Research
The internal publication programme: question, model, campaign, result, and explicit claim boundary.
Open topology →INDEPENDENT COMPUTATIONAL R&D / KAZAKHSTAN
We build purpose-made models, run controlled computational campaigns, and establish what actually follows from the equations—including negative findings and validity boundaries.
PUBLIC PROGRAMME STATE
Internal research, independent audits, and the educational environment are separated by provenance, purpose, and admissible claim strength.
02 / EVIDENCE DOMAINS
The site is organised by evidence provenance and status, not merely by subject.
The internal publication programme: question, model, campaign, result, and explicit claim boundary.
Open topology →Independent evaluation of an external hypothesis against predeclared controls and claim-admission criteria.
Open audit ledger →Normalised educational instruments for learning controls and bounded inference, without presenting demonstrations as experiments.
Enter Phase Lab ↗03 / CURRENT EVIDENCE
A compact view of the current corpus; full context, sequence, and archive remain in the research topology.
1 completed audit
Independent numerical audit of a collaborator-proposed FLG interface-strengthening hypothesis: interface organisation supported, mechanical strengthening not supported.
A reproducible interface-organisation signal survives the public controls, but three mechanically distinct downstream transfer models fail their predeclared matched-control gates.
Audit dossier →04 / CLAIM PIPELINE
A favourable trajectory is admitted to a conclusion only after balance, comparative, and numerical controls.
Translate the idea into state variables, observables, assumptions, and an outcome that can fail.
Choose equations, discretisation, boundaries, initial conditions, and diagnostics appropriate to the question.
Add passive, null, negative, ablation, conservation, and numerical-validity controls before interpreting a favourable branch.
Sweep parameters, seeds, perturbations, and resolutions; retain negative and failure results.
Separate model-level evidence from assumptions, artefacts, untested extrapolation, and laboratory claims.
Deliver reproducible methods, diagnostics, limitations, and an explicit laboratory or engineering follow-up.
Only reproducible, robust, and bounded results pass. A negative finding remains part of the evidence record.
Full method →05 / APPLIED R&D
The project outcome is an auditable evidence package, not a promise of a favourable conclusion.
A falsifiable question, minimum defensible model, and numerical proof-of-concept.
Purpose-built modelling, parameter sweeps, controls, and an operating-region map.
Stability, sensitivity, numerical artefacts, and negative findings.
A technical report, physical-validation plan, or staged R&D programme.
06 / AUTHOR & DISCLOSURE
Independent researcher and author of Fieldmind Lab, Kazakhstan.
The site explicitly separates internal research, external audits, normalised demonstrations, and physical validation that requires a separate stage.
Authorship and disclosure →07 / PROJECT ENQUIRY
Start with a non-confidential problem brief. Response within 3 working days.