HCD.2 - Compose and Compare Human Capability-Development Programmes for Later Work

Type: Human Capability Development practitioner programme-composition pattern Status: Stable

HCD.2:0 - Use This When

Use this pattern when one named person has a recoverable later-Work demand and a practitioner, manager, or that person needs several feasible ways to develop the required contributions. Begin with the person’s current mastered and supported configuration, the complete later-Work tasks, and the conditions of participation; do not begin with a topic list or a preferred course.

The first useful result is a programme-comparison account: comparable descriptions of the actual programme candidates, their intended task closure, contribution and access allocations, burdens, uncertainties, evidence needs, applicability conditions, and the smallest change that would reopen the comparison. If only one candidate remains feasible, describe that candidate and the missing comparison rather than inventing a rival or calling it dominant. A separate authorized choice consumes the comparison.

The practical gain is a programme that can be inspected as a finite change from what the person can already do with available support. The practitioner can see whether a proposed component completes a usable contribution, preserves a prerequisite, adds a future option, or merely displaces practice that another required result still needs. When another start competes with unfinished learning, §4.2a compares continuing, pausing and starting together.

Do not use HCD.2 to design a universal course for an unknown audience, diagnose a person, choose on behalf of the decision owner, promise learning, or establish provider competence. When the current recipient is a learning-product author who does not yet know the learners, return the outputs, audience assumptions, existing alternatives, resource boundary, and individualization questions to HCD.28. It constructs and compares the whole learning product; its conditional variants and requirements can then supply this personal comparison. A named person’s compatible Work, entry performance, support, and participation conditions are the retry inputs for personal comparison.

HCD.2:0.1 - Working Terms

The programme, its description, a later decision to adopt it, the Work that enacts it, and observed human development are different objects. HCD.2 composes and compares descriptions; it does not make the later events true.

HCD.2:1 - Problem Frame

A person may need to recognize when a professional Method applies, retrieve a precise rule, perform a calculation, challenge an AI proposal, and obtain a specialist result in one piece of later Work. A syllabus can mention every topic and still fail to make that combination attainable. Conversely, requiring the person to memorize every detail can waste scarce time and remove useful support already present in the Work.

Programme candidates start from different baselines. One person may already perform the routine operation but miss a hazardous mismatch. Another may need a prerequisite before a whole task is useful. A third may be capable with a reference and specialist return but not under unsupported conditions. Time, participation, project access, teacher capacity, and the cost of later checking can change which programme is feasible.

The practitioner therefore needs to compose finite alternatives around complete work tasks, make their human and external contributions visible, and compare them without treating detail, duration, novelty, or a technique label as evidence of superiority.

HCD.2:2 - Problem

The common shortcut is to translate a demand into topics, allocate hours, and call the result a programme. This hides whether the pieces combine in representative Work, whether practice and correction fit inside the budget, and whether the assumed teacher, data, AI, specialist, or project access exists.

A second shortcut starts with one polished programme and compares every rival with its documentation rather than with the same required result. The detailed candidate then appears to dominate because its assumptions and costs are visible while the others remain sketches.

A third shortcut assigns an audience profile to every future learner. It converts a course-design hypothesis into a personal diagnosis and can recommend development without same-holder evidence. The practitioner needs a useful return for both situations: a personal comparison when compatible inputs exist, and a precise learning-product design question when they do not.

HCD.2:3 - Forces

• Complete tasks expose coordination and critical errors, while some prerequisites need preparation before meaningful whole-task practice.
• Internalized performance can reduce access delay, while reliable references, tools, AI, and specialists can be legitimate parts of later Work.
• A detailed programme is easier to inspect, while unequal detail is not evidence that it contributes more.
• Personalization can remove redundant work, while role labels and audience averages do not establish one person’s starting capability.
• A finite time budget forces displacement, while hidden reading, preparation, provider Work, and retries are easy to leave outside the count.
• Early supported performance can enable practice, while it does not establish later independent, changed-condition, or delayed performance.
• One candidate may be the only feasible option now, while feasibility alone does not establish dominance, effectiveness, or a final choice.
• Programme design can reveal missing access or provider capability, while the description does not make that support available.
• Future options can be valuable, while speculative option value must not replace the required near-term work result.

