Steemit Proposal for Developer and Community Evaluation

Response to @dantheman original post seeking input on how to improve Steemit:

People Rank - Using Page Rank Algorithm for Better Curation and Rewards  

Preface 

The current Steemit system, although working very well, possesses some design deficiencies, limitations, and is not as representative as intended as it scales to larger audiences.  

@dantheman – Steemit Dev, proposed some changes and opened up the conversation to users for their ideas. The following is just that. A user submitted proposal to be use in part or whole for whatever purposes that support the improvement of the Steemit platform.  

This proposal will briefly touch on the Problems, cover Design Principles which should not be undermined, and define the Goals this proposal is intending to address.   A main Proposal Overview will include Examples.  Changes to the system as well as Mechanism will be outlines and a Final Experience will be detailed.  A secondary and tertiary Proposal Overviews will cover ancillary areas.  Conclusion will give my final thoughts and a Glossary is at the end to cover terminology. 

Problem 

Persistent challenges:

• People will attempt to game the system for their benefit
• Everyone wants to “vote selfishly” to give themselves the most reward possible
• Large stake holders want to vote in a manner that maximizes the value of the currency
• Accounts can go rogue, get hacked, or otherwise start acting in ways that harm the wider community

Situation:

• Large stakeholders have enough stake to unilaterally allocate far more than even the largest group of smaller users
• When a small stakeholder goes bad the system can safely ignore their votes and hide their posts with a reputation score, but when a whale goes bad things are not so easy

Result:

• Whales need to spread their votes across 1000 times as much content as normal users or refrain from voting all together
• Need is a scalable solution for managing distributed authority over content generation and deciding which is most worthy to be promoted

Steemit Design Principles

• Individuals acting alone shouldn’t have much power, but the more people that work together the more power they have collectively. i.e. two people voting together is more than twice as powerful as either voting alone.
• The root of all authority must be derived from stakeholder Steem Power 
• System must remain scalable for millions of users

Goals 

• Retain the value principle and current structures of Steem Power as a mechanism of 'mass' to move rewards 
• Encourage a steady stream of good content creation by ever more of the community 
• Identify and reward the best of the content, regardless of topic, author, or self-interest reward 
• Identify people as domain experts, trustworthy, and valuable contributors as part of a reputation system 
• Not overly burden the system with new complexities or new measures of value 
• Promote active and regular participation by all Steemit members (reading, voting, etc.) but not allow for dilution via too many votes by bots or mass-voting tactics. I.e. vote power controls must remain in place. 
• Give no advantage to those who would create multiple accounts to vote. Kill-off new accounts, after a period of time, that do not contribute to content. (unused accounts or accounts just used as part of an upvote bot-army) 
• Empower all accounts in good standing (even new accounts) to have some measurable meaning to voting and participation in a system which rewards good behaviors (the act of just voting) with more weight (outside of the SP and SD system - don't want to muck with that if at all possible) that users can witness benefits to their participation 
• Tie downvoting to expertise of the voter and provide automated specific weights as to the reason 
• Empower whales to more easily facilitate and oversee dispersion of their influence to trusted parties (ie. like @smooth) and give them the ability to designate topic experts 
• Make allocation of steem power a voter-controlled variable to facilitate scale-ability of up-votes (from minnows to whales) Ex. so whales don't have to upvote hundreds every day to disperse their value to the system.  

Proposal Overview 

Part 1 – All votes are meaningful (enough minnows can overtake a whale), drive dispersion of Steem across diverse content (better curation), tunable allocation amounts (ease for multiple upvotes)

• Steem Power remain the base for assigning value to content. The total allocation at any moment is shown to the user
• Vote Power % is replaced with a pool amount that is ‘spent’ when upvoting. The base limiters of Steem Power x Vote Power remain, to restrict unlimited voting. The quantity of available power is SpVp
• Users can upvote Content with a slider, allowing the tuning of how much SpVp is assigned to an upvote. Min is .01/Max can be set by Dev’s for any single post
• However, SpVp is split among various Categories, so the user is limited to spending only what is available for certain types of content 
• Once the SpVp is depleted for a category, no more upvotes for that category can take place until SpVp is restored over time (like Vote Power currently does).
• Allocation of SpVp is correlated to the value increase of a post. If user upvotes a single piece of content with .1 SpVp, the value of that content rises exactly .1 SpVp
• Allocation of SpVp results in more steem or faster recovery time for users (algorithm up to Dev’s), promoting curating/up-voting

