- JavaScript 64.5%
- CSS 30%
- HTML 5.5%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
|
All checks were successful
ci/woodpecker/push/woodpecker Pipeline was successful
|
||
| crosswalks | ||
| froscon-2026-oss-metadata | ||
| .gitignore | ||
| .woodpecker.yml | ||
| AGENTS.md | ||
| CRA_WISHLIST.md | ||
| DATA_FLOW.md | ||
| EU_POLICY_CONTEXT.md | ||
| FUTURE_WORK.md | ||
| GLOSSARY.md | ||
| LEVY_VISION.md | ||
| LICENSE | ||
| MARKET_SURVEILLANCE_EVIDENCE_REQUIREMENTS.md | ||
| METHODOLOGY.md | ||
| MODEL_ASSESSMENT.md | ||
| MODEL_OBLIGATIONS.md | ||
| MODEL_PRESERVATION.md | ||
| MODEL_SOVEREIGNTY.md | ||
| PITCH.md | ||
| PROCUREMENT_EVIDENCE_REQUIREMENTS.md | ||
| PROPOSAL.md | ||
| README.md | ||
| RESEARCH.md | ||
| RISK_ANALYSIS.md | ||
| ROADMAP.md | ||
| TOOLING.md | ||
How to Read This Project
This guide helps you navigate the documentation based on your role and what you need to learn.
Quick Start: 60-Second Summary
The Core Idea
Open source procurement is held back not by the absence of good software, but by the absence of shared, machine-readable evidence — the foundation of data-driven procurement decisions rather than purely relationship-driven ones. The north star of this project is decentralized metadata exchange: every stakeholder — projects, vendors, deploying organizations, registries — publishes structured data from their own domain. That data can be crawled independently by any catalog or procurement tool, and further annotated with external verification, trust labels, or policy signals. No central authority controls what gets published or how it is interpreted.
Every step toward this vision has value on its own terms. If some components of this proposal are adopted, others handled by adjacent standards, and still others deferred or never needed — the ecosystem still moves in the right direction. Read the proposal's breadth as a concrete map toward that destination, not a prerequisite checklist.
What This Is
- A proposal to improve
publiccode.ymland related ecosystem standards. - A practical path to make open source easier to discover, assess, and procure in the public sector.
- A bridge between technical metadata standards and procurement/policy enforcement.
Why It Matters
- Public institutions often support digital sovereignty in principle, but procurement still defaults to proprietary vendors.
- Procurement officers face institutional risk when deviating from familiar frameworks.
- Security is often perceived as a proprietary advantage, even though roughly 96% of software products already include open source components (OSBA).
- Missing piece: reliable, comparable, auditable infrastructure for evaluating open source options at scale.
- A data-driven approach lowers perceived personal and institutional risk by replacing relationship-based choices with documented evidence, making it easier and safer to deviate from proprietary-first defaults.
What This Project Adds
- Better project metadata and discovery standards.
- Registry APIs and discovery mechanisms for decentralized evidence collection.
- Policy-aligned structures so procurement decisions can be transparent and defensible.
- A data-driven basis for procurement decisions that supports justified deviation from incumbent proprietary norms.
Positive Feedback Loop (Design Intent)
- Procurement policies and sovereignty mandates create demand for standardized metadata.
- Projects and vendors adopt the extended standards.
- Governments gain auditable evidence for procurement, security, and funding decisions.
- Open source adoption becomes more confident, sustainable, and resilient.
Who This Is For
- Procurement professionals
- Open source maintainers
- Policy makers and legislators
- Vendors and service providers
- Researchers and standards specialists
- Funders and grant managers
- Software catalog operators
Core Proposal Components
Improvements to publiccode.yml:
- Faceted Classification — multi-dimensional discovery
- Supply Chain References — SBOMs, policies, security evidence
- Accessibility Declaration — product accessibility baseline for procurement
- Certifications and Seals — attested proof, project-side
- Interoperability and Standards Declarations — per-standard support status
- Compliance Document References — privacy policy, terms, DPA
- Vendor Contribution Credit System Discovery — vendor and contributor visibility
- Deprecate
usedBy— usage moves to external registries (trust-model-tagged decentralized adoption signals) - Deprecate Temporal Fields — cleaner and more trustworthy data
- Package Distribution References — npm, PyPI, Maven, etc.
- Sanctioned Mirror Declarations — forge migration and mirrors
(Further improvements await consensus: CRA Steward Declarations, informed by the Commission's July 2026 CRA application guidance; Platform Lock-in Declaration, seeking community input; External Metadata Source References, defining co-existence with peer metadata formats such as CNCF .project and OpenSSF security-insights.yml; and Lifecycle Status Declaration. See PROPOSAL.md for details.)
