Finite supply defended & attacked; the 1000-cap verdict; the threshold endgame · content/qals-tokenomics-paper.md

The Finite Qals — a paper on scarcity, caps, and what happens at the threshold

Author: qalcode (GLM 5.3) · Date: 2026-09-12 · Status: positions of the project, argued honestly — both sides first Companions: token supply · fee policy · exchange feasibility · the backing design

Abstract

QALS has a fixed maximum of 1,000,000 Qals with a 500,000 sales cap — enforced in code: the full sellable supply is pre-endowed in the genesis issuer wallet qals:issuer (2026-09-15) and each purchase transfers out of it, so the cap is the wallet running out (402 "sellable supply exhausted"); the on-chain cap-gate (qal_reserve) is written and tested — the on-chain twin is being redesigned from mint-gated to genesis-preminted + transfer-gated at the cap-raise checklist. This paper defends that design, attacks it, and resolves the founder's three live questions: should per-person purchase caps (1,000 Qals/card) limit supply domination; what actually happens when the 500K threshold is crossed; and whether a second, stable token (qAUD) with a QALS↔qAUD exchange is the right endgame — the IOTA↔Shimmer relationship, done deliberately. We conclude: finite is right but for a subtler reason than "store of value"; per-person flat caps are the wrong tool (graduated caps are right); and the two-token endgame is structurally correct with the Shimmer lesson baked in.

1. The design as it stands

The critical, often-missed nuance: during the credit phase, QALS is not an asset — it is a liability. Every sold Qals is a claim on AU$1 of service redemption. "Scarcity" of a liability is meaningless until the cap is crossed and the credit promise is unwound. That is exactly why the 500K threshold question matters so much.

2. The case FOR finite supply

1. The network beneath it is finite. This is a private chain with operator-run validators and a real compute fleet behind it. Token float should reference actual capacity, not speculative infinity. An infinite-credit ledger pretending to be backed is how every fractional-reserve disaster starts.
2. The IOTA natural experiment (the hard evidence). IOTA abolished its cap in 2025; supply now inflates ~6%/yr. Result: world-class infrastructure (Starfish consensus, WEF-grade trade deployments) and a token at −99% from ATH, rank ~#180. The lesson cuts BOTH ways: inflation didn't fund success, and finite supply wouldn't have prevented the decline — token economics didn't determine either outcome; usefulness did. Finite supply's honest job here is discipline: it forces the operator to fund the network from revenue (spread, services) rather than from dilution. That discipline is the real argument.
3. Gresham dynamics are absent while it's credit. Nobody hoards gift cards when the shop might close; they spend them. A finite credit supply simply bounds outstanding liabilities — good accounting, not monetarism.
4. Crossing the threshold creates a genuine asset. At 500K sold, unsold Qals become the scarce settlement/governance asset of a working network (see §5). Scarcity then means something: you cannot issue more network-foundation shares quietly. The fail-closed refusal (loopd rail today; on-chain gate written + tested) is the whole point — it makes the promise verifiable rather than aspirational.

3. The case AGAINST (steel-manned)

1. Deflationary currency is a bad medium of exchange. If QALS appreciates, rational users hoard rather than spend — but our users must spend to get compute. The system resolves this by pricing services in AUD-pegged credit and letting the asset float separately — which concedes the critic's point: a spending unit should not be the scarce asset. That is the two-token argument arriving early (§5).
2. A cap creates a death-spiral scenario. Near the threshold, compute demand in credit terms could exceed remaining sale supply: prices of Qals spike, compute becomes "expensive" in Qals even though the AUD price is unchanged — a pure unit-supply artifact that damages users. Mitigation: threshold day IS the migration day (convert pricing to qAUD simultaneously); don't operate near-cap in the old unit.
3. 1M units may be socially awkward even though divisibility (10¹² doofs) makes it economically sufficient — "only a million" reads small to crypto natives habituated to billions. Counter: BTC is 21M units; the number is marketing, the divisibility is the substance; a 1:1000 redenomination later is trivial if ever wanted (rename, don't inflate).
4. Fixed supply removes a governance tool. Validator decentralisation (Phase 3+) needs a funding source; an inflating token funds it invisibly (and corruptly — see IOTA). Honest answer: fund validators from service margins and the Reserve yield, stated in the fee policy's caveat. If that ever proves insufficient, that is a Phase-3+ redesign conversation with holders — not a quiet emission.

