Downvote pool
Here are some of my thoughts on a downvote pool combined with 50/50 and the new curve discussed in this post by @vandeberg and of course also after his recent post talking about the downvote pool which seems to have received a lot of pushback from the community.
Even though I've given this a lot of thought and have some really long discussions with Steemians I feel are a big part of this ecosystem and understand it rather well I have to warn you that I may miss some points in this post. It's been quite an eventful week with a lot of discussions, debates and opinions, one thing is certain though, everyone cares about Steem and wants the best for it and what we have right now is clearly not.
First off let's talk about the linear curve. I have often in posts mentioned that even though it's not great for content discovery and the main thing taking the opportunity of it right now are bid bots I always hoped that it would only be the beginning phase and that over time there would be more services, innovations and other usecases for it. Unfortunately not many occurred and the toll has been on content discovery. Many who remained powered up on Steem have either turned to receiving passive ROI through bid bots/distribution bots/small part "SMT's" (such as actifit/steemhunt/etc) while many others have instead decided to vote-trade their way forward leading to closed circles of users rewarding similar content daily, having a guaranteed vote waiting for them thus quality declining much like when an author is on an autovote.
The current curve does not leave a lot of leeway for good curation rewards, as mentioned many have already circumvented that problem by delegating to bots and receiving a bigger piece of the pie than just 25% curation but the results of that are a trending completely filled by authors that buy these votes of these delegators. I'm not saying we should do anything to fix our trending, trending itself may not be under the best parameters right now as though how it calculates a post from being trending which mostly just depends on rewards + age. If we are to focus on content discovery again though and at the same time reward curators or investors who instead of delegating to bid bots would delegate to curation projects and receive about the same ROI or higher this could fix our content discovery and curation.
There are few cases today where curators are able to earn a 100% ROI or higher on their vote value for doing good curation, with 50/50 and the updated and discussed curve this chance would increase a lot more. Of course though bid bots will still exist, maybe not to the extent we see today but there will always be interest of buying attention in an attention economy. While they may exist they will have to adjust quite a bit and at the same time I think many investors will move away from them as their ROI won't be as safe as delegating to a curation project that focuses on rewarding general quality content. With this in mind let's start talking about the downvote pool.
I read quite a bit of comments in the downvote pool post by @vandeberg and many keep bringing up the "look at how flags downvotes are already pushing away new and existing users, how will this change bring in new users and keep existing ones. This is something quite shortsighted in my opinion because how do we know how many have refrained from giving Steem a try because of the low quality trending, low quality post rewarding and in general bad distribution due to there only being upvotes which has a lot of authors producing lazy content - myself included - since it's so safe for them to do so and they have gotten comfortable with it and downvotes have become this unusual thing which as soon as it happens they get so surprised by it that they go onto the defensive and retaliate or unfollow/mute/etc. At the same time, these few cases of random accounts flagging newcomers or pushing them away are so few that I don't think it's something we need to focus our attention on. It's similar to blockchain tech in general where early on many will abuse it and use it for all the bad ways you can imagine but at the end of the day the benefits, advantages and what it allows you to do will outweigh the cons every time, we just gotta get there.
With about 0.01-0.1% of downvotes compared to upvotes in this day and age it is quite clear that we need them. Reddit has around 10% downvotes and there the upvotes barely even mean anything. Now obviously they can't be compared but a system that allows you to earn from it's pool and investors will clearly need some more downvote activity than we currently have, things have gotten too safe and comfortable and close to Steem just becoming another proof-of-stake chain with "content" only being a placeholder to receive your ROI. I believe downvotes will mostly be used for the bigger reward pool abusers on the chain which don't need to be named as many of us are aware of them already, I'm hoping they will be used well and organized and without bias but I can't promise it will cause a lot of chaos early on - that's the way of open-source tech and innovation and new experiments and that's a good thing. Name other big projects today that do listen to the community, have witnesses who want the best for the economy and of course get rewarded for it and at the same time are so flexible with changing and trying out new things - I don't believe there are many thus this is one of our strengths.
