As you may know, a group of us old Hiveans have been brainstorming ways to influence the culture of the platform. Today, the Jabber Talkie team is releasing a very special podcast featuring @moeknows, focused on the topic of reputation score. But it was this very conversation that sparked a clearer vision—a way to gamify the Hive experience and put our fingers on the scale when it comes to encouraging desirable user behaviors.
I believe the credit for the visual concept goes to @buttcoins. The idea is that the Hive frontends—whether it's PeakD, Ecency, or any other—would display not just your avatar and the current number we mistakenly call "reputation," but instead, a flower, with each petal representing a different score. I loved that initial idea. But my dear wife suggested—and I think she was spot on—that a honeycomb structure would make even more sense, keeping the theme intact.
The breakdown could look like this:
- A hexagon for Experience (the current reputation)
- A hexagon for Reputation (as proposed by @moeknows)
- A hexagon for Voting Habits
- A hexagon for KE Ratio
- A hexagon for Comment/Post Ratio
I think the idea can be easily expanded. Adding a hexagon feels... organic, if I may say so. Of course, people contribute to Hive in many ways beyond posting, commenting, or curating—but ideally, a revamped reputation system would account for those contributions too.
Today, however, I'm going to attempt something I've never done here before. I’m a mediocre coder at best, but in my younger years, I built a few things I was proud of. That background still helps me think like a programmer, even if my Hive coding journey has mostly been limited to little scripts here and there.
This concept of a curation score is heavily inspired by recent conversations I’ve had with @acidyo. Again, my goal isn’t to burn down the house and build a new one. What I’m trying to do is encode the unwritten rules—the behaviors we old Hiveans recognize as good, but that newcomers often struggle to identify and internalize.
Not long ago, I shared a philosophically stained post about Hivean bias—a play on the idea of survivor bias. And I’d invite anyone who’s quick to dismiss the ideas we’re proposing to do some introspection on the topic. It doesn’t matter if you or I already know what a good Hivean acts like. These ideas—this gamification—aren’t for us. They’re for those still finding their way. They’re for the ones who need a map.
With all this in mind, allow me share a White Paper of a kind, I've been working on this week. It addresses specifically Curation Score, but as I've said, this can tie easily to our broader idea of building the concept of Hiveology. The statistics that show us clearly, desired and undesired user behaviors.
I worked on a little mockup this morning of how it might look. I think it will help us all get on the same page- at least for this conversation.
A Curation‑Quality Score for Hive
1. Introduction
Hive’s “proof of brain” ethos rewards those who discover, foster, and amplify high‑quality content. Yet it can be hard to distinguish between genuine curators and vote rings that simply recycle rewards among a small circle. We propose a simple, transparent Curation‑Quality Score (CQS) — a 1–10 metric that captures how broadly, fairly, and independently a user curates content over a rolling 7‑day window.
2. Design Goals
- Encourage breadth Reward users who seek out new voices, up to a healthy social limit (≈ 50).
- Incentivize outreach Credit vote weight cast beyond one’s core circle.
- Discourage self‑voting bias Penalize heavy self‑voting that inflates rewards.
- Keep it simple Three sub‑scores combined into a single 1–10 rating.
- Transparent & on‑chain All inputs are publicly available via Hive APIs.
3. Definitions (7‑Day Window)
| Symbol | Definition |
|---|---|
| Wₜₒₜ | Total vote‑weight cast by user (sum of all vote percentages) |
| U | Number of unique authors the user voted for |
| W₅₀ | Sum of vote‑weight directed to the user’s top 50 authors |
| Wₛ | Vote‑weight the user cast on themselves |
4. Sub‑Scores
Breadth (B)
[
B = \min!\bigl(\tfrac{U}{50},,1.0\bigr)
]
– 0 if you vote 0 authors; 1 if you vote ≥ 50 distinct authors.Outreach (O)
[
O = \frac{W_{\rm tot} - W_{50}}{W_{\rm tot}}
]
– Fraction of your weight going outside your top 50 recipients.Anti‑Self (S)
[
S = 1 ;-;\frac{W_s}{W_{\rm tot}}
]
– 1 if you never self‑vote; declines toward 0 as self‑voting increases.
Each sub‑score ranges from 0 to 1.
5. Composite Score
- RawScore = (B + O + S) ÷ 3
- Curation‑Quality Score (1–10)
[
\text{CQS} = \bigl\lceil ,\text{RawScore} \times 9 \bigr\rceil + 1
]
– RawScore = 0 → CQS = 1; RawScore = 1 → CQS = 10.
6. Example Calculations
| Curator | U | W₅₀/Wₜₒₜ | Wₛ/Wₜₒₜ | B | O | S | RawScore | CQS |
|---|---|---|---|---|---|---|---|---|
| A (narrow) | 10 | 100 % | 20 % | 0.2 | 0.00 | 0.80 | 0.33 | 3 |
| B (core‑only) | 50 | 100 % | 0 % | 1.0 | 0.00 | 1.00 | 0.67 | 7 |
| C (diverse) | 100 | 25 % | 0 % | 1.0 | 0.75 | 1.00 | 0.92 | 10 |
| D (self‑heavy) | 30 | 60 % | 40 % | 0.6 | 0.40 | 0.50 | 0.50 | 6 |
7. Implementation Steps
- Data retrieval
- Use Hive’s history API to fetch a user’s votes over the last 7 days.
- Compute inputs
- Sum vote weights (Wₜₒₜ), tally weights per author, count U, identify top 50 & sum W₅₀, measure Wₛ.
- Calculate sub‑scores & CQS
- Apply formulas from Sections 4–5.
- Display
- Show “Curation‑Quality Score: X / 10” on user profiles in PeakD, with optional mini‑chart for B, O, S.
8. Benefits to PeakD & Hive
- Enhanced discovery: Spotlight users who genuinely explore beyond their circles.
- Community trust: Expose collusion patterns and promote fair curation.
- Gamified improvement: Encourage curators to raise their own CQS through broader engagement.
- Aligns with Proof of Brain: Reinforces Hive’s founding principle—reward effort and quality, not just reciprocation.
9. Next Steps & Feedback
We invite the PeakD developer community and Hive curators to:
- Review the proposed formulas and suggest refinements.
- Prototype a UI mock‑up for displaying the Curation‑Quality Score.
- Test with a small group of volunteer users.
- Iterate based on real‑world data and community feedback.
Together, let’s strengthen Hive’s curation economy and celebrate those who truly seek out fresh voices.
—
End of Proposal
Before I go, let me say
I'm giving this idea to Hive—to all of you. I’m not planning to submit a proposal or seek DHF funding for it. My motivation is simple: a rising tide lifts all ships. That’s really all there is to it.
I honestly don’t know if the PeakD or Ecency teams would be interested in implementing something like this. But because I’m passionate about these ideas—as are all of us on the Jabber Talkie team—I’m more than willing to discuss, debate, and fine-tune them if needed.
Anyway, I hope that, at the very least, I’ve inspired someone to think more deeply about this concept of behavior gamification. Just imagine how much better this place could be if we all aimed to look our best—for the Hivean picture.
Much love,
MenO