HCD.2:4 - Solution

Compose each candidate as a finite change from the person’s current mastered and supported configuration. Keep complete later-Work tasks, human and external contributions, practice and correction, resources, evidence limits, and future options visible on one comparison basis. Return the comparison without making the final choice.

HCD.2:4.1 - Bind the Personal Comparison

Name:

1. the exact human holder and the basis for their participation;
2. the receiving decision, decision owner, later-Work family, configuration, horizon, and protected result or critical-error conditions;
3. a compatible HCD.1 demand account or equivalent qualified later-Work input;
4. the current performance evidence, supported contributions, unresolved gaps, and any HCD.3 target or HCD.4 profile that the comparison actually uses;
5. the professional Methods, target-domain criteria, source access, tools, AI posture, specialists, provider contributions, environment, authority, and time that are available or still assumed; and
6. the development stake that the candidates must address without silently changing it.

HCD.1, HCD.3, and HCD.4 can supply these inputs, but their PatternIDs are not ritual prerequisites. Compatible direct evidence is enough for the positions the current comparison consumes. Use diagnosis only when rival causes would change the programme.

If personal evidence, participation, later Work, or support conditions are missing, stop the personal comparison at the affected claim. For an unknown audience, issue the learning-product design return to HCD.28. When a product variant comes back for a named person, test its entry assumptions, intended contribution, assistance and burden against that person’s compatible evidence before using it as a programme candidate. For a missing target-domain Method or correctness criterion, return the exact qualified outside-owner question. If no candidate can make the supplied development stake feasible, return that constraint and the surviving alternatives to the separate navigation or target decision rather than weakening the stake inside HCD.2.

HCD.2:4.2 - Compose Task-Closing Candidates

1. Select representative complete tasks. State the required result, important variants, critical errors, Work products, conditions, and stopping rules. A topic or Method name is only a discovery cue.
2. Decompose only far enough to allocate contributions. For every material contribution, state what the person must recognize and perform, what may be retrieved from a current reference, what controlled AI help may supply, and what requires a specialist or other provider. Keep the human check required for assisted output explicit.
3. Separate prerequisites from simultaneous complements. A prerequisite must be available before the relevant practice can proceed. A reference, tool, or specialist contribution may instead remain part of the same later Work. Do not teach every complement as if it had to become unaided personal performance.
4. Construct finite changes from the baseline. Add, substitute, remove, or retain programme elements relative to the person’s current configuration. Explain which task combination each change is intended to close, which future option it opens, and which existing contribution it might displace or weaken.
5. Include the complete development loop. Account for preparation, demonstrations or explanations where useful, representative attempts, feedback, correction, retries, changed-condition or later checks, and the access and provider contributions that make those activities possible.
6. Keep downstream elaboration explicit. HCD.5 can select the development Methods, HCD.6 can design the practice tasks, HCD.7 can establish available support, HCD.8 can compare obtaining or building a missing contribution and establish the preparation or operating test the selected route needs, and HCD.9–HCD.10 can specify practice and progression. HCD.2 needs enough of those contributions to make a candidate comparable; it does not pretend that an unperformed elaboration already exists.

Recover the relevant vertical within each complete task, using FPF B.1.5.EW when the connections are unclear. Ask what the person or available contributors must perform together while the encompassing work is under way. Adequate performance of isolated constituents can leave their coordination unavailable. Keep already attained contributions and include development only for the missing or insufficiently transferable ones. Include the variety of constituent Methods that the intended task variants require; a single rehearsed path may leave the person unable to choose a different continuation. Check shared time, attention, equipment and assistance as a combined demand rather than counting the same available resource separately for every part.

