How DALLEAU reads evidence for private residences in Bali

Step 1. Start with the question

Method question

How does a research finding become a useful project question without becoming a false design rule?

One-sentence answer

We examine the question, study design, sample, setting, exposure, comparator, outcome, finding, limitations and what the study did not prove. Then we ask whether the result survives Bali’s climate, operation and household reality.

Step 2. Identify the study design

Official guidance or standard

Useful for accepted definitions, measurement methods, limits and professional practice. It does not automatically prove that every project target is a universal threshold.

Systematic review or meta-analysis

Useful for understanding consistency across multiple studies. Its strength still depends on the quality and comparability of the included work.

Field study in a real building

Usually carries strong relevance for residential translation because the environment, duration and behaviour are more realistic.

Controlled laboratory study

Useful for mechanism and controlled comparison. Transfer to long-term residential life may be limited.

VR, image or screen-based experiment

Useful for preference, perception and hypothesis generation. It should rarely carry a strong claim about a real occupied residence by itself.

Expert or vendor source

Useful for implementation context or current product capability. It is not independent evidence of a human benefit.

Step 3. Read the sample

Ask:

• How many people took part?
• Who were they?
• Were they healthy adults, older people, children, patients, office workers or another group?
• Were age, culture, chronotype, acclimatisation or prior experience relevant?
• Was the study large enough for its conclusion?
• Did many participants drop out?

A result in a small group of young adults does not automatically apply to children, older owners, international guests or staff in Bali.

Step 4. Check the setting

The setting affects transferability.

A full-season study in occupied homes answers a different question from a ten-minute VR exposure. A sealed office in a temperate climate answers a different question from an open or mixed-mode residence in Bali.

DALLEAU gives greater residential weight to evidence that includes:

• real buildings;
• real occupancy;
• realistic exposure duration;
• actual controls and behaviour;
• operational and maintenance context;
• relevant climate conditions.

Laboratory and VR work can still explain mechanisms. It should be labelled accordingly.

Step 5. Define the exposure

A useful summary records what people were actually exposed to.

For light, that may include timing, eye-level intensity, duration, spectrum and prior light history.

For sound, it may include average level, event peaks, low frequency, tonality and event count.

For thermal comfort, it may include air temperature, radiant temperature, humidity, air speed, clothing and activity.

For indoor air, it may include pollutant source, ventilation, duration, particle level, moisture or filtration.

Vague labels such as “healthy lighting”, “biophilic room” or “smart home” are not sufficient descriptions.

Step 6. Separate the outcome types

Four outcomes are often confused.

Preference

What people liked or chose.

Behaviour or performance

What they did, how quickly they completed a task or how they navigated.

Physiological response

A short-term measurement such as heart rate, skin conductance, hormone level or brain activity.

Long-term health or life outcome

A durable clinical or population-level result.

A preferred space is not automatically a healthier space. A short-term physiological change is not proof of a durable benefit. A building POE is not a clinical trial.

Step 7. Read the finding precisely

Ask:

• How large was the effect?
• Was it statistically and practically meaningful?
• Was the result consistent across outcomes?
• Was it replicated?
• Did the authors qualify it?
• Were alternative explanations considered?
• Was the study funded by a party with a commercial interest?

DALLEAU does not rewrite “associated with” as “caused”. It does not rewrite a preference result as a health result. It does not rewrite a short exposure as a permanent effect.

Step 8. State what the study did not prove

Every Study Note must include this section.

Examples:

• The study did not prove that the result applies to every household.
• It did not isolate one design feature from the rest of the intervention.
• It did not test long-term occupation.
• It did not take place in a hot-humid climate.
• It did not measure a medical outcome.
• It did not test construction quality, maintenance or staff operation.
• It did not compare the result across seasons.

This is not an exercise in weakening the research. It is how the result remains useful without being overstated.

Step 9. Assign the evidence level

Established evidence

Supported by authoritative guidance, standards, systematic reviews or multiple converging human studies.

Growing evidence

Supported by several credible studies, but still dependent on population, setting, duration or method.

Emerging research

Promising findings with material limitations, such as small samples, short exposures or laboratory and VR settings.

Research question

A plausible question worth testing or monitoring, not a basis for a universal project rule.

A page may contain mixed evidence classes. In that case, the page-level label is only a summary and each material section receives its own qualification.

Step 10. Assign the application type

Evidence maturity and project use are not the same thing.

Official standard or guidance

A requirement, method or recommendation issued by a recognised authority or standards body.

Evidence-based recommendation

A project recommendation supported by the current evidence but still requiring context and professional judgement.

DALLEAU project target

A target chosen for a specific project. It must not be presented as a universal scientific threshold.

