Vox Metrics — DAO-controlled fair value scoring (Roadmap update for Polkassembly)

Vox Metrics is now live with an open API, a dashboard, and universal weights that apply fairly across all supported networks. All scoring parameters are DAO-controlled: the community can preview proposed changes, see precisely how scores would shift, and then adopt them by updating the on-chain settings pointer (or controls.json in the reference build). The aim is a transparent, on-chain alternative to opaque, centralised macroeconomic yardsticks.

What this is (for new and returning readers)

Vox Metrics produces a single Value/Trust Score (VTS) in [0,1] for each network, built from five normalised pillars:

PoV — Protocol of Value (throughput & resilience)
Sub-metrics: transactions, validator/effective set, uptime
OPI — On-chain Participation & Integrity (governance health)
Sub-metrics: voter turnout, stake ratio, proposal cadence
DLI — DeFi Liquidity (depth & efficiency)
Sub-metrics: TVL, DEX volume, slippage (inverted)
PTI — Policy/Transparency & Security (project hygiene)
Sub-metrics: transparency, security, sentiment
RWAI — Real-World Assets Index (tokenised RWA activity)
Sub-metrics: value locked, trades, diversity

All subscores are mapped to [0,1] on published scales, then rolled up with weights. The final VTS is reproducible, comes with a hash of inputs, and includes the controls version used.

Why DAO-controlled?

Fairness by design: one set of weights applied to every network in strict policy, so no chain is advantaged by bespoke tuning.
Public, predictable governance: proposed changes are previewed (no side-effects), discussed, then adopted by vote.
No black boxes: the full recipe (weights, ranges, formulas, sources) is visible; anyone can reproduce results.
Auditability: each response returns a reproducibility hash and the exact controls version used.
Neutrality: parameters are not set by a single company or committee; they are set by token-holders in the open.

Current state: the reference implementation reads from controls.json. The next governance step is to point the service at a DAO-managed location (e.g. IPFS/Arweave/Git tag) so an approved proposal updates a single pointer that all deployments can follow.

Why move away from centralised macroeconomic data?

Centralised macro data (CPI, GDP prints, central bank guidance) has three practical problems for crypto networks:

Opacity and lag: releases are scheduled, revised, and politically framed. They are slow, get revised after the fact, and the method is rarely transparent enough to audit line-by-line.
Jurisdiction bias: macro signals reflect specific economies and policy choices. They are not neutral for borderless, permissionless networks.
Single-source risk: markets can be whipsawed by one publication or a handful of institutions. That’s fragile.

Vox Metrics is built to be on-chain, transparent, and always-on:

Data are drawn from openly queryable sources (explorers/indexers/DEX APIs).
Normalisation and weighting are public and reproducible.
Method is portable across networks; comparisons are like-for-like.
Governance is collective and inspectable, not managerial.

This is not a macro substitute for national economies; it is a fair, network-native compass for crypto systems.

What’s live today

API & docs: /docs (OpenAPI).
Dashboard: /dashboard with a network selector, compare networks, all networks table, and a controls sandbox.
Universal weights: default policy is strict (same weights for every chain).
DAO-tunable sub-metrics: sub-weights per pillar can be adjusted by governance.
Major chains supported: Polkadot plus Ethereum, Polygon, BSC, Arbitrum, Optimism, Base, Avalanche, Fantom.
Self-test & diagnostics: range checks, normalisation checks, and reproducibility validation.
Reproducibility hash: every response includes a deterministic hashOfInputs.

Roadmap status

Phase 1 — Foundation ✅
Scoring engine, normalisation, hash-of-inputs, dashboard, strict universal weights.

Phase 2 — Multi-chain (first wave) ✅
Polkadot + major EVMs with consistent scaling.

Phase 3 — Governance tooling ▶ In progress

Preview pathway done (no write).
Next: DAO-managed controls pointer (IPFS/Arweave/Git).
Optional: guarded “apply controls” endpoint for authorised deployments.

Phase 4 — Data breadth & resilience ▶ In progress
More sources per chain, caching, graceful fallbacks.

Phase 5 — Polishing ▶ Next
CSV export, policy badges in UI (strict/adaptive), light anomaly alerts.

How governance changes the score (in practice)

Propose new weights/ranges in the thread.
Preview the impact (sandbox or POST /api/v1/controls/preview) to produce before/after tables.
Discuss and refine.
Adopt via vote; update the DAO pointer to the new controls version.
Verify: responses show the new controlsVersion and a fresh hashOfInputs.

By default, we recommend staying in strict policy (universal weights). An adaptive policy is available if the DAO agrees to rule-based, small shifts (e.g. temporarily down-weight RWAI where it is still maturing).

Useful endpoints:

GET /api/v1/networks — list available networks
GET /api/v1/{network}/metrics/all — full breakdown for one network
GET /api/v1/metrics/all_networks — table across all networks
GET /api/v1/weights — current weights (read-only)
GET /api/v1/selftest — system check
POST /api/v1/controls/preview — try new weights without saving

Call to action

Review the scales and weights. If you believe PoV should emphasise validator diversity more than raw throughput, propose new sub-weights.
Suggest sources. If there’s a reliable indexer or DEX dataset we should add for any chain, please share.
Bring a proposal. Use the preview endpoint to produce a clear before/after table and submit it for a vote.
Hold us to account. If anything is unclear or not reproducible, say so. This system only earns trust by being legible.

Our goal is simple: a fair, open, and genuinely useful way to compare networks that doesn’t depend on central banks, national statistics, or opaque rating shops. If that sounds right to you, help us tune it.
Screenshot 2025-09-15 101645.png Screenshot 2025-09-15 101510.png