HCD.1 — Derive Human Capability Demand from Representative Later Work

Type: Human Capability Development practitioner method pattern Status: Stable

HCD.1:0 - Use This When

Use this pattern when someone asks what a named person must become able to contribute in later Work and the available request is still expressed as a job title, competency list, course request, platform score, MethodDescription, or technology rollout. Begin with the exact human holder and representative later Work, not with a learning product.

The first useful result is a human capability-demand account recording which contributions that person needs to make in the named Work family, configuration, conditions, and horizon for the receiving decision. When the future Work is well framed but current evidence is not yet available, return a prospective demand hypothesis and the observations to obtain. When the question belongs to a clinical, target-domain, authority, organization, assignment, access, tool, or nonhuman-capability owner, return it there. When a conclusion consumes evidence that is absent, name each missing value and its retry predicate instead of calling the gap a learning need.

The practical gain is a demand that can guide diagnosis or profile comparison without pretending that training has been selected or that capability has already been observed. Use HCD.3 when the demand is known and the question is which changeable human limitation, if any, is supported. Use HCD.4 when the demand is known and simultaneous capability claims must be compared. A separate programme comparison (HCD.2) may use the account; deriving demand does not select a programme.

Do not use HCD.1 to prescribe diagnosis, treatment, curriculum, provider, employment action, safety judgement, release decision, or target-domain correctness. Those decisions remain with their direct owners.

HCD.1:0.1 - Working Terms

Name in this pattern	What it denotes
human holder	One identified human person whose capability demand is being considered. A role, cohort, roster, demographic category, or population is not that person.
later Work	A named family of future or recurring `U.Work` occurrences in which that person may contribute under stated conditions.
representative case	A case selected to examine a material variant of later Work, such as unfamiliar content, changed support, or a specialist interface. Its suitability depends on the conditions it represents and the conclusion for which it is used.
human capability demand	The contributions this person needs to be able to make under the named later-Work conditions. A demand account records those requirements; it does not establish the person’s current capability.
demand claim	A statement that this person needs to be able to make a specified contribution under the named later-Work conditions, with the evidence for that requirement, uncertainty, and limits.
prospective demand hypothesis	A provisional account of future demand grounded in a supported future-Work frame. It states which demand claims remain provisional and which observations could qualify or revise them.
demand-evidence-blocked	A result that names each missing input, the conclusion it blocks, and the condition for retry. It also distinguishes available material from the missing input when that material could mislead the receiver.
outside-HCD return	A return of the current question and available evidence to the direct clinical, target-domain, authority, organization, assignment, access, tool, or other owner.

The account records what later Work requires of the person. Evidence about present capability, causes, learning, or intervention effectiveness answers different questions and uses its own pattern.

HCD.1:1 - Problem Frame

Human development is often commissioned from a visible carrier: a course catalogue, certification, role framework, AI platform, assessment dashboard, or manager’s request. Those carriers can be useful, but none identifies by itself what a particular person must contribute in later Work.

Later Work is configured. The same named task can require different contributions when the product configuration, Method, information access, AI support, time, authority, specialist interface, or critical-error boundary changes. Assisted performance may also conceal which contribution belongs to the person and which belongs to a tool, another Agent, or an organization.

Demand therefore has to be recovered from representative Work and its intended result. Current observations are valuable, but future planning sometimes precedes an actual Work occurrence. The method must support honest prospective hypotheses without silently upgrading plans or demonstrations into current evidence.

HCD.1:2 - Problem

A role-first or course-first description bundles unlike things. It can call a missing Method, inaccessible source, broken tool, absent specialist result, poor assignment, unsafe workload, or organizational constraint a human skill gap. It can also average across a population and then assign the average to one person.

The reverse mistake is to refuse any demand account until the exact future Work has occurred. That discards useful architecture and delays the observations that would test it. The practitioner needs a result that distinguishes what can be framed now, what is qualified now, what belongs elsewhere, and exactly what evidence blocks the stronger conclusion.

HCD.1:3 - Forces

A short role label is easy to communicate, while the consequential demand usually lives in case variants, interfaces, and error boundaries.
A MethodDescription can make intended Work inspectable, while only enactment or qualified representative evidence supports claims about obtaining Work and its results.
Current task success is visible, while later transfer and unfamiliar variants must be established rather than presumed.
AI and other supports can improve the joint result, while assisted output is not automatically human capability.
Population summaries help allocation, while a person-specific decision requires a person-specific holder and compatible evidence.
Missing evidence should stop false certainty, while a useful prospective hypothesis may still guide observation.
HCD can organize demand evidence, while target-domain correctness, specialist judgement, assignment, safety, release, and employment authority stay elsewhere.