Companion specifications:
- Contribution Credit Registry API
- Registry Discovery Standard (
/.well-known/publiccode-registry.json) - Trust and Identity — trust models and entity identifiers shared across registries
- Usage Registry API
- Organization-Level Usage Declarations (
.well-known/publiccode-usage.json) - Project-Level Funding Declarations (
.well-known/publiccode-funding.json) - Assessment Registry API (for Sovereignty Checks and regional compliance assessments)
Policy layer and governance alignment:
- Procurement policies that prefer open source and use standardized criteria
- Selection frameworks based on metadata-backed evidence
- Legislative reference models (including Switzerland's EMBAG law)
Policy Primer (Optional)
For policy makers new to open source policy: Mirko Boehm, "How Open Source Coordinates: A Guide for Policy Makers" (SSRN, 2026): https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6303038
Trust Model in One Line
- Independent registries expose standardized data and declare trust models (
verified-domain,signed-attestation,self-reported) so evidence consumers can choose risk-appropriate sources. See Design Principles.
Documents at a Glance
| Document | Purpose | Length | Best for | Start here if… |
|---|---|---|---|---|
| GLOSSARY.md | Key terms explained | ~20 min | Everyone | You encounter unfamiliar terms or acronyms |
| RESEARCH.md | Comparative analysis of 5 standards | ~30 min | Researchers, policy makers, infrastructure decision-makers | You want to understand why publiccode.yml and not something else |
| DATA_FLOW.md | Ecosystem architecture and data flow diagram | ~15 min | Architects, system designers, visual learners | You want to see how the ecosystem connects together |
| PITCH.md | Why each actor benefits | 10 min | Practitioners in each role | You want to know "what's in it for me?" |
| PROPOSAL.md | Detailed technical spec covering the publiccode.yml improvements plus companion standards (registry APIs, discovery, organization declarations, assessment registries) | ~75 min | Maintainers, spec authors, technical implementers | You want the concrete technical specification and full ecosystem design |
| EU_POLICY_CONTEXT.md | How this proposal relates to CRA, NIS2, Interoperable Europe Act, and the Public Procurement Regulation | ~20 min | Policy makers, legal teams, procurement professionals | You want to understand the regulatory landscape and legislative hooks |
| PROCUREMENT_EVIDENCE_REQUIREMENTS.md | Format-neutral statement of what public buyers need to know, from whom, with what verifiability, at what point in time — written for practitioner review and for embedding in external standards efforts (IETF SCITT, OpenSSF, harmonised standards) | ~30 min | Procurement professionals, standards liaisons | You want the buyer-side requirements without the technical format design |
| MARKET_SURVEILLANCE_EVIDENCE_REQUIREMENTS.md | Format-neutral statement of what a market surveillance authority needs to establish, about whom, through which channel — centered on proactive oversight ahead of incidents (funding gaps, control transitions, behavioural change, composite health indicators), with reactive investigation and defensible enforcement built on the same evidence, all by observation with no duty upstream | ~30 min | Surveillance practitioners, regulators, standards liaisons | You want the authority-side requirements — acting before the incident, not after it |
| MODEL_SOVEREIGNTY.md | Exploratory, and the entry point for the model strand: the openness ladder that makes a sovereignty claim testable, the layers a model dependency actually consists of, development-time AI as a distinct dependency, and declaring model requirements in forges | ~17 min | Metadata and catalog work, policy makers | You're asking what happens to public-sector AI dependencies if the model hub changes hands |
| MODEL_ASSESSMENT.md | Exploratory: what a sovereignty record must carry about a third-party model, who can assert it, verdicts scoped to use profiles, geopolitical criteria written as compellability rather than nationality, and why a public assessor must observe rather than admit | ~12 min | Assessors, policy makers, procurement | You're asking who decides whether a model can be trusted, and how that avoids becoming a gatekeeper |
| MODEL_OBLIGATIONS.md | Exploratory: how AI models break machinery the CRA and AI Act already own — undetectable malicious components, vulnerability handling with no patch and no channel, an empty due-diligence branch, and a support period that promises more than the dependency guarantees | ~19 min | Compliance and regulatory policy | You need to know what a manufacturer actually owes for a model it cannot obtain, modify, or fix |
| MODEL_PRESERVATION.md | Exploratory: what holding models would take — completeness tiers, federation designed so no participant can hold others hostage, the building blocks that already exist, cost shape, and legal deposit as the instrument nobody has pointed at models | ~13 min | Infrastructure planners, preservation institutions, funders | You would operate or fund a model preservation federation |
| RISK_ANALYSIS.md | Identified risks and mitigations | ~20 min | Risk managers, decision-makers | You need to understand what could go wrong |
| ROADMAP.md | Phased implementation plan | ~20 min | Project managers, funders, coalition builders | You're thinking about how to make this real |
| TOOLING.md | Tooling ecosystem, gaps, and existing OSS building blocks — existing publiccode.yml tools, adjacent data sources, the new/adapted tools the proposal needs, and which open-source tools fill each gap | ~25 min | Tool builders, catalog/registry operators, implementers | You want to build or extend tooling that supports the proposal |
| METHODOLOGY.md | Research process documented | ~40 min | Meta/research/reproducibility-focused readers | You want to understand how conclusions were reached |
| FUTURE_WORK.md | Speculative directions recorded but not part of the core proposal (no committed design) | ~10 min | Contributors, researchers, anyone scoping the boundaries | You want to know what is deliberately deferred or out of scope |
| LICENSE | CC-BY-SA 4.0 | 2 min | Everyone | You need reuse permissions |
Reading Paths by Role
Pick the track that matches your available time:
- Quick path: 10-20 minutes for orientation and decisions
- Deep dive: 35-120 minutes for implementation and policy detail
1. Procurement Professional
You're evaluating whether open source is right for your organization.
