Additional user-supplied research edition · 2026
This is a complete additional source edition supplied for editorial comparison. Its claims, projections and proposed frameworks have not all been independently verified. For the selected and qualified argument, read the integrated study and the public case for cognitive liberty.
Historical synthesis, publisher cooperative architecture, linking charter, prototype and 12-month pilot · CognitiveLiberties.com · 11 October 2026
Executive finding
Hyperlinks provide a low-cost, voluntary and inspectable way for independent publishers to make one another discoverable. Curated directories, feeds and editorial references can reduce dependence on any single algorithmic feed without rejecting search engines or AI. The objective is not to engineer reciprocal backlinks, but to reconnect people to documents and authors they would not otherwise encounter. A useful network is a transparent collection of independent editorial judgments, not a central registry authorized to decide which lawful ideas may exist.
1. From exploratory web to intermediated discovery
In the early web, manually authored hyperlinks, academic reference pages, webrings, directories, weblog blogrolls and syndicated feeds let readers move along human-chosen relationships. The experience could expose distinctive local and intellectual communities, but it was also uneven: small sites vanished, curation reflected personal tastes, accessibility varied and much material was difficult to find without a directory. Search improved navigation of the enormous web, while social distribution and ranking made visibility increasingly dependent on intermediary incentives.
Modern SEO also changed publishing. Links became a ranking signal, inviting manipulation, but this does not mean all publishers stopped linking naturally or that authentic reciprocal references are inherently unethical. Search guidelines discourage link schemes intended to manipulate rank, not editorial citations made for readers. The proper remedy is clearly authored references, descriptive context, and transparent editorial independence—not a mass exchange of URLs.
2. Open protocols and realistic roles
| Mechanism | Contribution | Constraint / decision |
|---|---|---|
| Ordinary HTML links | Universal, resilient, readable, crawlable paths to sources | Needs editorial context and link-health checks |
| RSS / Atom | Subscribe to original posts without one feed algorithm | Feed format and update frequency vary |
| OPML | Export curated feed bundles and transfer subscriptions | Trust and quality cannot be inferred from presence |
| Webmention | Notify cited sites of references using an open W3C protocol | Incoming notifications require spam screening [1] |
| ActivityPub | Federated social distribution and replies | Governance, abuse and server moderation are necessary [2] |
| Sitemaps, canonical URLs | Improve source discoverability and canonical consistency | Do not guarantee indexing or a preferred rank [3] |
| Public APIs / static catalogs | Preserve machine-readable link and citation graphs | Cost, maintenance and licensing must be planned |
Use the web itself as the default integration. Optional ActivityPub or Webmention adoption should be staged, not required for participation. An organization that publishes accessible HTML and Atom feeds is already a valuable node.
3. Architecture: Cognitive Liberty Discovery Network (CLDN)
Decentralized content: authors retain their own canonical sites and update their own work. Local editorial catalogs: each participating publisher curates its own themed link collection. Open references: link items include title, original URL, short editorial rationale, subject tags, language, evidence type, license and last-checked date. Cross-directory discovery: catalogs can be exported as JSON and OPML and may point to other independently governed catalogs. Federated trust: no universal whitelist; participants publish their selection criteria, conflicts, corrections and exit terms. Reader controls: users can explore by topic, language, evidence type and viewpoint diversity, or show the original-source list without AI summarization.
A sample static JSON record:
{
"id": "cldn:essay:0042",
"url": "https://publisher.example/research/primary-source",
"title": "Source document and critical commentary",
"language": "en",
"topics": ["information-discovery", "intellectual-freedom"],
"source_type": "original-research",
"editorial_reason": "Contains a replicable method and cited counterargument",
"editor": "CLDN guest editor",
"editorial_disclosure": "No sponsorship; selection not an endorsement",
"checked_on": "2026-10-11",
"license": "Link only; see source for reuse permissions"
}
These fields are illustrative, not the assertion that a live API already exists. JSON records should support revision IDs, correction links and link-integrity checks. The original source's license is not inherited by the directory's commentary.
4. A Public Linking and Editorial Independence Charter
- Link because a source materially advances understanding, not because another publisher promised a backlink.
- Preserve direct access to original sources; summarization should supplement, not replace, reading.
- Cite strong criticism of one's preferred argument when relevant.