HCD.1:4 - Solution

Derive demand claim by claim from representative later Work. Preserve the truth posture of every case and source, separate human contributions from surrounding conditions and other owners, and return the strongest conclusion that the evidence supports.

HCD.1:4.1 - Pattern-Use Unfolding

Fix the use boundary and holder. Name the exact human person, receiving decision, later-Work family, subject configuration, effective horizon, and evidence window. If the question concerns a population, identify the separately observed members or representative accounts; do not manufacture one aggregate holder.
Recover the later-Work basis. For every case, distinguish what was observed from what is planned, stipulated, or assumed. Explain which later Work and material conditions the case represents, and why its evidence is suitable for the intended conclusion. Recover the intended result, quality and critical-error predicates, Work products, and stopping conditions.
Recover Methods and direct-owner inputs. Identify applicable target Methods or MethodDescriptions, specialist results, assignments, authority, access, environment, tools, and supports. A missing nonhuman value remains a missing nonhuman value; do not rename it a learning gap.
Select representative cases. Cover the variants that can change demand: familiar and unfamiliar situations, allowed and withdrawn assistance, time or workload limits, AI-output challenge, coordination interfaces, specialist-return conditions, and material quality or critical-error boundaries.
Derive human demand claims. For each case, state which contribution the named person must be able to make for the intended result under the protected conditions and why the evidence warrants that requirement. Contributions may include, for example, noticing a signal, handling an object, coordinating a specialist return, or justifying a decision. Record required tools and supports as conditions of the contribution. Keep each contribution distinct from the Method, task, tool output, team result, or personal trait.
Bind evidence and uncertainty. Relate each demand claim to the cases, observations, work products, inference rule, uncertainty, and remaining observation. Keep instruction events, learner events, learning outcomes, later transfer, and job performance distinct; one does not establish the next by label alone.
Return a conclusion-specific disposition. Emit qualified current demand, prospective demand hypothesis, outside-HCD return, or demand-evidence-blocked. Different questions may lawfully receive different dispositions in one use—for example, a prospective hypothesis for future framing and a blocker for qualified current demand.

HCD.1:4.2 - Record the Result

A human capability-demand account contains these result positions:

Result position	Required content
use boundary	Human holder, receiving decision, later-Work family, configuration, effective horizon, evidence window, and direct owners whose results or authority matter.
later-Work basis	Cases, what was observed or remains planned, stipulated, or assumed, and the later Work and conditions they represent; reasons the evidence is suitable for the conclusion; intended result, quality and critical-error predicates, Work products, and stop conditions.
protected conditions	Allowed or required tools, AI posture, information access, time, environment, assignment, authority, supports, and specialist interfaces.
demand claims	Statements of the human contributions required in the named later Work, including material interfaces among simultaneous contributions. Derive these requirements from the cases rather than assigning capability from job labels or traits.
evidence account	Evidence and inference for every demand claim, uncertainty, incompatible evidence, and the observation or case still needed.
external gaps	Missing Method, description, specialist return, assignment, authority, access, environment, or nonhuman capability kept under its own kind and owner.
disposition	One conclusion-specific `qualified current demand`, `prospective demand hypothesis`, `outside-HCD return`, or `demand-evidence-blocked`.
continuation	Compatible inputs that HCD.3, HCD.4, or a future programme comparison may consume, plus scope limits and reopen conditions.

A demand-evidence-blocked result records one row per absent consumed value:

Missing-value field	Required content
missing kind	For example `later-Work-occurrence-missing`, `later-Work-result-claim-missing`, `result-record-missing`, `representative-observation-evidence-missing`, or a named external supplier result.
blocked conclusion	The exact conclusion that cannot be supported without this value.
available non-substitute, when relevant	If available material could plausibly be mistaken for the missing input, name it and explain why it cannot fill this gap. Otherwise omit this field.
retry condition	The input that must become available or be confirmed before reconsidering the named conclusion. State its required connection to the person, Work, and conditions, using only qualifiers that affect that conclusion.

Do not infer result-record-missing when qualified direct observations are sufficient for the conclusion. Conversely, an orphan record does not establish that Work occurred or produced the result described in the record.

