v1.10 — Proving Nothing

Release Notes — v1.10

Proving Nothing: A Layered Guide to Zero-Knowledge Proof Systems

Date: March 2026
Author: Charles Hoskinson

---

Title Change

The book has been renamed from The Seven-Layer Magic Trick to Proving Nothing: A Layered Guide to Zero-Knowledge Proof Systems. All internal references, build scripts, and metadata updated accordingly.

New Formats

• EPUB (Kindle-ready): Reflowable format per Amazon Kindle Publishing Guidelines v2025.1. MathML for mathematical content, HTML tables, custom CSS for Kindle Enhanced Typesetting.
• PDF (dark-mode): Updated with XeLaTeX-rendered cover page featuring classical painting artwork.

Cover Art

New cover art: a classical painting depicting a figure presenting a sealed certificate to a gathered audience — a visual realization of the book's prover/verifier metaphor. Cover text rendered via XeLaTeX for professional-quality typography.

Chapter-by-Chapter Editorial Changes

Chapter 1: The Promise of Provable and Programmable Secrets

• Expanded "The Phenomenon" with Fiat-Shamir transform bridge (the conceptual barrier to non-interactive proofs, Fiat & Shamir's 1986 insight)
• Expanded "Three Converging Forces" with eIDAS 2.0 selective disclosure detail, rollup mechanics explanation, and individual narratives for SP1 Hypercube, RISC Zero, StarkWare Stwo (940x throughput improvement), and EF zkEVM
• Expanded Layers 4-5 preview in "The Deepest Question" with concrete overhead example
• Introduced Midnight as second running example alongside Sudoku, with author disclosure
• Added polynomial definition and $2^{128}$ notation gloss
• Voice fixes: removed lecturer tics, reduced grandiosity

Chapter 2: Layer 1 — Building the Stage

• Expanded SRS section with geometric elliptic curve explanation, discrete logarithm grounding, and pairing explanation
• Added pairing bridge sentence for knowledge gradient
• Expanded Option-Value Analysis with 3-scenario sensitivity table and HNDL/NSA parallel
• Added bridging transition before Quantum Shelf Life section
• Deduplicated capex/opex sentence (3 verbatim repetitions → 1 canonical + 1 callback)
• Removed draft scaffolding (BGM17/BCTV14 parenthetical author's note)
• Tightened Peter Todd narrative, fixed "three mechanisms" → "four mechanisms" count
• Rewrote chapter coda as synthesis rather than recap

Chapter 3: Choreographing the Act (Layer 2 — Language)

• Filled PLACEHOLDER_PROVE_REST with honest description of proof-generation developer experience
• Expanded Philosophy C (RISC-V) from 321 to 650 words explaining why RISC-V won
• Moved Polygon zkEVM cautionary tale into Philosophy A section (was standalone)
• Added Sudoku running example in Compact (verify_sudoku circuit)
• Added Chapter 2→3 transition callback
• Glossed "nanopass" terminology inline
• Added TVL timestamp qualifiers

Chapter 4: The Secret Performance (Layer 3 — Witness)

• Developed "privacy is a luxury good" argument with implications for technology roadmaps
• Added witness verification subsection (property-based testing, differential testing, formal verification)
• Added Sudoku scaling context (80 field elements → thousands → billions)
• Added equity gap in synthesis connecting four technical gaps to hardware stratification
• Fixed 95% → 96% population arithmetic
• Removed defensive "This is not science fiction" line
• Added timestamp qualifiers to all hardware prices

Chapter 5: Encoding the Performance (Layer 4 — Arithmetization)

• Strengthened Chapter 4→5 transition with causal bridge
• Added mid-chapter checkpoint after CCS section
• Added developer-facing overhead rule of thumb
• Connected ZKIR to Chapter 3's Sudoku example
• Voice fixes: removed condescension and lecturer tics

Chapter 6: Layer 5 — The Sealed Certificate

• Added Nightstream generalizability note
• Added Solana ZK ElGamal date qualifier
• Connected Proof Core observation to Chapter 10 preview
• Added Midnight devnet performance caveat

Chapter 7: Layer 6 — The Deep Craft

• Expanded Sections 10-12 (algebraic hash functions, structural advantage of lattices, maturity and readiness)
• Reduced "miraculous" repetition (5→3 instances)
• Softened Forrester Q-Day 2030 attribution
• Added 2-adicity explanation inline
• Added curve cycle and non-native arithmetic glosses
• Added production blockers for lattice adoption
• Differentiated decision frameworks (Cascade Effect vs. One-Way Door)

Chapter 8: Layer 7 — The Verdict

• Expanded Solana ZK ElGamal example with attack specifics
• Added proving-system-specificity caveat to pricing attacks table

Chapter 9: Privacy-Enhancing Technologies

• Expanded Midnight/Kachina connection (disclose() as Kachina's information-flow control)
• Added Part III transition bridge
• Voice fix: "key insight" → direct statement

Chapter 10: The Synthesis — Three Paths, Not Two

• No edits needed — publication-ready as written (DAG with 14 edges, three paths, trust decomposition delivered)

Chapter 11: zkVMs — The Universal Stage

• Added synthesizing conclusion "The Stage Is Set" (three conclusions: ISA war over, proof core visible, competitive axis rotating)
• Added Midnight exclusion note in landscape table
• Added forward bridge to Chapter 12

