You are viewing a single comment's thread:
RE: Reflections on the past, present, and future of VYB (Part 1)
The biggest accomplishment of free downvotes was, it eliminated bid-bots and upvote selling/buying. It was able to achieve this with the implementation of Economic Improvement Proposal that also increased curation rewards from 25% to 50%. It was actually impressive how quickly bid-bots either shut down services or became curators.
There used to be about a dozen bid-bots and since they offered good returns compared to curating and upvoting for free, many chose to delegated their stakes to bid-bots. Trending page would be filled with posts upvoted by bid-bots. And this was destroying the entire proof of brain concept in practice. It is free market, people always find more efficient ways of profiting as long as protocol allowed, and it did.
Proof of brain was disappearing. Thanks to curation projects like Curie and OCD, proof of brain was preserved and authors still had opportunities to get genuine upvotes for free for sharing their talents. But if it continued that way, it was a just matter of time for Proof of Brain to be gone from the ecosystem completely.
Free downvotes may have its problems too, I can see that too. However, these issues are tiny compared to bid-bots reigned world. Layer two solutions are not immune for upvote buying/selling phenomenon either. Once layer two solutions grow in size and value, and content rewards reach consistently high values, people again will try to come up with ways to maximize the profits. But layer two solutions have more freedom to experiment without breaking the overall network and come up with better solutions.
One of the ideas for decentralizing the abuse fight can be, taking into account of the downvotes by several people and increasing this threshold higher over time. For example, there used to be steem-promo project that have several curators, and in order to automate the process of consensus among curators without deliberating about the content, they wanted to create a script that would keep track of upvotes by certain curators. If four or five of the curators upvoted the same content, then the main account would also upvote it. I wrote them a python script to do just this.
Similar process can be implemented for muting temporarily or permanently based on the downvotes the content receives from certain curators. Let's say there is a group of 100 high reputation community members to start with, and this number can increase over time. If a certain content receives 5 downvotes from any of these 100 curators/members, then the account will be muted for a month. If there are 10 downvotes from any of these 100 members, then the account will be muted permanently. Of course these numbers can be changed, and continue changing as the community/tribe grows.
On layer one level, there can also be changes implemented to if there is enough interest, I doubt there is at this time. Free downvotes also are limited. There is only certain amount or percentage of free downvotes each stakeholder gets and it replenishes over time just like HP. If this amount/percentage of free downvotes could be turned into a parameter set by witnesses, just like they set parameters for account creation fee, APR, etc, then witnesses will have an ability to increase or lower the amount/percentage of free downvotes based on the current conditions and needs of the network. They will also have to reach some sort of consensus on this.
I've been thinking about the benefits of lowering the downvote power. It does seem a little high now; overkill. I'd support that. It's been brought up numerous times now since Hive's inception.
What happens though if the downvote power is lowered to say 50% or even 75% less than what's offered today. That's far less damage one account (or several accounts run by one) can do if acting nefariously. More people would then have to downvote in order negate instances of abusive/exploitative behavior.
Has anyone been paying attention lately. Do you folks see how extreme things can get if you even so much as speak to someone who's used their downvote button... ?
I like that idea of the account being muted, temporarily. Being put in time out. I had suggested something similar awhile back. I also suggested affording members the opportunity to be able to write an appeal (similar to our current proposal system) if they feel they're being pushed around by downvotes. A platform where they can handle these disputes professionally (rather than exploding on everyone and everything) and somewhat privately (away from their blogspace). Community members can voluntarily review their case and if it's agreed the downvoter is out of line, they can have their ability to downvote muted for a month.
Downvotes is actually very interesting social and economic experiment. Because it kinda comes with some sort of negativity, I try to avoid this topics and don't pay attention to such events. There are always better things to do.
Downvotes is actually a brilliant consensus mechanism for rewards distribution. The system is designed that stakeholders need to decide on what rewards what content or account should receive after 7 days voting window. Without downvotes the system wouldn't be complete. So it is very rational tool. The paradox is that we as rational thinkers we humans can comprehend this concept easily, but as emotional being we may react in an opposite way.
For example, I received a downvote just a couple of days ago that slashed post's pending rewards by half. Reason dictates - "great, system is working and rewards don't belong to anybody until they appear in their wallets". But emotions wouldn't be happy and question the motives.
The answer is I think in understanding where rewards come from in the first place. All Hive rewards are basically go out of collective stakeholders pockets. If stakeholders majority decide that there should be no content rewards at all, code can be changed and there will be no content rewards. Hive can function just fine without content rewards, and layer two solutions can produce content rewards.
Since the inflation dilutes stakeholders' shares, it makes complete sense that they get to decide how the rewards distribution is done. In the end, stakeholders are the ones who lose or win the most.
Ultimately, the system will grow and get more efficient as it gets more decentralized. I think we are on the right path overall, and ahead of many other attempts at creating decentralized standards for social platforms and web.
Decentralization requires participation. It looks like your content and art haven't been participating as of late. When are you going back to producing content?
View more
As someone who is being downvoted to zero and all upvotes are being downvoted, I still believe the current system is ideal.
There will always be a few bad apples, but the majority of people don't downvote. They don't want to take the risk of reducing their own rewards, they don't care about fighting abuse, they are not concerned with long term vision of the platform, at least not enough to do anything.
As @geekgirl said, the downvotes were the main reason we no longer have bid bots, Haejin is not farming with 100% efficiency with 1.4M HP that isn't even his, and many other farming operations have been shut down.
As I have said many times to trost, there are far more malicious upvotes (poor judgement, bad quality, farming, favoritism, etc) than malicious downvotes, I'd wager something like 1000:1 or more. It's not really a problem, the few cases it happens the community can step and correct it (like downvotes, this unfortunately rarely happens due to the potential loss of curation rewards doing so).
View more
Thanks for the perspective and feedback.
I agree that free downvotes were needed and appropriate when they were instituted, and probably saved the platform from devolving into a Mad Max arena.
However, as a relative newcomer, I clearly see distinct downsides of the free downvote. It is a blunt instrument that has faithfully served its purpose, but needs to be replaced with a more precise tool. And that more precise tool must be experimentally developed and demonstrated on Layer 2.
That’s what I’m endeavoring to do with VYB. Hopefully other potential solutions will also be instantiated, with at least one proving itself worthy and effective.
I appreciate and respect your perspective and welcome your continued feedback.