HCD.1:4.3 - Population and Reuse Boundary

A population use may return a distribution over separately identified member or representative demand accounts. It does not create one person’s demand by averaging, and it does not turn the population into a capability holder. A population whole qualifies as a holder only through the independent-System condition of E.23.CDI, outside this one-human-holder pattern.

Any downstream use must preserve the same human holder, configuration, later-Work conditions, effective horizon, compatible evidence window, and receiving decision. HCD.3 may consume the account as evidence input; HCD.4 may consume it as a scope and result frame. Programme comparison (HCD.2) may use the same frame. Deriving demand requires no programme-comparison result.

HCD.1:4.4 - What Changes in Practice

A practitioner stops asking, “Which course matches this role?” and first asks, “What must this person contribute in which later Work, under which conditions, and what evidence makes that demand current?” The answer either supports a bounded demand, makes a future hypothesis testable, returns the question to its owner, or exposes the exact missing evidence.

HCD.1:5 - Archetypal Grounding

HCD.1:5.1 - Constructed Positive Case: ImpactReview-17

ImpactReview-17 is a constructed pedagogical case. Its answer key, ImpactReview-17-Domain-Oracle-v1, supplies the engineering correctness and criticality criteria for PumpWorks-Config-Baseline-42 and AI-Assist-M4. The review events and evidence below are stipulated case data.

The human holder is Engineer-P14. The later-Work family is independent review of AI-proposed change-impact claims for mixed physical/software pump-controller changes before separate integration, assurance, and release decisions. The constructed 2026-Q4 conditions include the named configuration and AI assistant, current source access, a forty-five-minute case limit, recoverable audit records, and bounded safety and maintenance specialist returns. Final integration, safety, release, and employment authority remain with their direct owners.

The stipulated work permits a routine reminder to construct and test an alternative configuration hypothesis. The engineer must supply that hypothesis and its test; neither another person nor the AI may supply the case-specific rival or rejection decision. Remembering to initiate the operation without a reminder is a separate, stronger question.

The evidence packet stipulates eight representative review occurrences: four familiar and four unfamiliar variants; four include a plausible but wrong AI dependency claim; two require a specialist return; and every case requires a recoverable audit record. HCD.1 returns this human capability demand for Engineer-P14:

recognize when the independent-review Method applies;
trace each release-affecting claim to the current source and configuration;
independently challenge plausible AI output rather than treating supported output as the person’s contribution;
request, delimit, and preserve required specialist returns;
return auditable evidence; and
detect every seeded safety-critical omission within the protected time and authority boundary.

Within this example, the stipulated criteria and observations support qualified current demand for Engineer-P14. The case shows how to derive required human contributions from representative Work; demand does not establish the person’s current capability. Operational use needs qualified current target-domain criteria, safety or specialist inputs, and representative evidence for the actual person and Work.

HCD.1:5.2 - Missing Case: ReleaseWork-R9

For Engineer-E27, the accepted architecture supports a prospective demand hypothesis for 2026-Q4 ReleaseWork-R9: AI-output challenge, configuration-impact reasoning, evidence return, specialist coordination, critical-error detection, and adaptation across unfamiliar mixed physical/software variants may be required. This hypothesis guides observation; it is not qualified current demand.

The stronger conclusion is blocked by distinct absent values:

Missing value	Blocked HCD.1 use	Non-substitute and retry
`XRI-21-supplier-result-missing`	Any demand claim that consumes current engineering-culture enactment, its bounded practice-change decision, retained alternatives, or evidence gaps.	The current `SYSE.21` MethodDescription and its constructed PumpWorks demonstration can shape questions but are not an obtaining supplier result. Retry with a compatible obtaining `SYSE.21` result—or a receiver-permitted qualified direct substitute—for Engineer-E27, the Method, Work, period, and evidence use.
`later-Work-occurrence-missing`	Treating ReleaseWork-R9 itself as current later-Work evidence.	A plan, architecture, MethodDescription, or constructed occurrence does not establish enactment. Retry when an admitted performed Work occurrence identifies Engineer-E27, configuration, period, and actual performer.
`later-Work-result-claim-missing`	Any demand claim whose evidence consumes the occurrence-produced result, quality, error, or consequence.	A planned result or demonstration result is not the occurrence’s result. Retry with the produced result or a supported direct claim tied to the admitted occurrence.
`result-record-missing`	Any record-dependent inference about that occurrence or result.	An orphan document would not establish the Work or result. Retry when a current result record links the admitted Work occurrence to its produced result; omit this requirement where qualified direct observations are sufficient.
`representative-observation-evidence-missing`	Upgrading the prospective hypothesis to `qualified current demand`.	A target repertoire, curriculum, course completion, platform score, or constructed case is not representative observation evidence. Retry with holder- and condition-compatible observations or a named qualified representative case set covering performed Method, configuration, work products, quality and critical errors, AI/support posture, specialist returns, variants, and evidence window.