For example, in a stipulated dance case, strength and recall of the figure are adequate but balance during rotation is not. A programme adds suitable rotational-coordination practice and return to the figure under varied relevant conditions; it need not repeat the already mastered strength and recall work. HCD.3 distinguishes that gap from other causes, HCD.6 designs the practice and HCD.10 distributes its variation. A domain practitioner supplies the exercise and performance criterion.

When candidate programmes rely on different ways of developing the same contribution, carry the relevant HCD.4 local model into this comparison. Name which entry evidence makes each route usable, which qualitative change it proposes and what evidence supports the relations it relies on. HCD.6:4.2.1 supplies task and help alternatives; compare their whole learner/provider burden and the observation that could redirect them. Do not count crossed model nodes as equal units of gain or require every candidate to follow one presumed ladder. In the N24 case (HCD.6:5.6), contrast-first and supported-performance-first routes have the same target but different entry reliance, preparation and learning opportunities. The selected plan remains revisable; it is not the person’s recorded trajectory.

HCD.2:4.2a - Compare a New Start with Unfinished Learning

When another component would start alongside unfinished learning, recover what is still needed for the contributions already being developed before treating the addition as feasible. A learning line here is the remaining development towards one declared task contribution or a bounded future option. Several exercises can belong to one line; completing an exercise need not close that line.

Start from the person’s current programmes and recent attempts, including paused work that still needs maintenance or would incur a restart cost. For each relevant line, recover the intended later use and its threshold under allowed support, what the person has already demonstrated, and what remains unsupported. Then describe the remaining practice, feedback, learner correction and checks needed for that use. Include the provider and access windows that make them possible. Include a delayed or unassisted check only when the intended use requires it.

Compare the proposed new start with continuations that finish an existing contribution, preserve necessary practice, or deliberately pause or retire a line. Compare remaining benefit and burden; time already spent is not a reason to finish. A future option can justify a new start, but name what the bounded exploration will make possible or settle and when it returns for a decision. Calling every new topic an option leaves no stopping condition.

Construct a feasible boundary on concurrent learning for each candidate. Place the remaining attempts, feedback and correction in the available learner and provider windows; check attention blocks, shared tools and any assistance needed at the same time. Count a shared activity once when it serves several lines, while retaining different observations needed for different claims. Include maintenance, transition and restart costs of a pause. A sum of course hours can fit even when feedback arrives too late for correction or two lines need the same scarce practice window.

State which lines can remain active together and under what condition another can start: for example, an existing contribution is closed to its declared criterion, one line is paused with its consequences accepted, or additional usable capacity becomes available. The result can be a conditional set of compatible lines rather than a fixed count. Splitting one line into more modules does not change its remaining demand. Carry this boundary and its protected work into the programme comparison; the authorized decision owner still chooses whether to start, continue or pause.

Return actual attempts and help from HCD.9 to this account. A successful episode removes only the remaining work its evidence settles; the wider contribution can still need a changed-case attempt or another required check. HCD.11–13 supply stronger observations only where the intended claim needs them. Failed correction, a changed later task or a lost provider window changes the remaining work or feasible boundary through HCD.14/.15. When the required result is established, remove the completed work from the open set while retaining any needed maintenance. These changes can tighten or widen the boundary.

HCD.2:4.3 - Make the Descriptions Comparable

Describe every candidate on the same decision-bearing positions:

Unequal detail is itself a comparison gap. Develop only the smallest missing fragment needed to compare the same task, output, conditions, support, and resource coordinates. Do not require complete competing courses when one closed task family can discriminate the alternatives.

HCD.2:4.4 - Compare the Finite Programme Changes

Use C.11.CRC to compare each realizable candidate configuration with the person’s current configuration. Name the baseline, finite addition or substitution, candidate configuration, horizon, scenarios, result and resource coordinates, interactions, uncertainty, transition, reversibility, and option effects.

