8000 Create a first-class research-notes system with institute cross-linking and trust labels · Issue #4 · Mnehmos/Mnehmos · GitHub
[go: up one dir, main page]

Skip to content

Create a first-class research-notes system with institute cross-linking and trust labels #4

Open

Description

@Mnehmos

Summary

The site has a blog, but the current research program needs a structured publication layer rather than isolated portfolio posts.

Create a research-notes system that can publish ongoing machine-mathematics work with clear evidence status, reproducibility links, and cross-linking to the Mnemosyne Research Institute.

Required content model

Each research note should support:

  • title, summary, publication date, last-reviewed date
  • authors and institute affiliation
  • research program
  • tags and theorem/problem identifiers
  • evidence status
  • repository, commit, issue, dossier, and artifact links
  • environment/toolchain hashes where relevant
  • prior-art and citation-review status
  • disclosure status for proof bodies and benchmark-protected material
  • related notes and supersession links

Trust labels

Support visible, consistent labels such as:

  • kernel_verified
  • certificate_checked
  • externally_cited_unreviewed
  • human_reviewed
  • experimental_evidence
  • informal_strategy
  • open_obligation
  • superseded

These labels must describe evidence without styling unverified claims as equivalent to proof.

Site changes

  • Add a dedicated /research or /notes index separate from generic projects.
  • Add program filters for Proof Search, MathCorpus, open conjectures, Erdős campaigns, ANLU, OTCF, and applied verification.
  • Add RSS/feed support if not already present.
  • Add social preview metadata and canonical URLs.
  • Add institute branding and links on every research note.
  • Add a Cite this note block with a stable citation format.
  • Add a visible correction/supersession mechanism.

Initial notes to support

  • Shift-8 forensic proof-search case study
  • Erdős 647 campaign report
  • Proof Search MCP architecture and zero-to-one trajectory thesis
  • MathCorpus machine-curriculum and RL-transition design
  • Wall mapping as a research objective

Acceptance criteria

  • Research notes are distinguishable from product/project pages.
  • Evidence labels are rendered consistently and documented.
  • A note can be updated without silently changing its original publication date.
  • Superseded or corrected claims remain auditable.
  • Institute and repository links are present.
  • npm run build passes.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions

      0