A guide to curation incentives on the Steem blockchain

This post discusses some of the curation incentives that are associated with Steem account sizes, curation rewards, and the STEEM token value in order to help illuminate the path towards optimal curation strategies.

---

Experimental: Please leave a relevant and substantive reply to this post in order to be considered for a gratuity from @penny4thoughts after the post pays out.

---

Here on the Steem blockchain, from the smallest account to the largest, we all have the same boss. Namely, the blockchain itself. This arrangement can be challenging, because the blockchain is very limited in the ways it can tell us what it wants or expects from us. Basically, the blockchain's mechanism for communicating with us involves the use of author rewards, curation rewards, and the market value of STEEM. For purposes of this post, I'm going to ignore author rewards, and begin examining the ways that curation rewards and the market value of STEEM can incentivize curation accounts of various sizes.

It is my opinion that for Steem to ultimately succeed, it is necessary for Steem's curators to seize control of the trading price from the market traders and speculators. This only happens by making curation a more profitable endeavor than trading.

It may be important to understand these incentives because differing incentives may suggest that optimal curation strategies differ for accounts of varying sizes. Towards that end, I have created the following table mapping account sizes to the relative priority which those accounts place upon curation rewards and STEEM's market value.

Account		Priority
Class	Steem Power	Curation Rewards (TRX + SP)	Value of STEEM
Plankton	< 500	Low	Low
Minnow	500-5,000	Medium	Low
Dolphin	5,000-50,000	High	Medium
Orca	50,000-500,000	High	Medium
Whale	500,000-5,000,000	Medium	High
Blue Whale	> 5,000,000	Low	High

Note that this table just reflects my own opinion at the moment, and it's obviously an oversimplification. I welcome discussion about how to improve it.

Section 1 - Account Definitions, Priority levels, and the nature of curation

Account Definitions

I'm not aware of an "official" standard for account sizes, so I made some of these categories up. In short, I defined six categories: Plankton, Minnow, Dolphin, Orca, Whale, and Blue Whale where the account steps up in categories at 5 times the next power of ten. At the base, the Plankton category covers three orders of magnitude, and at the other end of the scale, the Blue Whale category covers a theoretically unlimited number. All others cover a single order of magnitude.

Priority Levels

Priorities are defined as the subjective values of {High, Medium, and Low}. This is oversimplified, but it's the sense that I have for the relative importance that the two reward mechanisms (curation rewards, and Steem's price) play when making voting decisions.

What is curation on the Steem blockchain?

Note that the rewards are called, "curation rewards", not "voting rewards". We often talk about voting and curation as interchangeable, but that's technically incorrect.

curation: the selection of something such as documents, music, or internet content to be included as part of a list or collection or on a website:

In my opinion, the reason the blockchain is offering curation rewards is because it wants a ranking of posts that will attract two things: attention and engagement. The blockchain wants authors to write content, and it wants curators to appraise the value of the content so that guests can easily find the most attractive and engaging material.

If curators do a good job, STEEM prices go up and the blockchain is happy. If curators do a bad job, traders take over, STEEM prices go down, and the blockchain is sad.

Now, it's important to note that there are two possible types of errors that curators can make when it comes to appraising something. We can undervalue it or overvalue it. If content is misranked in either direction, then it drives the audience away and it saddens our boss, the blockchain.

So with all that out of the way, let's move on and look at the curation incentives for the varying account sizes.

(Side note: the intended solution to overvalued content is the "downvote", but IMO, five years of practice shows that this is a failed mechanism. That's outside the scope of this post, however. For more about that, see my various posts on modeling curation rewards after second-price auctions.

Section 2 - Curation Incentives

In this section, I'll discuss the incentives for each of the account levels, and consider how the incentives drive voting decisions.

Plankton

Stake: 0-500 SP
Priority for curation rewards: Low
Priority for the price of STEEM: Low

Plankton-level accounts simply don't have much influence on the rankings, so the goal here is just to increase Steem Power either through markets or by creating content. The Plankton account owner can happily vote for anything they want, and if enough Plankton vote for something, it might come to the attention of a higher ranked voter.

For Plankton, I recommend avoiding downvotes because the risk of an anti-social "Flag War" is high, and the vote doesn't have any quantitative effect anyway.