Example 

• @dan has 100 Steem Power and his vote power is at 90%. @dan SpVp is 90
• There exist 10 defined Categories. Therefore @dan has 9 SpVp to allocate in upvoting actions in each of the Categories. (SpVp / 10 category)
• Dev’s set a max limit of 5 SpVp for any single upvote (could be variable based upon user net-worth). Min limit is .01 SpVp for any upvote. 
• @dan can choose to upvote in the News category and with the slider assign anywhere from .01 to 5 SpVp, which is then withdrawn from his SpVp for the News category, but not affecting the SpVp for other categories
• When @dan is at less than .01 SpVp he can view posts in the category but not upvote them until SpVp is restored, which will occur normally over time 
• @dan can choose to distribute SpVp or not, in other categories
• As @dan votes, he receives a benefit. Either a steem reward at some ratio (allocate 10.00 SpVp in votes = .01 reward) or a reduced cool-down time for his SpVp to return (so he can vote faster in the future) 

Changes to System

• Categories to be defined (by dev’s) and are aligned to tags. Ex. A new category of Art is defined and all posts tagged with Stories, Pictures Drawings etc. are aligned
• A new variable SpVp is created to show each user how much they may allocate, for each category
• All new users start with.01 (the smallest amount of SpVp allowable for upvoting)

Mechanisms 

• New users begin with only .01 SpVp, therefore can only vote once until their SpVp is restored
• Voting results in benefits to users (either steem rewards or shorter recovery time of SpVp)
• Whales can vote on many posts, with variable amounts per post, and are compelled to vote across all the Categories if they wish to spend all their SpVp over time.
• Minnows can vote, which makes a difference, but cannot vote as much as Whales, because they don’t have the Steem Power
• SpVp must be tracked for each category for each user (this may be burdensome, haven’t thought about it yet…)
• Even if users self-promote (vote for their stuff), the overall negative impact is marginalized because the categories limit exclusive voting (unless they have content in every category, which is an offset in itself)
• As a counter to granting all new accounts .01 SpVp, a normal process must run to purge accounts which are dormant or do not post any Content over a period of time (probation), to reduce risks of farms spawning accounts to conduct mass upvoting

Final Experience

• All level of players can vote and carry weight
• Steem Power will determine the overall SpVp resource which can be allocated by upvoting
• The more Steem Power you have, the more you can vote and the more each vote can count
• Introduction of Categories, puts a structure in place to drive more diversity of upvotes and promotes users to curate content topics they may normally not view, but get benefits for doing so 
• With the ability to upvote becoming a fixed resource, careful curation is promoted
• Rewards for voting will encourage full participation by users and persistence over time

Proposal Overview

Part 2 – Reputation as a system of down-vote expertise

• Upvotes in a Category will then add to the reputation of the Content author for that Category
• Category reputation then becomes an indicator of expertise for those subjects (which would be cool for ranking of content as well)
• Reputation then plays a role in down-votes, as those with expertise can spot harmful content. 
• Those with high reputation (experts) in a Category will then be compelled to manage ‘their’ domains, promoting better down-vote curation
• Option: Enough downvotes by experts (number threshold or % for quorum) and the content is removed 
• Option: Enough upvotes by experts and the content is stickied (remains longer on site, but no voting)

Proposal Overview

Part 3 – Delegation of SpVp by Whales

• Whales can delegate SpVp to others or other accounts which can specialize in a topic, ie Category
• This allows for subject matter experts to dole out SpVp
• Example: @smooth hires a News category delegate which will focus on being an expert for that content and upvote using the News SpVp granted by the @smooth

Conclusion 

• I believe the Steemit platform can be improved while retaining all the core principles
• Enhancements can reinforce good behaviors and empower all users to affect the value ratings of content
• Steemit is worth the time it took for me to write up my crazy ideas and for you to read them!

Vocabulary 

• Content = blogs, stories, replys, etc. created by Steemit users
• Categories = Higher order (macro) descriptions which tags (micro) are be associated with. One to many can exist if Devs allow
• Tags = current keywords assigned to blogs/stories. First Tag is the primary sorting variable. Ex. news, life, introduceyourself
• SpVp = Steem $ value available to a user to vote within a category. SpVp’s are tracked for each category, for each user
• Whales = users with an abundance of Steem Power
• Minnow = users with very little Steem Power, ex. a new user