For HCD.2, the decision-bearing coordinates normally include:

• closure of the required later-Work outputs and protected critical-error conditions;
• learner, teacher, assessor, specialist, tool, and transition burden;
• dependence on references, AI, cues, project access, authority, and provider availability;
• preservation of current strengths and practice needed by other contributions;
• the qualified expectation and evidence plan for supported, independent, changed-condition, transfer, and delayed performance; and
• future development options and the cost of reopening a closed route.

Use dominated, non-dominated, beneficial, harmful, or indeterminate only for the candidates, coordinates, horizon, conditions, and evidence actually compared. A programme is not non-dominated merely because it was generated, and one feasible candidate is not dominant merely because no rival is ready. HCD.2 supplies the comparison claim; C.11 supplies any later bounded choice from the complete choice inputs.

HCD.2:4.5 - Return the Programme-Comparison Account

Return one account containing:

When a final choice is requested, pass this account to the authorized decision owner and the applicable decision Method. Preserve the option set, preferences, belief state, outcome model, protected conditions, probe worth, and authority required there. Do not turn the HCD.2 comparison into authorization, commitment, launch, or proof of effect.

HCD.2:4.6 - Recognition and Assurance

Recognition. Use HCD.2 when the live question sounds like “Which way could this person become able to do the whole job?” and the alternatives differ in preparation, practice, support, time, or future options. The smallest useful result may be one comparable task fragment rather than a complete curriculum.

Assurance. Stronger reliance needs same-holder and condition-compatible inputs, qualified target-domain tasks and error criteria, credible provider and access claims, complete resource coordinates, and explicit evidence limits. A programme can be compared before it is chosen or run. Effectiveness, transfer, retention, provider operation, and personal development require their own observations.

HCD.2:4.7 - What Changes in Practice

A practitioner stops asking, “Which course covers all the topics?” and asks, “Which finite programme change closes this person’s complete later-Work tasks, with what human and external contributions, at what burden, and what evidence is still missing?” A detailed course becomes one candidate; an unknown audience receives a design question rather than an invented personal profile.

HCD.2:5 - Archetypal Grounding

HCD.2:5.1 - Constructed Personal Case: Learner-FP6

Learner-FP6 is a constructed pedagogical holder. The stipulated entry attempt shows that this person can build a device model but confuses a flow unit, the recipient of the result, and the organizational permission needed for the proposed action. The intended later-Work outputs, mandatory tasks, 340 learner-hour limit, available teacher and worked examples, and protected critical errors are held fixed for this comparison. No observation of actual learning is claimed.

Two six-hour candidates replace harder optional cases rather than add time:

HCD.2 returns two comparable descriptions, the common output and evidence plan, and the unresolved provider-load and changed-condition evidence. It does not choose between them. An independent attempt showing three separable errors would strengthen focused preparation; success on the components but failure to combine them would strengthen the whole-task candidate. Either observation reopens only the affected comparison claims.

HCD.2:5.2 - Unknown Audience: FoundationsProgramme A, B, and C

A learning-product author has two intended outcomes: checked performance of complete selected work tasks, and lower effort to learn a previously untaught Method while preserving correctness. These are design aims, not observed gains. The author has an audience hypothesis, three 340-hour compositions, and no named learners. A allocates 202/96/42 hours to general/domain/joint application, B allocates 88/204/48, and C allocates 136/136/68. C is described in more detail; that detail does not establish dominance.

The HCD.2 boundary returns a learning-product design question to HCD.28: preserve the two outputs, audience assumptions, three alternatives, learner and teacher resource coordinates, support assumptions, and individualization questions. Use C’s existing material and develop only the smallest missing A/B fragment for one closed task family with the same output, entry assumptions, representative attempt, feedback, retry, access, and resource account. Do not fabricate one future learner or require three complete courses before comparison becomes useful.

