Evidence and coverage
A map you
can inspect.
The sources, the limits and the work required to turn a promising idea into established knowledge.
What has been checked?
On 13 September 2026, the supplied 41-record manuscript collection was re-extracted and its text matched the existing reading corpus for all 41 records. Abstracts and conclusions or closing discussions were read, with missing continuation pages recovered. The PDFs contain 1,098 pages in total; this is extraction coverage, not a claim that every proof on every page has been independently verified.
The presentation groups the work into eight contributions. The register retains 41 previously consolidated research families and adds the 41 original numbered items from the five explicitly enumerated lists inspected here: hormesis (5), metabolic-control attenuation (6), the four-role architecture (17), aluminium pharmacology (6), and ecological currents (7).
Those counts overlap. Some statements are retrospective reclassifications; some are conditional mathematical consequences; some are prospective hypotheses. A separate index retains relevant passages on 471 main-text pages across the entire collection.
What does “complete” mean here?
Complete against this declared 41-record inventory and those five numbered lists. Keyword retrieval cannot certify that every implicit prediction, every equivalent formulation or every mathematical corollary has been found. The coverage record makes that limit visible, and the original passages make further corrections possible.
Evidence labels
- Conditional mathematics: a result stated under explicit premises. Website inclusion is not independent proof verification.
- Simulation: a result inside a specified computational model. Solver accuracy and implementation checks matter.
- Secondary analysis: a reported analysis of existing data, with its original sampling and measurement limits.
- Hypothesis or prospective test: an empirical commitment awaiting an adequate test.
- Review or synthesis: an organization or interpretation of existing results, with any normative premise retained.
Corrections belong in the achievement record.
The later work restricts earlier universal-coordinate claims, separates current from abundance, distinguishes transient recovery from durable fate, and preserves failed predictive additions. The historical aluminium “elliptic signature” is retained with an explicit warning: its stated geometric rationale is unsupported. These corrections are part of the programme’s development.
What is not certified?
Global priority, independent proof checking, clinical efficacy, universal physical coverage and online-version parity. An archive hash certifies source identity only. No number of manuscript records substitutes for these separate tasks.
Publication and privacy
This is the new public presentation of the boundedness programme. The Observatory’s previous code and experiment history are preserved in their existing repositories. Only the supplied manuscript content and already public programme results are included. The site itself is informational: its two dials execute declared local formulas and do not call hosted AI or collect enquiry records.
Third-party fonts are loaded from Google Fonts; the artwork is AI-generated illustration. No analytics or advertising scripts are added.
Reproduce or correct the map
Download the public content and source receipts, or inspect the repository. The static site is generated from that content. A correction should identify the paper, version, page and affected claim, with a reason and supporting evidence.