- Explain significant editorial selections and factual limitations.
- Distinguish inclusion from institutional endorsement.
- Disclose sponsorship, ownership conflicts and paid placements prominently; do not sell purported independent citations.
- Permit participants to leave and take their content, records and audience routes with them.
- Allow multiple overlapping directories so no curator holds exclusive gatekeeping power.
- Maintain abuse controls against fraud, malicious redirects, harassment and automated link floods.
- Correct broken links, misleading blurbs and falsely attributed claims, with public revision records.
5. Technical prototype specification
Static-first MVP: a searchable directory site rendering HTML on the server, backed by a versioned links.json file; Atom/RSS reader; optional OPML import/export; a submission form that does not auto-publish; lightweight full-text topic search; and accessible 'show source / show criticism / show related ideas' navigation. Store editorial notes in Git or a similarly auditable repository. Validate outgoing URLs, prevent cross-site scripting in metadata, and check links without excessive scraping or intrusive user analytics.
Submission workflow: (1) author proposes URL and rationale; (2) editor checks access, provenance, originality, credibility labels and copyright boundaries; (3) second editor evaluates any controversial subject; (4) public listing carries selection rationale and review date; (5) correction requests are triaged with visible outcomes. A contribution can be selected for intellectual relevance even if editors disagree strongly with its conclusions; misleading claims still receive meaningful labels. Never promise open inclusion of harmful or fraudulent operations.
AI layer: optional retrieval draws only from permitted catalog metadata and linked pages with lawful access. Every generated overview links to original documents and marks missing source text as missing. When a document contains disputed claims, the interface offers a counter-source. AI should not hallucinate links, metadata or permission to train on content.
6. Twelve-month empirical pilot
| Period | Delivery | Testable success criterion (pilot target, not result) |
|---|---|---|
| Months 1-3 | 20 founding publishers, 200 curated records, editorial charter | 95% valid source links; 100% conflict disclosure |
| Months 4-6 | Catalog API, feeds, multilingual schema, correction forms | At least 5 languages; median correction handling under 14 days |
| Months 7-9 | Optional Explore Alternatives navigation and AI source tracing | 90% of AI claims attributable to real linked documents in audit sample |
| Months 10-12 | Independent source-diversity audit and external replication | 20% relative increase in unique independent publishers encountered versus controlled baseline, if causally supported |
All targets are exploratory and should be revised after feasibility testing. Randomize volunteers between a conventional topic directory and diversity-aware routes; count unique credible sources reached, link-following to the original, comprehension and susceptibility to false balance. Do not count manipulative reciprocal links or traffic from bots as success. Use aggregate or consented metrics, and assess whether the system creates new ideological echo chambers.
7. Risks and funding
Risks: curator bias, spam, broken links, coordinated narratives, defamation, malicious pages, and gatekeeping by high-status founding members. Mitigate through plural editorial boards, distinct topic directories, provenance labeling, transparent complaints, rotating reviewers and portable catalogs. Budget a 12-month small pilot at $55,000-$125,000, assuming part-time editor, developer, accessibility/localization help, domain/hosting and periodic independent review. This is a planning estimate, not a market quote. Publish actual expenditure and in-kind contributions.
8. Recommendation
Start from CognitiveLiberties.com's existing network pages and original-source links, but place editorially useful outbound pathways at the center of the home and research experience. Run the linking initiative as a public service, not an SEO arrangement. Preserve the existing site's privacy choices as an option, while making voluntary sharing, discovery and external referral its visible core.
Sources
- W3C, Webmention, Recommendation. https://www.w3.org/TR/webmention/
- W3C, ActivityPub, Recommendation. https://www.w3.org/TR/activitypub/
- Google Search Central, AI features and your website and foundational SEO. https://developers.google.com/search/docs/appearance/ai-features
- CognitiveLiberties.com, Cognitive Liberty Network. https://cognitiveliberties.com/cognitive-liberty-network
- IETF, The Atom Syndication Format (RFC 4287). https://www.rfc-editor.org/rfc/rfc4287
Continue independent inquiry
Follow the cited primary sources, compare their evidence, and draw your own conclusions. Disagreement is part of discovery—not evidence of wrongdoing.
Browse the full research collection · Follow the resource navigator · Review corrections