These rows do not collapse into “insufficient data.” Each names a different missing input, blocked conclusion, misleading substitute, and retry condition. Recovering one row changes only the conclusions that consume it.

HCD.1:5.3 - Unlike Case: Clinical Ownership

Return an adult’s request for treatment of social anxiety to a qualified clinical specialist as an outside-HCD return. HCD.1 may address a separately posed, legitimate question about that person’s non-clinical participation in later Work, supplied by the person or qualified clinical owner. Diagnosis, treatment choice, provider competence, safety, consent, and clinical outcome interpretation remain with the qualified clinical route.

HCD.1:6 - Bias-Annotation

Recurring bias	Likely drift	Repair
training-solution bias	The requested course or certification is treated as the demand.	Recover the human holder and representative later Work before considering any learning product.
role-average bias	A job-family profile is assigned to one person or one population average is reified as a person’s demand.	Keep holder-specific accounts and report population distributions separately.
current-arrangement bias	Today’s Method, tool, or division of labour is assumed to be the only future demand.	Name the effective horizon, alternatives, supports, and configuration; preserve prospective uncertainty.
technology bias	AI-assisted or tool-supported output is attributed to the human holder.	Separate human, tool, specialist, team, and organization contributions and test material support conditions.
evidence-convenience bias	Completion, self-report, platform prediction, MethodDescription, or visible document substitutes for later-Work evidence.	Trace each demand claim to representative cases, observations, inference, uncertainty, and remaining evidence.
human-deficit bias	Missing access, authority, assignment, Method, environment, specialist result, or nonhuman capability is called a skill gap.	Return the value under its actual kind and direct owner.
occurrence maximalism	No future question is allowed until the exact Work occurs.	Return a prospective demand hypothesis with observations-to-obtain, while blocking only the stronger current conclusion.

HCD.1:7 - Conformance Checklist

One exact human person is named as holder; any population use preserves member or representative accounts.
The later-Work family, configuration, effective horizon, evidence window, and receiving decision are explicit.
Each case distinguishes observed facts from plans, stipulations, and assumptions, and explains which later-Work conditions it represents and why its evidence is suitable for the conclusion.
Intended results, quality and critical-error predicates, protected conditions, tools, supports, and specialist interfaces are recoverable.
Each demand claim names a human contribution supported by representative Work rather than a job label, trait, task, MethodDescription, or course.
Human, AI, tool, specialist, team, organization, assignment, authority, and environment contributions or gaps remain distinct.
Each demand claim has evidence, an inference boundary, uncertainty, and any observation still needed.
The result uses one conclusion-specific disposition and does not suppress a lawful prospective hypothesis merely because qualified current demand is blocked.
Every blocker names the missing input, blocked conclusion, and retry condition; available material is distinguished from that input when it could mislead the receiver.
Downstream reuse preserves holder, configuration, conditions, horizon, evidence window, and receiving decision.
The first result selects no curriculum, provider, intervention, assessment programme, treatment, safety action, release action, or employment action.

HCD.1:8 - Common Anti-Patterns and How to Avoid Them

Anti-pattern	Why it fails	Repair
“The role requires these competencies”	The role label hides case variants, conditions, and the exact holder.	Derive claim-sized contributions from representative later Work for the named person.
“The course objective is the capability demand”	A learning-product intention is not a later-Work result.	Start with later Work and let a future programme comparison answer the demand.
“The MethodDescription proves the Work”	A description states an intended Method; it does not establish enactment, result, or observation.	Preserve its use as a question frame and recover the missing occurrence or qualified representative evidence.
“The tool score is the person’s capability”	The score may describe joint or model-dependent performance.	State support conditions and recover direct task evidence and inference limits.
“One missing record means no evidence exists”	Direct observations may support the conclusion without a record, while an orphan record may support nothing about the occurrence.	Name the evidence actually consumed and require a record only for a record-dependent use.
“The average employee is the holder”	Aggregation erases person, condition, and evidence identity.	Keep individual accounts or an explicit distribution; use `E.23.CDI` before treating a population whole as holder.
“HCD.2 must come next”	Numbering is mistaken for a lifecycle and invites premature programme design.	Route independently to HCD.3 or HCD.4 when their qualified inputs exist; programme comparison needs its own qualified basis and decision.
“Poor outcome means training”	Method, tool, assignment, access, organization, nonhuman capability, or specialist causes may dominate.	Return those values to their owners and use HCD.3 only when a diagnostic question remains.