Minnow

Stake: 500-5,000 SP
Priority for curation rewards: Medium
Priority for the price of STEEM: Low

It is now possible for you to begin earning some small curation rewards. In a fairly recent post, I observed that accounts of all sizes were receiving curation rewards in the range of 20-30%. You are dependent on other voters, however, to prop up the value of your STEEM so that your gains in curation rewards aren't swamped by loss of value for the token.

The reality here is that the Minnow is mainly interested in guessing which posts will get votes from higher ranked accounts and voting on those posts before the high-stakes accounts vote. As with the Plankton, if enough Minnows vote for a post, that post might gain the attention of a larger account.

Of course, we all want to see the value of STEEM rise, but minnows can't effect much change in that area, so it's low priority when making voting decisions.

For Minnows, I also recommend avoiding downvotes because the risk of an anti-social "Flag War" is high, and the vote still doesn't have much of a quantitative effect.

Dolphin

Stake: 5,000-50,000 SP
Priority for curation rewards: High
Priority for the price of STEEM: Medium

The dolphin is in a position to start gaining some nice curation rewards, so this is their highest voting priority. The dolphin still doesn't have much influence on the price of STEEM, but a pod of dolphins acting together can have influence here, so the priority that a dolphin places on price is "Medium" at this level.

As with Minnows, then, the Dolphin wants to vote on posts that will catch the attention of larger accounts. However, Dolphins should also begin to weigh the potential attention and engagement that a post might draw when casting their votes.

Dolphins may want to cast the occasional downvote on posts that are particularly egregious as spam or other abuse, but should err on the side of caution when deciding whether or not to downvote.

Orca

Stake: 50,000-500,000 SP
Priority for curation rewards: High
Priority for the price of STEEM: Medium

The Orca doesn't differ much from the Dolphin in voting priorities, but the balance begins to shift away from curation rewards and towards the value of the STEEM token. The Orca is beginning to have some serious influence on the overall ranking of posts, especially if a handful of Orcas agree on a voting decision.

Like the Dolphin, downvotes may sometimes be appropriate for the Orca, but conservative use of downvotes is probably warranted.

Whale

Stake: 500,000-5,000,000 SP
Priority for curation rewards: Medium
Priority for the price of STEEM: High

Before we talk about voting strategies for the remaining two groups, let's ask the question "what is a whale"?

Binance has this to say:

The term “whale” is used to describe an individual or organization that holds a large amount of a particular cryptocurrency. There is no exact cutoff threshold for this definition, but some say a Bitcoin whale should hold at least 1,000 BTC. A whale may also be defined as a person that has enough coins or tokens to cause a significant impact on the market prices, either by buying or selling large amounts.

Steem is fairly unique among blockchains, because a Steem whale can move markets not just by buying or selling cryptocurrencies, but also by staking and curating with it. This is why the incentive-balance for this level of account shifts even further from curation rewards towards token value. A Steem whale with half-a-million tokens today is worth less than $100,000, but at the all time high the same account would have been worth $4,000,000. To generate $4,000,000 in value now, from curation rewards, would take decades, but if the whale is able to use its voting power to increase the token value, that value might be realized much sooner.

What innovative things can a Whale do to try to make the token more (or less) attractive to investors?

• Shape the trending page, the first posts that visitors see when they arrive on the site. If those posts attract attention and engagement, the token may go up. If not, it may go down.
• Make sure that curator-investors are rewarded when they vote for attractive and engaging content. This might mean intentionally voting late, making votes easy to predict, and even announcing an upcoming vote in order to let other voters pile on first to claim curation rewards.
• Reinvesting curation rewards into tools and techniques that eventually lead to blockchain value growth due to better content discovery.

In summary, the Whale's best voting strategy may be to look for posts that the account owner believes will attract audience and engagement. Voting in ways that will drive the price of STEEM up is far more important than voting to generate curation rewards.

In keeping with the desire to avoid overvalued posts in the ranking, the Whale should consider the judicious and temperate use of downvotes, too. Whales may want to get more liberal with downvoting over rewards disagreement in order to help shape the curated list of posts in a way that will be pleasing to blockchain visitors.

