongheaded. It's just wrong. It's exactly the wrong direction for solving the problem you say it's supposed to solve.
If you want to reduce the advantage that bot voting has over humans, reduce the advantages that bots have over humans. Make the curation reward window a day long, drop the decaying payoff aspect, and simply treat all upvotes within the window completely equally. Then humans and human curation actually has a chance to find this content and vote it up and aren't penalized for not being a bot with nothing else to do.
HF 20 is going to increase the number of bots that people use to get maximum curation awards. You get what you reward, and you are rewarding and automated process which merely looks for new updates on feeds that are likely to be popular, tries to wait for the optimum number of minutes, and then puts in a vote. It has nothing to do with the content. This effectively makes the content mean even less in the context of human interaction.
Congratulations. That's quite impressive.
This will better serve the original mission of the curation rewards budget: to ensure that the Steem blockchain distributes rewards to the most valuable content.
Wouldn't it be much simpler to simply remove self-voting from the system and maintain the rest of the architecture as is? That would immediately and completely reduce the author's rewards from any curational activity without pouring any funds back into the top of the hopper – in theory rewarding people who weren't even involved in either creation or curation of the piece in any way.
In fact, that would suggest that the greatest stakeholders have an even higher motivation to choke off rewards to people they don't like rather than reward content that they do like.
All of that money goes back into the hopper and has a much greater statistical likelihood of landing on their head, after all.
I reiterate, you get what you reward.
Stop rewarding self voting by making it impossible and you'll still get curation and creation. Simply "redistribute" the authors' cut that you think they don't deserve and you'll get less authors and more people interested in "redistributing" those funds largely back to themselves.
In hardfork 20, this “vote dust threshold” will be removed. After this change users with any amount of SP will be able to cast votes so long as they have sufficient bandwidth. Votes that are below the threshold will be posted to the blockchain but will have no impact on rewards.
Will accounts which actually possess SP be able to lodge effectively zero SP votes on content? This might actually provide us an opportunity to differentiate content which "more people should see" (social upvote) from "I think this author deserves some money for what they're doing."
And yes, those are different ideas and different things which cannot at this point be signaled to the system differently. For a social network, it's amazingly non-social.
Allowing that finer grained expression of intent will actually be a positive help.
... it is also important to disincentivize rewarding content with respect to which no other stakeholders see value.
Why?
This is a contention without any support.
Why would it be important to the platform to dis-incentivize voting up content that you legitimately see as valuable, no matter what other people think of its value? After all, the underlying assumption is that if I think it's valuable than somebody out there is also likely to think it's valuable and I should reward it.
The alternative is in many respects what we see now: chasing the Dragon. Everyone is looking to jump on board the next big viral hit and playing the numbers to do so because the actual content doesn't matter, rather than simply seeking out what they find to be good and valuable and rewarding it. Likewise, on the creator side, chasing the Dragon ends up with a vast number of crappy, minimal effort, minimal investment posts about cryptocoin and cheerleading for STEEM – which are rewarded because everyone sees that they are valuable in the sense of can easily acquire tons of upvotes, and you get the social network equivalent of incest.
That surely can't be exactly what you want. And yet – that's what you reward.
The changes required to add support for PoW mining for discounted accounts will be included in hardfork 20, but the actual PoW mining will be added later as a softfork on top of HF20.
Isn't this really just a backdoor way to get proof of work mining into the STEEM blockchain? With the removal of the powerdown restriction, even with the account creation fee getting burned rather than turning into SP for a new account, this just looks like one more way to get swarms of vote dust bots to leverage hovering around the SP baseline. Worse, the people that can afford to run bots in numbers are exactly the kind of people who can afford to run proof of work mining systems in order to create bot accounts in numbers at an even steeper discount.
Along with the changes to the payout timing window, this is one more place that more bot interaction is going to become ever more frequent rather than diminish.
I admit, I am not feeling particularly sanguine about these changes. From a user perspective, particularly one who is interested in the social network aspects of the system, these look to dis-incentivize interaction with the blockchain as a person, dis-incentivize being a creator even over what the situation is now, and open the door to even more bots at every turn.
It's as though the expected use case was never for people to be using the system at all.
I feel like that's kind of a problem. I may be the only one, but that's where I'm at'.
at the end of day we have to ask ourselves do we need curation reward that much? With any set of rules imaginable it will be bots playground.
This is why I have proposed eliminating curation rewards altogether. I have also proposed delinking VP from SP, and instead weighting VP with reputation that is actual community vetting.
It isn't stake-weighting that draws people to Steemit, but rewards. Stake-weighting was intended to incentivize investors to use SP to direct content development, but instead it has merely become a vector for financial manipulation.
Coupled with a mechanism that precludes bots, Steemit would be truly a social media platform - but investors would be offered only the incentive of capital gains by which to profit from their stake.
Those with the stake prefer the immediate returns gamification of rewards provide, and content be damned.
No. There's no need to broke the whole economy to eliminate curation bots.
If you think this economy isn't broken now, you have no idea what you are talking about.
It would be hard to make it worse, it's so broken.
I'm pretty sure it can be worse )
It can be worse - but not much. 93% of Steem is concentrated in 1% of accounts. It would actually be difficult to more concentrate Steem than that.
I believe in observing the ecology to determine what's going on in it and what those who participate in it want and need.
Given that the vast bulk of the people operating on the blockchain are engaged in curation manipulation in order to acquire reward, then I think it's safe to say – judging by the rules that Steem, Inc. have laid forth, people really do need curational awards that much. They want them. They want to engage with them.
Just not personally. They have no interest in actually playing the curation minigame themselves. It's far more rewarding to build bots who engage in playing the curation minigame algorithmically.
You get what you reward.
I cannot imagine a number of sets of rules which will not advantage automation significantly, but all of them ultimately hinge on making a change in basic assumptions about what STEEM means in the blockchain.
At this point, the basic commodity itself exists purely as an authoritarian, top-down assessment of your value to either create or successfully bet on pseudo-viral content. Not content which is good or reportable to a community, but content which other people will judge to be sufficiently viral.
And that's really all it means. If people like myself who are creators but who don't particularly make things which are likely to be runaway popular get rewarded, that's a pleasant accident. It means that some people probably lost their bet on me and my work.
Worse, it means that if they did vote my stuff up with the intention of me being rewarded for providing them something that they liked – the mechanics of the game say that they're playing it wrong. That's not how to get maximum reward out of the system.
From my perspective, that's very sad.
Maybe I spend a lot more time thinking about how to create sets of rules that people want to play with, but it certainly not incomprehensible to me that a set of mechanics could be put together which incentivizes an individual engaging with the platform in ways that rewards them with things other than bigger numbers that they want. To do that, it takes an effort to understand what else they might want.
The current population of Steemit? Most of them just want bigger numbers. And that's all.
It's no great mystery.
It is quite sad.
I think you've got some decent objections. Maybe you should write it up as a post on its own?
I think bots are a fact of life. Perhaps so much so that they ought to just be built-in to the system and integrated. No matter what algorithm you can come up with, someone will find a way to game it. The only way to win is not to play.
The way bots work is that they auto-upvote posts after 'x' delay from authors x, y, and z. (There may be other ways, but those are the ones I know about). What if you just say "the people I follow automatically get upvoted?" I dunno, maybe that makes the world worse.
The thing that worries me is that the number of ways to maliciously game curation awards is high, and the number of ways to do it relatively trivially is still pretty high.
I don't know the solutions for that, but you've definitely started me thinking.