HCD.1:9 - Consequences

The organization gains an inspectable demand account that can support diagnosis, profile comparison, evidence collection, or later programme comparison without making those later decisions in advance. Prospective planning remains possible, and missing evidence becomes a substantive return rather than an administrative “TBD.”

The cost is more careful case selection and relation keeping. Demand may vary across configurations, supports, horizons, and holders; some visible requests will return outside HCD. Evidence collection may reveal that the main constraint is not human capability.

HCD.1:10 - Rationale

Beginning from representative later Work preserves the direction of justification: intended results and conditions constrain the required human contributions; learning products do not invent their own demand. Holder identity prevents a role model or cohort aggregate from masquerading as a person’s capability.

The four dispositions avoid a false choice between unsupported certainty and silence. A prospective hypothesis is useful for future framing, while a discriminated blocker protects a stronger current conclusion. Keeping external gaps under their own kinds prevents HCD from taking authority over target Methods, specialist results, assignments, safety, release, or employment.

HCD.1:11 - SoTA-Echoing

The working question is: How should a practitioner derive a human capability demand that is useful for later development decisions and still warranted by later-Work evidence?

The selected current line combines representative-work analysis with the workplace-instruction evidence chain summarized by Kraiger and Ford’s 2021 review (COG-10): instructional events, learner events, learning outcomes, transfer, and job performance must be distinguished, and claims about each need corresponding evidence. Transfer must be established under material learner, design, opportunity, support, and work-environment conditions. Direct target-domain evidence supplies what counts as correct, safe, critical, or useful in the actual Work. Where AI assistance is material, Dell’Acqua and colleagues’ Navigating the jagged technological frontier (AI-03) supplies a bounded cue to test independent error detection under the named task and model conditions. It establishes neither a universal AI-task boundary nor this person’s engineering capability.

A serious operational alternative is a role-first competency framework or training-needs inventory. It is cheaper to publish and supports broad workforce communication, but it commonly loses holder identity, representative case variation, tool and specialist interfaces, critical-error predicates, and the distinction between a learning outcome and later performance. HCD.1 retains those inventories only as discovery cues and mutates them into evidence-bound claim rows.

The selected line does not establish one universal demand, transfer trajectory, or intervention for ReleaseWork-R9. It also cannot supply target-domain truth. Reopen a demand claim when its holder, Work family, Method, configuration, support posture, quality or critical-error predicate, effective horizon, evidence window, direct source, or receiving decision changes. A later stronger source reopens only the claims whose move or limit it changes.

HCD.1:12 - Relations

A.15.1 and current Work ontology: keep Method, Work occurrence, performer, result, and result episteme distinct.
E.23.CDI: governs whether a population whole is an independent System capable of holding capability; HCD.1 otherwise stays at one-human-holder granularity.
E.23.CAE: supplies capability-evidence boundaries for downstream observations; evidence does not itself authorize intervention or employment action.
COG-10: supplies the event-to-transfer distinction and the requirement to qualify workplace opportunity, support, and environment.
Dell’Acqua and colleagues, Navigating the jagged technological frontier (AI-03): supplies the bounded independent-error-challenge cue where assisted Work makes that contribution material.
Target-domain Methods and results: determine correctness, quality, criticality, and specialist boundaries. HCD.1 consumes them; it does not replace them.
SYSE.21 / XRI-21: an obtaining compatible cultural-continuation result may supply later-enactment observations, a bounded practice-change decision, retained alternatives, and evidence gaps. Its MethodDescription or constructed demonstration is not that supplier result.
HCD.3: may consume a compatible HCD.1 account as evidence input for a limiting-capability or non-training disposition; no target is entailed.
HCD.4: may consume a compatible HCD.1 account as the scope and result frame for a condition-qualified profile comparison; no mastery or choice is entailed.
Programme comparison (HCD.2): may consume the compatible demand account and evidence needs. HCD.1 derives demand without selecting a programme or requiring a programme-comparison result.
Clinical, safety, legal, employment, specialist, organization, provider, and release owners: retain their direct decisions and authority.