Blue Whale

Stake: > 5,000,000 SP
Priority for curation rewards: Low
Priority for the price of STEEM: High

For the Blue Whale, the price of STEEM is of ultimate importance. Movement, up or down, in STEEM's price can easily dwarf the account's curation rewards. and the Blue Whale has the ability to influence the token value. Every vote, upvote and downvote, should be cast in order to "curate" - truly curate - the list of posts and comments.

Because the "Blue Whale" can't be harmed by flag wars, these accounts should feel free to downvote over reward disagreement, but all attempts should be made to avoid negative perceptions by authors, who could be driven away by the perception of undeserved downvotes.

Blue Whales are where we find the Steemit, Inc. accounts, and hopefully other commercial enterprises in the future. If I ran a Blue Whale account, I would consider taking one or more of the following steps:

1. Hire someone(s) as a dedicated manual curator
2. Develop AI and natural language tools to automate some or all aspects of curation
3. Pay manual curators and tool developers a bonus that's tied to STEEM's market value.

Section 3 - On downvotes and the use of bots

Two topics have been controversial since I first arrived on the Steem blockchain. These are the use of downvotes, and the use of bots. If we adopt this framework, it may offer some insights into both topics.

On downvotes

We work for our boss, the blockchain, and the blockchain has a downvote capability, so the boss is obviously not militantly against downvotes.

However, I think that the boss wants us to attract people for an audience and engagement, not to drive them away. We have seen many times in the past the great potential for harm that arises from overuse of downvotes.

So, as with any other form of defense, I suppose the position on downvotes is to use them with care and only when necessary. There is an imperative to try to use them in a way that does not alienate people who are not attacking or abusing the blockchain.

On the use of bots

I have argued since arriving here that the use of bots is just another way of doing the same thing that Google does with its search engine. It's an automated form of content discovery. I have also argued that a mix of human and automated voting is probably best. Those arguments remain consistent with this framework, and I still believe them, but we also see now that the answer might be different depending on the size of the account.

If a Whale or Blue Whale account has excellent AI or natural language capabilities, then it would make sense to use a bot, but as far as I know, no such thing exists yet for the Steem blockchain. With these accounts, the primary concern is the value of the STEEM token, and at present I think they can support that goal better with manual voting, perhaps supported by automated content search and filtering tools. As noted above, in the long term, it may be wise for these account owners to identify and automate excellent curation practices.

For accounts that are at the Orca level or smaller that are primarily chasing curation rewards, I think it makes sense to automate the search. If the Whales and Blue Whales are doing their job, then this will steer the other voters into voting for "the right" posts.

Conclusion

Of course, all of this is just my opinion based on 4 ¹/₂ years of observation, but I thought it would be worthwhile to continue the ongoing conversation about curation on the Steem blockchain. In my opinion, one of the key signs of ultimate success for Steem will be the day when the curator-investors wrench control of the price movement away from the traders and speculators. In order to get to that point, I think the community needs to develop an understanding of how Steem's incentives drive curators and how that, in turn, drives value and investment.

---

Related:

• The Ultimate Guide to Voting Power (with cartoons, formulas, and code references!) (@biophil)
• If you're voting open-loop, it's optimal to vote earlier than 30 minutes (2020-12-31: now 5 minutes) (@biophil)
• Discussing the factors that drive Steem's curation rewards
• Simulating a Steem curation rewards distribution that is modeled after a 2nd price auction
• If you want Steem's minnows to use their downvotes, protect them with quorum sensing
• A game theory requirement for steem downvotes has not been shown
• Down-votes: Steemit's Achilles' Heel?
• On bots

---

Beneficiaries

• Experimental - penny4thoughts - 36%
• Public Steem API - justyy - 5%
• Public Steem API - steemit - 5%
• SteemWorld.Org Tools for Steemizens - steemchiller - 5%

---

Thank you for your time and attention.

As a general rule, I up-vote comments that demonstrate "proof of reading".

---

Steve Palmer is an IT professional with three decades of professional experience in data communications and information systems. He holds a bachelor's degree in mathematics, a master's degree in computer science, and a master's degree in information systems and technology management. He has been awarded 3 US patents.