Site-specific finding

A measured or documented condition from a named or anonymised project under stated conditions.

Project hypothesis

A question the project will test or monitor. It is not a promised result.

Step 11. Test the Bali transfer

Before a finding enters a residence brief, ask what Bali changes.

• Does high humidity change the result?
• Does open-air living undermine a sealed-building assumption?
• Does rain change the acoustic condition?
• Do insects or vector control alter the opening strategy?
• Does heat gain make a daylight solution unusable?
• Can the local team maintain the proposed system?
• What happens during owner absence?
• What happens during power or internet failure?
• Does staff operation change the intended mode?
• Does the project need local measurement before adopting the recommendation?

Global evidence becomes useful only after this filter.

Step 12. Translate the finding into a project action

A research conclusion should enter the project through one or more controlled objects:

• owner brief question;
• specialist scope;
• design criterion;
• DALLEAU project target;
• mock-up requirement;
• hold point;
• commissioning test;
• staff operating protocol;
• Building Passport record;
• POE question;
• unresolved-question register.

If the finding cannot influence a decision, test, record or question, it may belong in the library rather than the project brief.

When specialist review is required

Subject review is required when:

• the public page uses technical thresholds;
• the interpretation could affect health, privacy, life safety or regulated scope;
• evidence is mixed or controversial;
• the project application exceeds DALLEAU’s qualifications;
• measurements need specialist instruments or interpretation;
• the content discusses acoustics, circadian light, moisture, thermal modelling, age-ready design, cybersecurity or POE in technical depth;
• a named specialist or institution may appear publicly.

Legal and privacy review is required where the overall impression could imply medical benefit, where people-level data are collected or where project evidence may identify an owner, guest or staff member.

What this method can show

It can show:

• why a source deserves more or less weight;
• what question the source actually answered;
• what remains uncertain;
• what is transferable to a Bali residence;
• whether the result should become a standard reference, recommendation, project target or research question;
• what specialist or site validation is required.

What this method cannot show

It cannot:

• convert weak evidence into strong evidence;
• replace a subject specialist;
• establish a clinical outcome;
• guarantee that a recommendation will perform on site;
• remove individual and cultural variation;
• eliminate the need for commissioning and POE;
• prove that DALLEAU’s interpretation is permanently correct.

Unresolved questions

• How should DALLEAU score evidence quality without creating a misleading universal number?
• When should a source update trigger a full page rewrite rather than a correction note?
• Which topics require two independent subject reviewers?
• How should contradictory standards and field evidence be displayed without confusing non-specialist readers?
• Which local Bali measurements should be required before a recommendation enters the protocol system?
• How should reviewer disagreement be recorded and, where useful, disclosed?

Source ledger and corrections

The source ledger should be grouped by:

1. standards and official guidance;
2. systematic reviews and consensus papers;
3. human field evidence;
4. laboratory and VR evidence;
5. project evidence;
6. open questions.

Each entry should record:

• title;
• author or institution;
• year;
• source type;
• DOI or stable URL;
• claim supported;
• evidence level;
• known limitation;
• last checked date;
• correction or retraction status.

Material updates require a visible change note and a footer corrections log.

Sources

The source finding and DALLEAU's residential interpretation are kept separate. Links open the original publisher or authority record.

1. CIE System for Metrology of Optical Radiation for ipRGC-Influenced Responses to Light

   Established evidence. The measurement quantities and metrics used to describe stimulation of the five retinal photoreceptor types. Limit: A complete lighting application, a health treatment or prediction of a person-level outcome.

2. WHO guidelines for indoor air quality: dampness and mould

   Authoritative guidance. Prevention and remediation of persistent dampness and microbial growth. Limit: A universal safe microbial threshold or proof that hidden mould is absent.

3. Thermal Environmental Conditions for Human Occupancy

   Authoritative guidance. Thermal comfort as an interaction between temperature, radiation, humidity, air speed, activity and clothing. Limit: One thermostat setting that guarantees comfort for every person.

4. IoT Device Cybersecurity Capability Core Baseline

   Authoritative guidance. A starting set of device cybersecurity capabilities for IoT acquisition and integration. Limit: A complete project threat model, legal advice or a guarantee of security.

5. The Commissioning Process and commissioning requirements for new buildings and systems

   Authoritative guidance. Commissioning from owner requirements through acceptance, documentation, training and operation. Limit: Permanent performance under every future condition or a clinical human outcome.

Related content

• Evidence, review and corrections
• Methods Note: Why commissioning matters more than handover
• Protocols hub
• All four Study Notes

Closing

Better questions before stronger claims

DALLEAU Atelier uses research to sharpen the brief, define what must be tested and keep uncertainty visible.