It's been a little while since our last large-scale update, so over the next few days we'll be sharing a short series of posts covering what our development team has been building across the Hive ecosystem over the past month or so. We also shared these updates at the recent monthly Hive core developer meeting — this series expands on them with the full detail and live links you can try yourself. This first post covers our core-project work: a new network-analytics layer for HAF, and the Block Explorer dashboards built on top of it.
The work in this post spanned these Hive projects:
- haf_stats: a HAF analytics app that turns raw chain data into chart-ready network statistics and per-account metrics.
- Block Explorer (UI): the block explorer interface. We run a live instance at hivescan.info.
- Balance Tracker: HAF app providing endpoints for user account balances on Hive.
- HAF Block Explorer: HAF app providing information about transactions/operations in Hive blocks.
- Hivemind: Hive's social API layer.
- Health Checker: a React component for checking, fetching and switching between Hive nodes.
Everything below is live on our self-hosted HAF API node, kept continuously at develop-HEAD with the full stack deployed. The Block Explorer interface is at hivescan.info, backed by the node's APIs at testapi.hivescan.info — so most items include a Try it here link you can click right now.
The major areas of our work involved:
- A new network-analytics endpoint suite. A full family of read-only statistics endpoints built on HAF: daily active users, network vote stats, content volume, engagement, per-account content stats, and a financial summary. Each turns raw chain data into chart-ready time-series, and every value is verified against a real 5M-block replay so it matches mainnet exactly. (Try it: daily active users, vote stats)
- A network Top-Accounts leaderboard. Rank accounts across the whole chain by activity, financials and more (Try it: top accounts).
- Governance & wealth metrics. A governance influence-concentration endpoint reporting the Gini coefficient — a 0-to-1 measure of stake inequality (0 = everyone holds an equal share, 1 = a single account holds it all) — alongside top-1% / top-10% and median / mean shares (Try it: influence concentration). An HP-distribution breakdown then splits the network into wealth buckets with real totals and percentages, not midpoint estimates (Try it: HP distribution).
- A full Resource-Credit analytics system. Network RC utilization and per-account RC footprint, priced at read time from a live sampler of per-operation RC rates. custom_json operations are resolved to human DApp names across a 9-category registry, feeding a per-account DApp-footprint endpoint — so "who is burning RCs, and on what" is finally legible (Try it: RC footprint, DApp footprint).
- Cohort & retention analytics. An author-retention endpoint that follows whether creators keep posting after they join, and an account-funnel endpoint measuring 7/30/90-day activity milestones per signup cohort — with a historical backfill so the charts read correctly all the way back to genesis, not just forward from deploy day.
- A new analytics dashboard in the Block Explorer. All of the above wired into the explorer's home and reports pages as 8 new dashboard widgets — Daily Active Users & Operations, Voting Activity, Network HP Distribution, the Top-Accounts leaderboard, an Account-Retention Funnel heatmap, and Operations-by-Type — each a compact home card plus an expandable detailed view with its own charts, KPIs and range filters.
- HP Momentum & Power Activity tab. A new Power Activity tab with HP-momentum cards tracking power-up and power-down flows over time, with correct amounts, banner rates and a synced custom date-range filter.
- Proposals Votes-History modal. A dedicated modal for any proposal's voting history — time-series chart, KPI strip, filters and ASC/DESC sorting.
- Revamped front-page widgets. Recent Activity (per-category sizing + rolling 48-hour operation counts) and LastBlocks (live highlighting + slide-in animation, resize-aware and reduced-motion-friendly), plus a Transfer-Volume card that now handles NAI amount types with a HIVE/HBD selector and hourly granularity.
- Extended balance & explorer endpoints. Balance Tracker's top-holders endpoint gained optional min/max-balance range filtering for any coin type, while preserving each holder's true global rank (Try it: top holders). Liquid HBD-interest accrual is now tracked block-by-block, surfaced as pending HBD interest on the HAF Block Explorer's account endpoint so wallets can show interest accrued but not yet claimed.
- Hivemind refinements. Pending author and curation rewards are now exposed as explicit reward-basis buckets instead of a single flat value, and a mention-notification no longer renders the ungrammatical "mentioned you and 0 others."
- Better charts & 12-language support. Trend math was upgraded from naive 2-point deltas to linear-regression trend lines across the widgets, and the entire interface — new dashboards included — is delivered across 12 locales, with Arabic-digit and chart-locale reactivity so charts re-render correctly on a language switch.
Some screenshots below from hivescan
For the more technical
This cycle the team merged roughly 80 MRs across these repositories, shipping 9 new API endpoints and giving 10 endpoints a major performance rework as their datasets scaled to full mainnet volume. A sustained optimization program moved the timeout-prone endpoints onto processor-maintained rollups and served the top-accounts leaderboard from monthly + all-time pre-aggregates — cutting its all-history query from ~11 seconds to 0.09 seconds. That work also let us standardize what an omitted date range means: every get_network_* endpoint now defaults to full chain history (matching the explorer's "No Filter" option) instead of silently capping at 30 days, while still returning well within the gateway timeout.
The full set of merged work is browsable in each repo's merge-request history — haf_stats, block_explorer_ui, balance_tracker, haf_block_explorer and hivemind.
This is the first in a short series. Next up: a post about HiveFY — our new algorithmic "For You" discovery layer for Hive, and the app we built on top of it (and many other projects/updates in following posts). We look forward to contributing more to the Hive ecosystem and will share the next update in a couple of days.
HDev Team
@actifit on Hive — vote for us or set us as proxy on actifit profile, or via peakd, ecency, hive blog or hivesigner.
@actifit-he on Hive-engine — vote for us on Tribaldex.