Methods · How to read this site
Every claim on this site sits somewhere on an evidential pyramid.
Research passes through a fixed sequence of layers, from an open question to a public-facing explainer. Each layer has an explicit standard of evidence and a distinct visual treatment. A claim only advances a layer when the layer below is strong enough to carry it. Most work will spend most of its life in the middle of the pyramid; that is not a failing of the work, it is a property of honest science.
The evidential pyramid
The pyramid is a permission structure, not a checklist. A topic is not required to reach the top; it is required to earn every layer it publishes. In particular, no explainer is written for a claim whose underlying evidence would not survive the L1–L4 tests.
-
L0
Open question
The question stated plainly. What would settle it. Why it matters. No claims made.
Standard of evidence: the question is well-posed and not already answered.
Serves everyone · shortest form
-
L1
Pre-registration
Frozen hypothesis, protocol, chosen dataset or population, and the falsifier that would count as disconfirmation. Recorded before results are known. No outcome data attached.
Standard of evidence: an external timestamp, or a hash committed publicly on a date the site can cite.
Serves peers · methodologically legible
-
L2
Primary evidence
Data, code, benchmark, replication package. Whatever a competent reviewer would need to try to falsify the L1 hypothesis themselves. Where licensing or clinical constraints prevent full release, the L2 page states exactly what is available, what is not, and why.
Standard of evidence: the record is complete enough for an independent replication attempt, or the gap is named and justified.
Serves peers and technical collaborators
-
L3
Research record
What the evidence showed against the pre-registered falsifier. Limitations named. Results that did not fit the hypothesis reported as such. Next test, if any, stated with its own falsifier.
Standard of evidence: an outcome that can be checked against the L1 record, and would be published identically whether it supported the hypothesis or refuted it.
Serves peers · read alongside L1 and L2
-
L4
Specialist essay
A dense, cite-dense argument that places one or more L1–L3 records into their scientific context. Distinguishes established results, this site's conjectures, and testable propositions using an explicit claims box.
Standard of evidence: every non-established claim links to its own L1 or L3 entry, or is labelled as conjecture.
Serves peers and technical clinicians
-
L5
Cross-domain synthesis
Where the same structural pattern shows up across two or more of the pillars, and each instance is itself supported by an L4 essay or an L3 record. This is where the QBD research-engine worldview reads through: the argument that recurring structure across biological, physical and computational systems is a fact worth naming, and the specific pattern being named.
Standard of evidence: at least two L4 or L3 anchors under the pattern, and a stated way the cross-domain claim could be wrong.
Serves peers, technical generalists, and collaborators
-
L6
Public-facing explainer
Plain-language rendering of a claim whose L1–L4 layers already exist and are cited from the page. Written to be understandable without prior specialist knowledge, without inflating the strength of what the lower layers actually support.
Standard of evidence: no claim appears at L6 that is not already carried by a linked L4 or L3 entry on this site.
Serves generalists · longest form · lowest priority when time is scarce
Reading the site. Each publication carries a layer badge (L0 through L6) above its title, and a status tag (see below) beside it. A page at L4 that cites an L3 record links to it directly; a page at L6 links to every L4 or L3 entry it depends on. If a claim's linked layer is missing, the claim is not made.
Status taxonomy
The pyramid says how strong a claim's evidence is. The status tag says where the object itself is in its life. Publication state and evidence state are separate axes and are labelled separately.
| Tag | Meaning | When it changes |
|---|---|---|
| planned | A page exists for a research object that has not yet begun. States the question, the layer at which it will first appear, and the target quarter. | Advances to pre-registered when L1 is written, or is removed if abandoned. |
| pre-registered | An L1 record exists with a frozen hypothesis and falsifier, timestamped externally. | Advances when primary evidence (L2) is complete enough to publish. |
| in progress | L2 evidence is being generated. Status is public, results are not. | Advances to under review, published, or superseded. |
| under review | An L3 or L4 draft has been submitted to a named venue or to specific peer readers. | Advances to published, or returns to in progress with reasons. |
| published | An L3, L4, L5, or L6 object is stable and citable. May carry a DOI. Revisions are versioned, not silent. | Advances to superseded if a later version replaces it substantively. |
| superseded | Kept accessible for the record. A visible pointer names the object that replaces it. | Terminal. |
Every page carries a last-updated date, and that date is refreshed even when nothing else changes on a monthly review. A page that has not been touched in more than four months without a stated pause carries a one-line pause note.
Corrections and versioning
Errors are corrected in place with a dated changelog line at the foot of the object. Substantive revisions — a changed result, an added exclusion, a withdrawn conclusion — produce a new version with its own layer badge and status; the previous version remains accessible and is marked superseded with a pointer forward. A published object is never quietly rewritten.
The site does not attempt to be a preprint server. Where a persistent identifier is warranted, it is issued through an external primitive (a public repository commit, or a DOI) and the site links to that record rather than mirroring it.
Holdback and release discipline
Some work is withheld until it can be published in its intended form and forum. The site is transparent that this holdback exists, and does not attempt to hint at withheld conclusions inside otherwise-public work. Where a research object depends on a piece of framework or data that is not yet released, that dependency is named and the object waits.
Independent research on financial systems, where any is undertaken, is treated as a constrained research domain and is not written or presented as a personal recommendation, market forecast, or invitation to invest. Quantum Biodynamics is not an investment vehicle, does not offer financial advice, and does not list adviser relationships.
What Quantum Biodynamics is, and is not
Quantum Biodynamics Ltd is an independent research company led by Dr Robert English. It is a small operation by design. Work is done in the open when doing so improves the work, and kept private when doing so protects the people or systems it touches. The company does not offer services, take on external mandates, or accept unsolicited pitches.
The pyramid, the status taxonomy, and the release discipline described on this page are how that work is organised. They are also how the site itself is meant to be judged: not by the count of published objects, but by whether each object sits honestly at the layer it claims.