Chapter 12: Midnight — The Privacy Theater

• Expanded Layer 4 (ZKIR) with concrete Sudoku DAG example and abstraction cost analysis
• Expanded Layer 5 (Halo 2) with selection rationale and four-phase pipeline detail
• Expanded Layer 6 with Jubjub curve role, Poseidon dual nature, field size tradeoff
• Expanded governance analysis with verifier key upgrade mechanics
• Expanded side-channel gap with Poseidon, indexer, and network timing specifics
• Fixed "seven dedicated layer analysts" phrase
• Strengthened Lesson 5 with generalizable principle

Chapter 13: The Market Landscape

• Added Midnight reference in rollups section
• Added aggregate $20B TVL framing and Stage 2 maturity gap
• Added concrete coprocessor example (DeFi lending TWAP query)
• Fixed ZKML "700x faster" baseline
• Added eIDAS 2.0 scale (450M users, 4 Large Scale Pilots)
• Consolidated Privacy Pools data from Chapter 9
• Added market segment breakdown table with estimated percentages
• Added CAGR methodology for projections

Chapter 14: Open Questions and the Road Ahead

• Deepened Q1 (GPU witness) with team names and "solved" definition
• Deepened Q4 ("trustless") with explicit seven-assumption enumeration and actuarial concept
• Expanded Convergence section mapping three forces (Ch1) to three frontiers
• Expanded Privacy Frontier with three additional dimensions
• Expanded Coda with pharmacy and Madrid personas, difficulty acknowledgment
• Connected open questions to frontier blockers with risk scenarios

Cross-Book Improvements

• Midnight introduction naturalized: Glossary entry, Chapter 1 planting + author disclosure, Chapter 2 bridge paragraph, Chapter 12 framing paragraph
• Knowledge gradient fixes: Pairing bridge sentence in Chapter 2, polynomial definition in Chapter 1, NTT gloss in Chapter 4, 2-adicity in Chapter 7
• Voice consistency: 18 AI-tell markers across 5,511 lines (0.33 per 100 — far below threshold)
• Cross-reference integrity: 87 chapter references verified, zero broken
• All forward promises fulfilled: DAG in Chapter 10, three paths, trust decomposition, magician metaphor farewell

Build System

• New Kindle EPUB pipeline: build_kindle.py with kindle.css, kindle-filter.lua, metadata.xml
• Cover generation: generate_cover.py (Pillow) and coverpage.tex (XeLaTeX)
• Font fix: STIX Two Text installed, Consolas → DejaVu Sans Mono fallback
• Files renamed from the-seven-layer-magic-trick.* to proving-nothing.*

File Inventory

File	Description
proving-nothing.md	Master manuscript (5,512 lines)
proving-nothing.pdf	Dark-mode PDF (357 pages)
proving-nothing.epub	Kindle-ready EPUB
assets/coverart.jpeg	Source cover painting
assets/cover_print.jpg	Cover image (portrait)
assets/cover_kindle.jpg	Cover image (Kindle landscape)
coverpage.tex	XeLaTeX cover page
build_kindle.py	EPUB build script
kindle.css	Kindle stylesheet
kindle-filter.lua	EPUB Lua filter
tex/chapter1-14.tex	Standalone LaTeX chapters

v1.0.1 -- Added CC License

Added a Creative Commons Attribution 4.0 International (CC BY 4.0) license page to the PDF, placed on the verso of the title page following standard book publishing convention.

The license page includes the copyright notice, a plain-language summary of CC BY 4.0 terms (share, adapt, attribute), and a link to the full license text.

No content changes to the manuscript itself.

v1.0.0 -- First Edition

v1.0.0 -- First Edition

The Seven-Layer Magic Trick is a complete guide to zero-knowledge proof systems, written for engineers, executives, and researchers who want to understand how the technology actually works -- not just what it promises.

The book's argument: ZK proofs don't eliminate trust. They decompose it into seven independent, weaker assumptions. Each layer of the stack introduces one assumption. Each assumption can be tested, audited, and replaced without touching the others. The word "trustless" is marketing. The accurate word is "trust-minimized." This book shows exactly how much trust remains, where it lives, and what breaks when each assumption fails.

What's inside

• 14 chapters across three parts: the invitation (what ZK proofs do and why now), the craft (six technical layers from setup ceremonies to on-chain verification), and the synthesis (trust decomposition, zkVM landscape, Midnight case study, market analysis, open questions)
• A running example -- a 4x4 Sudoku proof traced from program to witness to polynomial constraints to sealed certificate, so one concrete computation threads through every layer
• 64 references covering the primary literature from Goldwasser-Micali-Rackoff (1985) through Symphony and Arguzz (2025)
• Three reading paths -- 45 minutes for executives, 2 hours for engineers, 4+ hours for researchers
• Seven open questions framed with executive risk assessments and timelines, not just as research prompts

Formats

• PDF (attached below) -- Midnight Network dark-mode branding, Outfit headings, STIX Two Text body, custom syntax highlighting with brand-derived colors. Designed for screen reading.
• Markdown and LaTeX sources are in the repo if you want to rebuild or adapt the content.

Technical coverage

The book covers the full ZK stack as of March 2026: R1CS, AIR, PLONKish, and CCS constraint systems. Groth16, PLONK, and STARKs. The sumcheck protocol. Lookup arguments from Plookup through Lasso to Jolt. The folding genealogy from Nova through HyperNova, LatticeFold, Neo, and Symphony. Circle STARKs and Stwo. The hybrid STARK-to-SNARK pipeline. Small-field arithmetic (BabyBear, Mersenne-31, Goldilocks). The quantum threat and the lattice response. Fiat-Shamir vulnerabilities. Governance attacks. Proof aggregation. Seven zkVMs compared through all seven layers.

It does not assume prior knowledge of cryptography. It does assume the reader is willing to think.