While I understand that authors may fear these changes and may turn to say things such as "we're already earning this little" I am hoping they see the bigger picture. As a manual curator myself I can already tell you that I'll be curating authors a lot more, a 50/50 change will already incentivize me to undelegate from @ocdb and use it towards curation again and at the same time I won't feel the need to post mediocre/lazy content and instead reward the real authors. I firmly believe that the 50/50 change will quickly cause authors to earn more rewards than what they are doing now. There is so much stake locked up in bid bots and distribution bots (the lesser of evils @ocdb has 4.3M SP right now) when it could instead be used for curating authors. The more we downvote what we perceive to be low quality content and clearly seeing that it has been bidded up the more we will make good curation even more effective. Remember that when you downvote a post you're at the same time rewarding everything else by that much spread out over all accounts - I know it feels like nothing but if enough of us do it we will quickly see the right kind of content being rewarded.
Having said that I fully expect there to be retaliation of authors comfortable with the way the system works today, especially those who have only been looking to take out as much value from our ecosystem as possible without giving back much. I also expect new secret bid bots to pop up but at the end of the day if we make downvoting & curating great again these authors will at least have to produce quality content they bid on, not what we currently are seeing on top of trending.
Looking forward to discussions about this, as I mentioned above there may be some examples, aspects and simulations I may have missed and would love to be pointed to in the comments so we can continue the discussions there. Right now though I believe these changes will only strengthen our platform and have been long due.
I think you have made a reasonably good assessment of the Steemit EIP. I think is important to emphasis that there are 3 parts to the strategy (curation, downvote pool, and rewards curve). These three are used to increase the incentive to curate.
I understand the downvote pool is most likely to face the most resistance. There are a few ways this can be implemented. Offering curation rewards for downvotes would be a bad approach. It is good that is not being proposed. A downvote pool that offers no reward but does not take from a users voting power is far more reasonable. As it stands, downvoting costs the downvoter almost as much as the downvotee (assumption of 1 lost selfvote). That should not be the case.
I think part of the problem with all of this is that there are so many related moving parts it is hard for people to get a grasp on the entire picture. This means that people focus on a narrow view with a couple of points that affect them but don't necessarily consider other aspects.
I would say that in combination, a slight curve, 50/50 and some flags reduces a great deal of abuse and incentivizes better content as there is more in the pool and more manual curators (hopefully). To win support now, send a bid. With manual curators it means supplying what is in demand well enough that it gets support.
The bidbots can still operate but the "main"
abusers will likely have to change their content or stop using them as they won't want to incur loss. If they really just want eyes on their post, the loss won't matter as it is the cost of advertising. Also, with the curve, the larger the votes are on a post, the more impact a downvote will have (as I have come to understand it). This means trending could look pretty damn good soon.Because it is so complicated to visualize, I think the only way to really get a good feel of it, is to try.
I think retaliation is the biggest reason people hesitate to downvote the biggest, most powerful abusers. Especially if their Steem power is so low and barely puts a dent in the authors rewards. This is a comment I left for @whatsup that I will copy here to get your opinion on:
“Downvoting is important but there are at least two general limitations. 1) you need Steem power to effectively reduce overvalued rewards and 2) you put yourself out there for retaliation, which in my opinion is the main reason the current proposal will not work ... and why people hesitate to downvote now.
One solution would be for the community (aka a witness or Steemit) to create a downvote bot that people can submit requests for downvoting. The name could be @communitywatch. Importantly, people could submit these requests as encrypted memos so at the end of the day no one knows who flagged them, eliminating the possibility for retaliation. Now, the problem is how will people abuse this system? Good question! I don’t know. The specifics would need to be worked out but the bot could limit the number of of downvotes submitted by a specific user, such as 2 a day. Or it could limit it so one user could only downvote another user so many times a week. Something like that. I don’t know what the right balance would be.
This type of system would also give minnows more power and since they make up the most accounts, watchdogs would be everywhere.
Another question would be how strong a downvote should be cast for a given submission? I don’t know the solution to this. If one post gets many submissions for downvoting, maybe the strength of the downvote exponentially increases.
