This project develops the cryptography part of the PV080 lecture/seminar materials. It is intended to become a structured, maintainable source for rigorous definitions, explanations, examples, exercises, and related teaching outputs.
The initial focus is cryptography. Other PV080 topics will be added later without changing the project’s general goal: keep the underlying knowledge precise and reusable, while generating explanations suitable for different levels of learners.
The repository is at the project-initialization stage. The current working material is a draft corpus of cryptography definitions and a design for a machine-readable knowledge graph. Before large-scale atomization, the definitions need expert review and the corrections in the source notes need to be incorporated.
The terms-only vocabulary inventories that guided extraction (from the supplementary AI literature notes and the broader legacy scan) are archived at archive/vocabulary/ — fully cross-checked into definitions/precise-definitions.md with no remaining unincorporated content.
- Establish a technically precise vocabulary for cryptography.
- Organize concepts by prerequisites, composition, security properties, attacks, and protocols.
- Separate normative definitions from pedagogical explanations, analogies, and examples.
- Support progressive disclosure: beginner, intermediate, and advanced views of the same material.
- Maintain provenance and version information for facts.
- Generate reusable outputs such as lecture notes, diagrams, exercises, quizzes, flashcards, and assessments.
- Leave room for additional PV080 subject areas as the project grows.
The first definition layer covers:
- data, alphabets, strings, blocks, messages, plaintext, and ciphertext;
- mappings, injectivity, bijections, permutations, inverses, and one-way functions;
- keys, key generation, keyspaces, security parameters, and efficiency;
- encryption schemes, correctness, confidentiality, integrity, and authenticity;
- attacker capabilities and security experiments;
- block ciphers, stream ciphers, modes, IVs, nonces, counters, and keystreams;
- asymmetric primitives, key pairs, hard problems, key agreement, and signatures;
- hash functions, digests, compression, and resistance properties;
- randomness, entropy, RNGs, PRNGs, and CSPRNGs;
- diffusion, confusion, avalanche properties, non-linearity, and computational indistinguishability.
Authentication mechanisms such as MAC/HMAC and authenticated encryption are planned as a separate batch. Statistical test suites are not part of the precise definition layer.
The intended internal representation is one machine-readable source of truth. Every meaningful concept, condition, attack, requirement, protocol, role, mitigation, or example should be represented as a node; meaningful connections should be typed edges. Free text is reserved for explanation, not inference.
The graph should distinguish, among other things:
- fact types such as
definition,primitive,classification,property,mechanism,protocol,protocol_step,constraint,attack,adversary_model,security_goal, andexample; - conceptual relations such as
depends_on,defined_from,is_a,part_of, andprecedes; - security relations such as
enables,motivates,condition_for, andsatisfies; - pedagogical content versus precise content, including formulas and symbol mappings;
- abstract protocol roles versus names used in teaching examples;
- current facts versus deprecated facts, with provenance and migration information.
The first tooling milestone is a validator for the graph data. It should check schema validity, references, formulas and symbols, protocol roles and ordering, deprecated targets, and cycles in the combined conceptual-precedence graph.
The structure will evolve as implementation begins. A likely organization is:
README.md
definitions/ consolidated definition material (305 entries, done)
knowledge-graph/ logic.md — the consolidated taxonomy/schema/activity logic;
next phase (fact collection) works from this
archive/ fully-consumed source material and superseded process docs
(former ai/, extraction/, sources/, tools/, vocabulary/,
review/, pedagogy/, and the pre-consolidation
knowledge-graph/ files — see DOCUMENTATION.md)
data/ canonical machine-readable facts
schema/ data schemas and relation definitions
src/ import, validation, and generation tooling
examples/ worked examples and sample graph fragments
tests/ validation and generation tests
The consolidated working documents are indexed in DOCUMENTATION.md. The original source notes remain in the teaching workspace and are not modified by this repository.
- Define primitive concepts before dependent concepts.
- Keep each fact atomic and give it a stable identifier.
- Use
is_afor genuine type hierarchies andpart_offor composition. - Store canonical relation directions and derive inverses rather than duplicating them.
- Represent machine-actionable conditions and consequences as graph facts, not embedded prose.
- Keep formal notation consistent, especially the security parameter
λand notation for randomized algorithms. - Mark uncertainty, draft status, and provenance explicitly.
- Treat correctness, invalid ciphertexts, decryption failure, key-generation distributions, and adversary capabilities precisely.
The initial corpus was prepared in:
/home/sysox/Teaching/2027/lecture_seminar_text_inputs
The key course reference is Paul C. van Oorschot, Computer Security and the Internet: Tools and Jewels from Malware to Bitcoin, Second Edition. Its local reference copy is kept under the ignored archive/sources/crypto_books/toolsjewels_2e/ directory. Lecture exports are kept under the ignored archive/sources/lectures/ directory. See archive/sources/reference-materials.md for the source registry and controlled extraction workflow (archived; direct extraction into definitions/precise-definitions.md is now the active workflow).
The prompt-driven atom-extraction phase (its prompt and collaboration protocol) has been superseded by direct extraction and is archived at archive/ai/.
The most relevant inputs are the precise-definition drafts, the consolidated cryptography knowledge-graph proposal, the database recommendations, and A_corrections.txt. The corrections identify issues to resolve before treating the definition layer as normative—for example, randomized encryption, injective rather than necessarily bijective encoding, probabilistic correctness, ⊥ on decryption failure, formal security experiments and advantages, and the distinction between CPA/CCA capabilities and security notions.
- Review and reconcile the definition drafts.
- Incorporate the blocking technical corrections.
- Add a provenance inventory and consolidation notes.
- Incorporate the existing PV080 practical activity model.
- Establish a working taxonomy of cryptography information and fact types.
- Freeze the first normative definition layer.
- Define the graph schema and canonical relation vocabulary.
- Implement the validator and cross-file checks.
- Atomize the reviewed definitions into graph data.
- Add pedagogical projections, examples, exercises, and diagrams.
- Add further PV080 topics using the same underlying approach.
Not decided yet. Until this section is completed, treat the repository as teaching-development material and discuss substantial changes before publishing or redistributing it.