If an investor wants to auto piolet their votes they can follow a trail, nothing can stop that ever, anyone can always follow someones votes. However, that isnt the point of attack, the point of attack is when the initial vote is cast. If an investor is following a trail of a manual curator, it makes the ones doing the work, the manual curators that get rewarded and investors still earn but just less because they cant front run.
With lazy autovoters, IE people who vote at 4min mark of all "popular" creators and just leaves it, that is destructive to the platform. We need to encourage better curation, which means rewarding manual voters. What is the best way to punish lazy voters and reward manual voters? That is the 100 billion dollar question IMO, if we get curation right, we explode in popularity.
If you make curation rewards more linear the more successful an author gets over time then it would encourage true content discovery. For example you could incorporate the reputation score of the author being voted on into the post curation rewards algorithm in a way such that curation of low reputation authors would be subject to a more exponential reward curve and curation of high reputation authors would have a more liner curation reward curve. This could stop the early vote stacking on historically popular authors as there would be minimal exponential advantage to voting early on those posts and shift the people chasing exponential rewards towards true content discovery of less known accounts and spread the votes over a wider distribution. I know reputation is not a perfect indicator after the bidbot years but it's somewhere to start. I guess the downside it would make self voting of accounts with high reputation slightly more profitable. But you would not have to make it completely linear just at a comparative disadvantage to the posts with lower reputations.
I have an idea as to how to fix that in the next hard fork. It's not too complicated and does not require too much coding or testing.
https://peakd.com/hive-136578/@markkujantunen/taxing-too-frequent-votes-on-the-same-authors
The issue with that proposal is if I discover a 'new' person and I want to reward a few posts at once while looking at their portfolio I would essentially get 'taxed' to do that. One would also have to keep track of everyone they voted for and when they could vote again and not be hit. This is complex and shifts the balance of power back to autovoting and away from manual curation , as you can schedule posts votes after a certain amount of time has lapsed.
I think we need to separate the "content discovery" issues which need to be dealt with by a front end algorithm that does not rank simplistically by HIVE reward, from the discussion about the perception of an unfair blockcahin reward pool distribution in a stake based system. Content discovery is abysmal. We need a proper way to search the platform, we need an open tag system like Instagram hashtags where we can see posts grouped into areas rather than the current 5-10 meaningless tags which are taken up with tribes and trending tags. We need algorithms that suggest trending users and posts based on spikes in community interaction .
HIVE is by definition a stake based system, should we really be discriminating on what people can vote on with their stake? Any top level token limitations will also have an effect on the development of other dapps etc who might use the reward pool. The real question is what are we trying to be here? Are we just going to specialize as a top level blogging site where the primary token is perfected to produce the best rank of posts using 'reward pool payout" as the ultimate form of post rank ? Or are we trying to be an open stake based Level 1 DPOS token without restrictions on the use of stake where people can build a variety of Web 3 DApps that can use their own level 2 token to rank posts (or anything else they might want to do like reward people for exercising, or providing the answer to question, or winning a level in a game, etc)?
Danny, how easy is to develop a Curator Reputation based on experience? I am bored to see the trending full of jerk circle abusers and manual curators receiving big upvotes as salary for their manual curation efforts.
If we are able to develop transparent curation metrics, we will be able to support those curators using a portion of the rewards pool or from a DAO proposal.
Curation by humans which scales with the platform, instead of the algos of Twitter and Facebook is a holy grail.
That is crushed by lazy autovoting.
Hive is close to getting it right but needs to be hyper careful with services like hive.vote. We all know that auto-voting is easy to code, but if the services aren't easy to use, fortunately, their use won't scale up.