Quick path (12-15 min):
- PITCH.md → Procurement Office
- PROCUREMENT_EVIDENCE_REQUIREMENTS.md — the buyer-side requirements, written for practitioner review; your corrections and additions are the input this project most needs
- RESEARCH.md → publiccode.yml strengths
- PROPOSAL.md → Policy Context
Deep dive (40 min):
- Add PROPOSAL.md → Design Principles and Faceted Classification
- Use GLOSSARY.md for unfamiliar terms
2. Open Source Project Maintainer
You're considering whether to adopt publiccode.yml metadata.
Quick path (15-20 min):
Deep dive (50 min):
- Add RESEARCH.md → publiccode.yml
- Add PROPOSAL.md → Design Principles
- Use GLOSSARY.md as needed
3. Vendor / Service Provider
You deliver expertise around specific open source projects.
Quick path (12-15 min):
- PITCH.md → Vendor/Service Provider
- PROPOSAL.md → Vendor Contribution Credit System Discovery
- ROADMAP.md → Phase 3
Deep dive (35 min):
- Add RESEARCH.md → publiccode.yml for background
- Use GLOSSARY.md for terminology
4. Policy Maker / Legislator
You're drafting open source procurement requirements or digital sovereignty mandates.
Quick path (20-25 min):
Deep dive (70 min):
- Add RESEARCH.md → Digital Sovereignty Score Initiatives
- Add ROADMAP.md (focus on Phases 0, 2, 3, and 5)
- Use GLOSSARY.md as needed
5. Researcher / Standards Specialist
You're studying open source governance, metadata standards, or supply chain transparency.
Quick path (20-30 min):
Deep dive (120 min):
- Add METHODOLOGY.md first for provenance and process
- Use GLOSSARY.md for terminology normalization
6. Funder / Grant Manager
You allocate money to open source infrastructure and want to understand this proposal.
Quick path (15-20 min):
Deep dive (50 min):
- Add RESEARCH.md → publiccode.yml
- Use GLOSSARY.md as needed
7. Software Catalog Operator
You run openCode.de, Developers Italia, or another registry/index.
Quick path (15-20 min):
- PITCH.md → Software Catalog/Crawler
- PROPOSAL.md → Registry Discovery Standard
- ROADMAP.md → Phase 4
- TOOLING.md → Tooling gaps
Deep dive (65 min):
- Add RESEARCH.md → Federated Architecture
- Add PROPOSAL.md (focus on Supply Chain References and Deprecate
usedBy) - Add TOOLING.md for what to build or extend
- Use GLOSSARY.md as needed
Tips for Reading
Terminology tip: If you see an acronym or unfamiliar term, check GLOSSARY.md first. It's designed to make everything accessible.
Links: Click links to jump between sections. Most documents cross-reference each other so you can read non-linearly.
Tables: Scan tables first—they often contain the key insights in compact form.
Technical sections: PROPOSAL.md has YAML examples. You don't need to understand YAML syntax; read the descriptions alongside the code blocks.
Scope boundaries: Each document tries to be honest about what it does not cover. If something seems missing, check the Risk Analysis and Methodology sections for why.
How to Contribute Feedback
This project is open for feedback and collaboration. Key places to raise ideas:
- On the proposal: See ROADMAP.md → Phase 0 → Actions for governance processes
- On research methodology: See METHODOLOGY.md → Limitations
- On specific risks: See RISK_ANALYSIS.md — if you see a missing risk, that's valuable input
- On implementation details: See ROADMAP.md — phases are sequenced but not set in stone
Project Status
- ✅ Completed: Initial comparative analysis (RESEARCH.md), proposal (PROPOSAL.md), pitches (PITCH.md), risk analysis (RISK_ANALYSIS.md)
- ✅ Completed: Possible roadmap and initial coalition mapping (ROADMAP.md)
- ✅ Completed: Supporting reference material — glossary (GLOSSARY.md), ecosystem architecture and data flow (DATA_FLOW.md), EU policy context (EU_POLICY_CONTEXT.md), tooling survey (TOOLING.md), research methodology (METHODOLOGY.md), future work (FUTURE_WORK.md), and peer-format crosswalks (
crosswalks/) - 🔄 Current: Community review for completeness and accuracy; early outreach and pilot preparation underway (engagement, procurement-pilot, and conference-submission drafts in
pilots/) - 🔄 Next: Find initial allies and open governance discussions (Phase 0 of ROADMAP.md)
License: CC-BY-SA 4.0 (see LICENSE file)