My initial thought is all requests should be 0.001 Steem or something like that so everyone can freely submit requests. Maybe it should be higher if needed to compensate the person running the account. Maybe Steemit could delegate stake to the account that is just sitting idle. Alternatively the community could delegate Steem power.
This is just a rambling idea I have had for a while. What are your thoughts? I have to imagine individuals would be more likely to go after powerful bad actors if they weren’t afraid their Steem account would get downvoted to dust.”
Good idea! At least better than developing anonymous downvote on-chain altogether.
No hardfork required. Another pro.
You just planted 0.11 tree(s)!
Thanks to @ucukertz
We have planted already
8443.60 trees
out of 1,000,000
Let's save and restore Abongphen Highland Forest
in Cameroonian village Kedjom-Keku!
Plant trees with @treeplanter and get paid for it!
My Steem Power = 21351.94
Thanks a lot!
@martin.mikes coordinator of @kedjom-keku
@steemflagrewards already serves that purpose.
Delegate and follow their downvoting trail if you like that solution.
More downvotes helps the abuse fighters currently serving the community.
It won't solve the distribution issue.
Which is the pinnacle of the problem, imo.
I address that here.
@steemflagrewards is a great service but the one vital difference is it is not anonymous. Are you associate with them? If so, would they consider taking requests via encrypted memo? Thst would help in my opinion.
This link is good for one day.
https://discord.gg/nPZuAb
It is not anonymous, but you can set the bot not to downvote anybody with more stake than you.
Cool. Thanks.
While I agree on many of what you mention above, at the same time I feel it's all wrong. Those who will receive the downvotes are more likely to "retaliate"...so chaos indeed. It'll definitely become personal. I don't really think the a separate pool for downvotes is what we lack of...
As for the good content that needs to be discovered....it can still be found...do we really need a 50/50 reward system to wake up the whales and be more active?
How's that gonna help a newcomer? or a guy with 500SP whose vote at the moment is almost none existent? How will they be benefited from that? Hoping that they will receive a vote from a big account that wouldn't have landed on their post in a 75/25 reward system?
Everything is already complicated and it's gonna get worse...that's my point of view...
I'm sure those doing the downvoting will be aware of this, if not there's always going to be a lot more upvoting power to help undo what others may find was unfair downvoting due to retaliation. There could even be guilds spawning to do just that.
How much stake is locked up in bid bots/other right now? 50 million SP? 80? It's not news that everyone will go for what is most profitable to them, in this day and age it is also pushing those that would want to curate to do the same thing cause else they're left with lower stake in the end than those going for max ROI and not caring about curation and rewarding authors.
If it starts with “If everyone would just...” It will never happen. People don’t behave like ants, acting as a unit in support of a greater system. You have to build systems that work based on people behaving selfishly. At the end of the day that’s what a corporation is as well as a military. In a corporation you work for your own personal enrichment in service to a larger system. In the military you’re legally obligated once you’re enlisted so this is also for self preservation.
Posted using Partiko iOS
I agree, for a long time many curators have had to just be altruistic for there to be any sort of natural curation happening. That's why 50/50 will do authors more good in the end while at the same time keeping curators/investors happy and not relying to delegating their stake to only selling it for attention to the highest bidder.
I honestly don't mind giving it a shot, I think the 50/50 split could very well improve curation. The thing I feel like won't work in this whole plan is the "group/community" flagging.
I'd rather we just put it out there, so people will see it doesn't work and we can move on to finding another solution.
I think for us to have trending page that doesn't suck, we'll have to use a more complex algorithm to adjust for the real life conditions of the platform. Bots, curation trails etc.
There is a clear tendency to focus only on the static effects of 50/50. It’s clear that at least some manual curators would become much more active after going 50/50. I’ve been told smoke.io has 50/50 curation there’s some actual data - possibly applicable - in existence already.
Posted using Partiko iOS
One of the main reasons I invested a little in Smoke early on was due to 50/50 and from what I've witnessed so far there's always quality content on trending. It's going to be interesting seeing how that will work on Steem, especially now with stake being more distributed.
