Is this a good approach? The honest verdict · content/conclusions.md
Conclusions — Is This a Good Approach? (Honest Verdict)
Date: 2026-09-06 · Question asked: "I'd like conclusions as to if this is a good approach or investigation." Method: 16 cloned repos, live market data, 20+ research docs — every claim traceable in this wiki.
The verdict, one paragraph
Yes — as infrastructure and an internal credit system, this is a good approach, with unusually good economics for its class: world-class chain code for free (Apache-2.0), a real AU$1 backing mechanism that survives scrutiny because it's structural rather than promised, and an immediate customer (qalarc itself) so the system never has a cold-start problem. The investigation has already paid for itself as research; the build is justified if — and only if — the Phase-1 discipline holds: closed-loop, non-transferable, full-reserve, no public sale, lawyers before liquidity. As a public token play or an IOTA investment, the answer is no — the live data (−77.5%/yr, rank #181, cap abolished, +6.1%/yr inflation) is the market telling you what ten years of IOTA history proved: infrastructure value does not automatically become token value.
Scorecard by question
| Question | Verdict | Confidence | Why |
|---|---|---|---|
| Fork IOTA's node for our own data-transfer network? | DO IT | High | Apache-2.0, Sui-lineage quality, private-network tooling included (iota-localnet, genesis builder, docker). Cost ≈ engineering time only. |
| QALS as prepaid credit across all apps? | DO IT | High | Kills billing silos; agent spend-caps enforced by bytecode; we're our own first customer; no licence needed in the closed-loop phase (get the written opinion). |
| AU$1 backing ("minimum support")? | DO IT — as designed | Medium-high | B-QALS mint-on-deposit/burn-on-use is the only honest version: the reserve can never under-cover. Reject any temptation to fractionally back "temporarily". |
| Qalx DeFi platform? | DO IT — internal first | Medium | Parity desk + receivables vault are genuinely useful (compute-invoice finance is a real moat). External venue only post-VASP. DeepBook port deferred — 2–3 engineers + US$150–400k audit isn't justified yet. |
| Compute tracking on a tangle-lineage chain? | DO IT | High | Object model fits jobs natively; receipts = AI provenance = client-facing trust product. Start with the hub's LLM proxy. |
| Digital identity (Qal ID) as a stack addition? | DO IT — arguably first | High | Cheapest to start (open-source lib), highest strategic leverage: agent authentication, signed AI outputs, KYC tiers, EUDI-readiness for clients. See the Qal ID Stack page. |
| Buying IOTA tokens as an investment? | DON'T | High | −99.1% from ATH, −77.5%/yr, inflation 6.1%, rank sliding, Shimmer dead 30 Sept. Great codebase ≠ good token. |
| Launching QALS as a public tradeable token now? | DON'T — yet | High | Legal exposure (Treasury digital-asset legislation pending), zero benefit to the internal use-case, and it would poison the prepaid framing that keeps Phase 1 licence-free. |
What would change our mind (kill criteria — pre-committed)
- Legal opinion says the prepaid carve-out doesn't hold for on-chain transferable credit → revert to pure ledger-internal points (no token objects) and keep the chain for data/compute/identity only. The chain remains worth it alone.
- No PSP will hold segregated client money for a crypto-adjacent reserve → partner-bank route or park the reserve idea; run G-QALS-only with published floor coverage (weaker, still viable).
- Move/Rust capability can't be built (can't hire, upstream too fast) → fall back to running Qalnet as a tracked fork without custom Move (data anchors + identity only) — 60% of the value, 20% of the effort.
- Compute demand never materialises beyond internal → compute marketplace stays internal cost-accounting (still useful); don't fund external provider acquisition.
Expected value, bluntly
- Cost to first working system (devnet + credit + receipts, Phases 0–2): ~3 months of one engineer's time + ~AU$2–5k cloud/misc. The corpus, tooling, designs and simulator in this project are already done.
- Upside if base case lands: ≈AU$16.2M cumulative revenue by month 36 with breakeven ~month 20–24 (business review) — plus the compounding non-revenue value: every AI output signed, every agent bounded, every app on one ledger.
- Downside if killed at month 6: a private chain nobody uses, ~AU$10–20k sunk, and a team fluent in Move on the single best-engineered open chain codebase. That's a cheap option with a fat tail.
Recommended decision set (this week)
- ✅ Approve Phase 0 (2 weeks, ~zero cost): fork →
iota-localneton superlocal → firstDataAnchor. - ✅ Commission the fintech legal opinion (prepaid carve-out + VASP timing) — the single highest-leverage spend.
- ✅ Choose the PSP conversation (segregated client-money account) — start with Airwallex/Zepto-class.
- ⏸ Defer: any public testnet, any transferability, any DeepBook work, any NFT drop beyond receipts.
- 📊 Adopt the monthly proof-of-reserves dashboard from day one — it's the product, not paperwork.
- 🆔 Green-light Qal ID Phase A (agent DIDs for the hub) — independent of everything else, immediately useful.
Final word
The IOTA decade is the most valuable failed experiment in this space: it spent hundreds of millions proving what breaks (feeless public economics, custom crypto, coordinator training wheels, token-first thinking) and then, right at the end, shipped the thing worth keeping — a clean, fast, object-based chain you're allowed to take home. Qals is that: keep the machine, refuse the casino.
Not legal, financial or investment advice. Projections are scenarios, not promises. Verify everything — that's what the wiki is for.