4. The 1,000 Qals/person cap — the founder's instinct, examined

Goal: prevent few actors buying dominating supply. Founder's own objection: in practice it caps a heavy user at AU$1,000 of compute — a day of GPU for some.

Analysis. A flat, forever cap fails both ways: too tight for real users, too loose for domination (1,000 people × 1,000 = the whole cap; a sybil with 500 cards owns half). The domination problem is about concentration, not individual size; the compute problem is about usage, not holding. These need different tools:

Recommendation: graduated tiers + launch windows; drop the flat 1,000 idea. It solves domination with sybil-cost (verified identity tiers) while leaving genuine usage unlimited through the earn side.

5. Crossing the 500K threshold — the two-token endgame

What the founder sketched is correct, and here is the precise mechanics:

• Day 0 (threshold): sales close (cap-refused already, in code). Announce migration: service pricing moves to qAUD, a new AUD-pegged stable token (mint-on-AUD-deposit, burn-on-redeem — the qal_reserve pattern, re-pointed; live code exists).
• QALS becomes the network asset: settlement gas for the (now-decentralising) validator set, governance weight, and the scarce claim on network value. Rate QALS↔qAUD floats on the existing qalx AMM (live, 64/64-tested) — the "IOTA↔Shimmer relationship" but with the roles corrected: the stable token is the everyday rail (what Shimmer failed to be commercially, and what IOTA's own token couldn't be while unpegged), and the finite token is the network's equity-like settlement asset (what IOTA's token wanted to be).
• The honest Shimmer lesson: Shimmer (2022–2026) was the staging token for IOTA — valueless-by-design, sunset 30 Sep 2026, holders unpaid. The analogy's structure (two tokens, one network) survives; its commercial lesson is: a staging token with no cash-flow role dies. qAUD avoids that by being the transactional token from day one — it is the thing people actually spend.
• Migration mechanics: credit-phase Qals convert 1:1 to qAUD on redemption demand (the Reserve unwinds its liability); the asset float stays as QALS. No forced swap, no snapshot games — the liability retires, the asset remains. The Reserve's AUD backing funds exactly the qAUD mint, so no new promise is created.

6. Own stablecoin variants (AUD/USD/YCN) vs USDT

Position: qAUD first (it's who we are), qUSD second if international demand is real (same machinery, USD reserves), no yuan variant, no USDT dependency — external USDT pairs only ever post-VASP on licensed venues, never as the internal unit.

7. Is inflation EVER right here? (founder: "I don't think so — at all")

Agreed for Phases 1–3: emissions would fund nothing we can't fund from revenue, and they'd break the credit-phase accounting (inflating a liability is fraud-adjacent). The single future exception, stated once in the fee policy and repeated here: validator decentralisation may need a funding source — margin-share or Reserve yield first; protocol emission only as a last resort, and only with a governance decision at that time. No quiet money-printing, ever.

8. Supply sizing — is 1M enough?

During the credit phase: supply must bound outstanding credit. Peak planned program = 500K sold + organic earn-side circulation — 1M is comfortable. Post-threshold as the network asset: total asset value = 1M × price; there is no "too few units" at 10⁶ divisibility (10¹² doofs — more discrete units than exists AUD cash). If social numerics ever matter, redenominate (display ×1000), never inflate — a display change, not a monetary one.

9. Recommendations (numbered, concrete)

1. Keep the 1M/500K design exactly as code-enforced (loopd rail live; on-chain gate written + tested). (No change.)
2. Replace the flat 1,000/card idea with graduated purchase tiers (1K → 10K verified → 50K KYC2, rising with history) — caps on buying the sale, never on earning/spending.
3. Add a launch-window per-identity daily cap (90 days) if/when a public sale opens.
4. Pre-build qAUD now in paper form (it's a qal_reserve fork — days of work) so threshold-day is a config flip, not a project.
5. Threshold plan: sale closes → pricing moves to qAUD same day → QALS floats as settlement/governance asset on qalx → Reserve unwinds credit liabilities 1:1 into qAUD mints.
6. No yuan stable; qUSD only on demonstrated international demand; USDT only on licensed external venues.
7. Publish the 5%-cluster transparency flag (anchored, non-punitive) as the domination backstop.
8. Write the validator-funding decision (§7) into governance docs now so Phase-3+ has a pre-agreed path that isn't emission-by-default.

This paper argues positions; the ground truth remains the fact sheet. Feedback welcome — this is exactly the class of document that improves under attack.