If the learner budget falls from 340 to 204 hours while both outputs remain fixed, none of the three original candidates is launchable as described. Multiplying every activity by 0.6 or hiding reading and practice outside the budget does not close the task. The return asks for a new candidate that identifies the removed or replaced 136 hours and preserves the mechanisms needed for the outputs. The separate 240-hour teacher calculation does not replace learner time or establish a competent available service. A 170-hour ThinkingFoundations-only offer answers a narrower output question and therefore is not an improved version of the same candidate set.

HCD.2:5.3 - One Feasible Candidate Is Not a Winner

A named person has a qualified near-term engineering task and strong entry evidence. Only one candidate currently fits the deadline and available project access; a second candidate depends on an unavailable specialist. HCD.2 may return the feasible candidate, its finite comparison with the current configuration, and the exact specialist result that blocks the other candidate. It does not invent a weak rival, call the feasible programme universally best, or infer that the programme will work. A later provider result, deadline change, or changed task reopens the bounded comparison.

HCD.2:5.4 - Two Unfinished Lines and an Attractive New Start

In this constructed case, Lena has six learner-hours and two 45-minute sessions with a suitably qualified teacher in the coming week. The sessions can each return feedback early enough for a learner correction and subsequent check. The programme must preserve preparation for Friday’s report. A second line concerns work next month; a simulation method might open a further project option.

Recover the remaining contribution of each line:

All three fit the six learner-hours but need 135 teacher-minutes; only 90 are available. Counting A’s completed calculation and B’s completed query as closed learning would conceal this competition. There are at least two comparable continuations that preserve Friday’s requirement:

• Continue A and B; defer C. Four learner-hours and 90 teacher-minutes are allocated to the two remaining task contributions. Completion is intended, not guaranteed by those allowances.
• Continue A and open C; pause B. The same four learner-hours and 90 teacher-minutes supply A and the exploration. On the stipulated pause arrangement, preserving B’s partial work takes another half-hour now and returning to it requires a further half-hour next month. Its remaining practice is still owed, and its later use must tolerate that delay.

The two continuations have different closure, option and transition consequences. HCD.2 returns those differences and a boundary of these two compatible active lines, not a universal maximum of two or a choice made for Lena. If the next project now requires simulation while B’s later use is cancelled, opening C and retiring B can become preferable; unfinished status is not a reason to retain obsolete work.

Under the original work estimates, a confirmed third timely teacher session would make the three-line candidate feasible at six learner-hours and 135 teacher-minutes. This widens the available comparison without establishing that taking all three is preferable.

Suppose instead that A’s next attempt reveals another correction requiring one learner-hour and a further 45-minute teacher session before Friday. A now consumes three learner-hours and both sessions. The previous two-line continuations no longer fit. With the same help, a candidate can keep only A active and pay B’s half-hour pause cost: 3.5 learner-hours and 90 teacher-minutes. Retaining B or starting C as described requires additional suitable help or a changed arrangement. HCD.9’s result changes the remaining work; HCD.14/.15 reopen the affected comparison.

HCD.2:6 - Bias-Annotation

Lenses: Gov, Arch, Onto/Epist, Prag, Did. Scope: personal human-capability programme composition and comparison under qualified later-Work, participation, support, and evidence conditions.

HCD.2:7 - Conformance Checklist

• One exact human holder is named for a personal comparison; an unknown audience receives a learning-product design return rather than an invented profile.
• The later-Work family, supplied development stake, protected outputs and errors, configuration, horizon, participation, evidence window, decision owner, and outside owners are explicit.
• The baseline contains only supported current human and external contributions under stated conditions.
• Every programme candidate closes or tests a recognizable complete-task combination rather than merely covering topics.
• Human recognition and performance, reference retrieval, controlled AI help and checking, specialist returns, and other provider contributions are allocated explicitly.
• True prerequisites are distinguished from supports and complements used during the same later Work.
• Preparation, representative attempts, feedback, correction, retries, changed-condition or delayed checks, and provider/access work are inside the candidate account.
• Candidates use the same decision-bearing output, error, burden, support, horizon, uncertainty, and future-option positions; unequal detail is reported as a gap.
• Learner, teacher, assessor, specialist, transition, tool, material, and displaced-work burdens are visible where they change the comparison.
• When another start competes with unfinished learning, the comparison recovers remaining work, compares closure with opening an option, and states which lines can remain active together, when another may start and what would change that boundary.
• Supported, independent, changed-condition, transfer, and retention claims remain distinct, and unobserved gains remain hypotheses.
• Any dominance wording is bounded to the candidates, coordinates, horizon, conditions, and evidence actually compared.
• The result contains a usable disposition and smallest reopen condition but no final choice, authorization, launch, provider-operation, or effectiveness claim.

