The "convergent linear rewards curve" (CLRC) part of the EIP worries me.
The idea of the change (as quoted from the article you link to) is:
"The scenario that it seeks to combat is a single entity spreading their stake over many small accounts in order to hide their actions in noise. A single entity doing this can hide their intentions and siphon off small rewards over time and be difficult to detect."
The CLRC achieves this aim by reducing the rewards for low earning posts / comments. Those rewards will be spread over all posts, increasing medium-higher earning posts by a small margin.
The issue is that the curve cannot differentiate between "a single entity spreading their stake over many small accounts" and actual small accounts. Since 70-75% of accounts earn less than 1 Steem per month, this is a pretty big issue.
I remember when I started. My early posts earned between 2c and 20c. I tried to contribute through good comments with rewards of the same level. I think most accounts start off in this manner.
In a pro / con analysis, the big cons are that:
(a) It reduces earnings for new / small accounts, thus harming mass adoption.
(b) It harms engagement by reducing the incentive to make great comments.
The pro is that it stops abuse from large accounts hiding their activity through dividing into many small accounts. But what is the reward for doing this? The reward under the current linear system is 100% of upvote. The same reward as circular voting, and very similar to delegating to a vote-bot. The economic incentive to do this is low.
I think that:
(a) The cons significant outweigh the pros.
(b) The original issue looking to be solved could be tackled through other avenues, such as using data analysis through MIRA to bring such accounts to light.
Does your opinion change if SMTs become the earning currency on applications rather than Steem directly? Would this incentivize applications to power up and be able to benefit more since there is less top-end abuse?
I like to think of SMTs as linked to dApps but also to Communities, with potentially many Communities having their own SMT reward token. Allowing different reward curves for each SMT would be very useful, allowing each dApp / Community flexibility in what they are trying to incentivise / achieve.
The problem for Steem is that it is trying to be many things at once. We want to find and reward the best content (i.e. content creation, a bit like youtube?) but we also want to incentivise mass adoption with many users and lots of engagement (i.e. social network, a bit like facebook?). The CLRC works towards the former (by aiming to remove some abuse, thus making it more likely rewards head to good content) but at the expense of the latter.
A Community / dApp that was only interested in rewarding the best content, or which wanted some form of one-person-one-vote functionality in addition to a dPoS vote functionality (i.e. a reason to remove sybil issues) would probably find the CLRC very useful.
Would this be a bad thing at this stage of the game considering while any want mass adoption now, are we actually ready for it?
Definitely.
I think this would be very useful for continued development of many kinds.
I think that if the idea of the EIP is focus on good content but to sacrifice mass adoption efforts / comments and engagement / the idea of Steem as a social network then that should be made very clear in the proposal. That would be a pretty big pivot.
In my view it's a bad move.
I don't think that is the idea, by the way, I think that solutions have been sought to a particular set of problems but without fully considering the other problems that could arise.
Also in general on abuse I think that if you aim to prevent people obtaining 90%-100% of vote value on things like:
Then accounts will obtain 90%-100% of vote value through:
In the end a vote has value and people will find ingenious ways to extract that value to its fullest extent. The harder it is to do just increases the marginal cut to middle-men for its organisation.
Every system is gameable in some way isn't it?
Yes! This is one of the reasons I like flat curation with the curation slider at the discretion of the voter. It removes all the incentives and just lets people vote for the content they like.
That is an experiement I have wanted to have run for 1.5 years too but I think it has less chance of happening.