BSAHI — First Non-SCCR Empirical Governance Observation Rank 2 of the BSAHI plan · Produced: 2026-09-10 · Author: Prateek Poswal (Independent Researcher, ORCID 0009-0005-2139-1877)
| Event | Height | Date | Source |
|---|---|---|---|
| Mandatory signaling window opens | 961632 | ~2026-07-20 | data/bip110.json, research/governance-boundary.md |
| Mandatory signaling window ends | 963647 | ~2026-08-22 | data/bip110.json |
| Lock-in | 963648 | 2026-08-23 00:48:47 UTC | Blockstream.info API (/blocks/963648) |
| Data snapshot (post-lock-in) | 966270 | 2026-09-09T21:32:30Z | data/bip110.json (GitHub Actions) |
| Blocks since lock-in at snapshot | 2622 | ~18 days | Computed: 966270 − 963648 |
Verified from blockstream.info API (primary): Block 963648 exists at height 963648 on the main chain. Block header: version 1073676288 (0x40028000), timestamp 1787446127 (2026-08-23 00:48:47 UTC), tx_count 4370, size 1568900 bytes, weight 3993884.
Verified from data/bip110.json (GitHub Actions, mempool.space): window.lockIn = 963648, window.passedLockIn = true, currentHeight = 966270, thresholdPct = 55, windowSignaling = 0.
Signaling percentage at lock-in: 0%. The last 10 sampled blocks (966261–966270) all have bit4: false. The daily series (Aug 10–Sep 9) shows 0% signaling every day.
BIP-110 (Reduced Data Temporary Softfork) is a User Activated Soft Fork that introduces consensus-level restrictions on data-bearing transaction constructions for approximately one year. Specifically:
OP_RETURN sizes and NOP sequences at the consensus levelscriptPubKey sizes and witness item sizesKey distinction from Core v30: Bitcoin Core v30 changed datacarriersize from 80 to 100,000 bytes — this is a relay policy change (non-consensus). BIP-110 changes block validity rules (consensus-level). Blocks remain valid regardless of individual node relay policy under Core v30; under BIP-110, certain constructions become consensus-invalid for upgraded nodes.
Deployment mechanism: BIP-110 uses a height-based timeout with NO_TIMEOUT flag. The mandatory-signaling window is blocks 961632–963647 (2016 blocks). Lock-in is guaranteed at height 963648 regardless of miner signaling percentage, because the enforcement clause is carried by node software, not mining pools.
Finding: BIP-110 is enforced by node software at a predetermined height, not by miner signaling.
BIP-110 is a UASF (User Activated Soft Fork). The enforcement mechanism is:
Post-lock-in data confirms this: at height 966270 (18 days post-lock-in), all sampled blocks continue to be produced normally. The version fields show various values (0x20000000, 0x20010000, 0x20912000, etc.) — none have bit 4 set. This is expected because BIP-110 does not use bit-4 signaling for enforcement; the activation is height-triggered.
What BSAHI can observe:
What BSAHI cannot observe from this data alone:
Finding: Miner signaling for BIP-110 was 0% during the mandatory window. Lock-in occurred anyway.
The data/bip110.json and data/bip110_daily.json files record zero bit-4 signaling across the entire observation period. The blockstream.info API confirms that blocks around the lock-in height (963648) do not have bit 4 set in their version fields.
This is the critical governance observation: a UASF can activate a consensus-level change with zero miner support. The 55% threshold (1109/2016 blocks) was the designed signaling threshold, but BIP-110's NO_TIMEOUT + height-based activation mechanism means the soft fork locks in regardless. The signaling threshold is a coordination signal, not a requirement.
Comparison to calibration case (BIP-148, 2017): BIP-148 (SegWit UASF) also used a flag-day enforcement mechanism. Miners pre-emptively complied via BIP-91, and SegWit activated without a split. BIP-110's case is different: miners did not signal, yet the deployment activated by design. The threat of enforcement changed behavior in 2017; in 2026, enforcement was guaranteed by mechanism regardless of behavior.
What BSAHI can measure:
What BSAHI cannot measure:
Finding: Chain production continued normally through and after BIP-110 lock-in. No orphaned blocks or measurable disruption.
From the blockstream.info API for block 963648 (the lock-in block):
Post-lock-in daily data shows 55–161 blocks per day, consistent with normal Bitcoin block production (~144/day average). The block_interval.json data shows intervals ranging from 10s to 1736s, with averages around 400–628s — within expected variance.
What BSAHI can measure:
What BSAHI cannot measure:
Finding: Transaction economics show no anomalous disruption at lock-in, but the data gap is structural.