About the issue of bid bots actually. Majorily one of the uses that bid bots supposed to have is to help your content gained visibility and exposure. I.e if your post is of quality content. But also this bid bots also can't work without delegated steem power. The bid bots have now been misuse by many to promote low junk quality content up there.
I am not trying to defend the bid bots non am I trying to discriminate the bid bots but I am just trying to show both the good and bad side of it all.
This statement you said "The more we downvote what we perceive to be low quality content and clearly seeing that it has been bidded up the more we will make good curation even more effective" yea. Thus statement is also talking about how the bid bots are been misuse like I have stated earlier. Low quality contents are been bidded up even though they are not worth it, or let me say the quality is low. There are good quality contents still out there but lack exposure, and visibility and this can even make the authors to leave the community out of frustration because I once thought of quitting steemit out of frustration.
If the manual creation can be address, it I'll surely help the community, helps the author who are feeling frustrated due to low value on their high quality contents .
Like I said earlier, I amM just trying to share my viewonn it. Nothing more. My comments is welcome with some contributions
That's where the downvote pool will have the most effect, today we can't even downvote low quality bidded up content. As in even if everyone active with stake would try there wouldn't be enough stake to downvote those posts down from trending.
The article says 'what we propose', so is this going to happen? The last time STINC, sent one of these out re: the downvote it happened real quick.
I dont delegate much of my SP to Actifit etc.., as you may or not be aware.. I like the power to vote on things.
What you mentioned about Circle Jerking can happen inadvertently due to sheer human connections and friendships. I'm all up for a different way of doing things, so I say lets try it.
The separate pool for flagging? Yes, you see down votes all the time on YouTube etc.. but where there's money involved it tends to get personal. I don't see it making a whole lot of difference, do you?
Yes but on the other side of the coin, where money is involved there is a lot more profit motivate toward milking rewards, which makes downvotes even more important.
There are two sides to it for sure, but it is important to indeed consider both sides.
Yeah and those are all great, after all it's a social media platform so no one is gonna judge someone for voting on their friends. There are though unfortunately "friends" that are only friends cause they own a similar amount of stake and enjoy getting the ROI and attention of trending posts no matter the quality/quantity.
I think the downvote pool will make a big difference, I was not sure about it being that high to begin with (25%) but considering those that don't want to use it won't I think it could be a good number to fight all of what users believe is abuse of the pool.
As @tarazkp said though, there's so many different variables in effect here, the best thing we can do is try.
Oh and about your first question, it will depend if the witnesses agree and accept the HF.
On reddit, they have some provisions so you don't see the votes right away ( at least with the comments)
The thing I do fear is down vote organization. Like minded people get together and push their votes or down votes for their gain. If this has a bunch more people on it, it will be hard to regulate.
I think interactions in posts should mean something. Trending posts with 1 comment is a bit ridiculous. I think we need more dialogue and interaction in posts other than the upvote.
Posted using Partiko Android
I agree with your ideas and we should be having more often conversation reviewing ways to improve the experience of people here.
I see a lot of repeating posts that in the end don't add any value and are just there to have quantity in order to get their daily votes. This also includes the 'spam' of some apps that post in your name to promote their services using your rating, activities or information and sometimes offer upvotes to do that.
Yeah the trending does definitely need a rework.
Unfortunately hiding votes does not work on a blockchain, but there are a lot of other good things that come out of this instead. :)
We need a downvote pool just with the current system or the old reward curve but not 50/50 curation..
Bid bots, vote sellers and other non-curators also profit from 50/50
Posted using Partiko Android
Yeah but the thing is that they won't profit as much as curating on good content that will get more votes on top or won't get downvoted. That's the idea afaik, if it's going to be more than delegators receiving both bids and curation rewards from a bid bot or distribution bot is what we will have to see. The free market will adjust quickly I believe.
I've also been thinking how it's going to affect @ocdb, we're gonna have to remove the "10% guaranteed profit on vote" but I don't think as many posts from whitelisted authors will be getting downvotes compared to bid botters.
But the bid bot market decides which posts have top payouts..
Posted using Partiko Android