From data/fee_history_blocks.json, the fee data around the lock-in period shows normal fee dynamics: blocks with average fees ranging from 240,513 sats (~$78,449) and varying patterns. The SCCR (data/sccr_latest.json) shows avg_sccr = 0.371939 (0.37x coverage), with 95.62% of blocks below 1× storage cost.
The key observation is what did NOT change: BIP-110's consensus restrictions on data-bearing transactions did not produce an observable fee spike, block size anomaly, or transaction pattern shift in the data BSAHI captures. This is consistent with BIP-110's design as a temporary (~1 year) restriction — the restricted transaction constructions may represent a small fraction of block space, and miners may have already adapted their transaction selection policies.
Comparison to SCCR context: The SCCR baseline (L_net = $5,627.80/block at N=32K) provides the economic frame. BIP-110's activation did not visibly alter the fee-to-storage ratio or block space demand patterns in the observed data.
What BSAHI can measure:
What BSAHI cannot measure:
research/governance-boundary.md| Metric | Value | Source | Frequency |
|---|---|---|---|
| Block height at lock-in | 963648 | Blockstream API | One-time |
| Lock-in timestamp | 2026-08-23 00:48:47 UTC | Blockstream API | One-time |
| Blocks since lock-in (at snapshot) | 2622 | Computed | Continuous |
| Bit-4 signaling share (post-lock-in) | 0% | data/bip110.json | Daily |
| Bit-4 signaling share (during window) | 0% | data/bip110_daily.json | Daily |
| Blocks per day (post-lock-in) | 55–161 | data/bip110_daily.json | Daily |
| Block size at lock-in | 1568900 bytes | Blockstream API | One-time |
| Block weight at lock-in | 3993884 | Blockstream API | One-time |
| Transaction count at lock-in | 4370 | Blockstream API | One-time |
| SCCR (post-lock-in) | 0.371939 | data/sccr_latest.json | Continuous |
| L_net (SCCR baseline) | $5,627.80/block | data/sccr_latest.json | Continuous |
| Block interval range | 10–1736s | data/block_interval.json | Continuous |
| Node census (N) | ≥32,000 | data/node_census.json | Stale (2026-08-02) |
research/governance-boundary.md, the economic majority constituency (exchanges, custodians, merchants) is not measured — this is the one undefined quantity in the governance boundary frameBIP-110 is the first non-SCCR empirical case study for BSAHI's governance-boundary observation program. It demonstrates:
1. The pattern is observable. BIP-110's activation parameters (threshold, window, lock-in height, timeout mechanism) are public and verifiable on-chain. BSAHI's data pipeline captures the relevant signals (block versions, fees, block sizes, intervals) and can track governance events as they unfold.
2. The mechanism matters more than the outcome. BIP-110 locked in with 0% miner signaling — not because miners "supported" it, but because the mechanism (height-based UASF) bypassed the signaling requirement. This is the governance-boundary insight: who can change Bitcoin's valid-state transition rules depends on the deployment mechanism, not just hashrate support. BSAHI can observe and measure this mechanism's effects.
3. The observation is falsifiable. Every parameter of BIP-110's deployment is a falsifiable claim:
4. The data gaps define the research frontier. The things BSAHI cannot measure (node adoption, block rejections, restricted transaction volume, economic majority) are precisely the gaps that a full Governance Boundary Index would need to close. BIP-110 defines the minimum viable observation — the governance event is observable at the chain-production level, and the deeper economic coordination questions remain open.
BSAHI's second empirical pillar is earned. BIP-110's lock-in at height 963648 on August 23, 2026 — with 0% miner signaling and normal chain production continuing — is a verified, primary-source-confirmed governance event that demonstrates the governance-boundary observation pattern: the mechanism of consensus change (UASF height-based activation) is measurable on-chain, even when the outcome (miner support, economic coordination) is not. This is the first empirical case study outside the SCCR framework, and it validates the approach: governance events are observable at the chain level, and BSAHI's data pipeline is already capturing the relevant signals. The Governance Boundary Observatory concept is not speculative — it has a live, parameterized, falsifiable experiment running on Bitcoin right now.
Sources: blockstream.info API (primary on-chain), mempool.space API via GitHub Actions (data/bip110.json, data/bip110_daily.json), research/governance-boundary.md, research/BSAHI_FUTURE_RESEARCH_BACKLOG.md, data/sccr_latest.json, data/fee_history_blocks.json, data/block_interval.json, tools/agents/26-bip110-signal.js, fork-tracker.html.