HCD.2:8 - Common Anti-Patterns and How to Avoid Them

HCD.2:9 - Consequences

The practitioner gains a comparison that connects development activities to complete later Work, exposes external support before commitment, and preserves uncertainty without becoming silent. A small comparable fragment can be enough to improve a decision, so the method does not require building several complete courses first.

The cost is explicit composition work. Human and provider time, access, transitions, retries, displaced practice, and later evidence can make an attractive programme infeasible. Some requests return to learning-product design, provider enablement, target-domain owners, or a separate target decision rather than yielding a personal recommendation.

HCD.2:10 - Rationale

Development is configuration-relative. What one programme adds depends on what this person can already contribute, what support is actually available, which complete tasks matter, and which protected errors cannot be traded away. Starting from an empty curriculum exaggerates every addition; starting from a role average fabricates the holder.

Whole tasks reveal coordination and applicability, while claim-sized contribution allocation prevents the whole-task label from hiding references, AI, specialists, or prerequisites. Comparing finite changes keeps resources and displaced work visible. Separating comparison from choice protects preference, authority, commitment, launch, and evidence of effect as their own questions.

HCD.2:11 - SoTA-Echoing

The working question is: How should a practitioner compose and compare a development programme for one person’s consequential later Work without reducing it to topics or promising an unobserved effect?

The selected best-known line combines sufficiently whole complex-task practice with expertise-sensitive support, explicit correction and retesting where those mechanisms fit, and configuration-relative finite comparison. The maintained 4C/ID model and its fourth-edition presentation supply the serious complex-task design candidate: whole learning tasks, support for non-routine reasoning, timely procedural information, and additional practice for recurrent parts. Paas and van Merriënboer’s 2020 review (COG-05) supports changing guidance with prior knowledge and task complexity. Winget and Persky’s 2022 review (COG-06) supplies the bounded formative-assessment, corrective-activity, retest, and threshold arrangement. HCD.2 adapts these contributions into §§4.2–4.3; none establishes later professional performance or the effect of a local programme.

The serious ordinary alternative is a syllabus-first or branded-programme comparison: list content and hours, match them to a role profile, then prefer the most complete or familiar package. It is cheaper to describe, but it hides complete-task closure, same-holder starting evidence, contribution allocations, provider dependence, displaced work, and unequal detail. HCD.2 rejects topic count and programme labels as the comparison basis, while retaining a syllabus as one carrier of a candidate once its contribution is made explicit.

C.11.CRC supplies the internal finite-comparison grammar used in §4.4. The HCD evidence chain keeps intended instruction, supported attempts, independent performance, transfer, and retention in their direct pattern owners; HCD.2 carries only the expectations and evidence needs used by the comparison. Bastani and colleagues’ 2025 school-mathematics trial (AI-04) is counterexample evidence that improved AI-assisted practice can fail to improve later unassisted performance; its population and task do not supply an adult workplace effect size or a universal AI prohibition. These limits produce the separate evidence states in §§4.3–4.6 rather than a promised causal programme effect.

This comparison adds design and evidence work, but it is non-dominated for the stated question because it preserves complete-task usefulness, personal fit, support dependence, finite burden, and uncertainty that the cheaper syllabus-first alternative loses. Reopen the selected line if a stronger source changes the complex-task, support, correction, AI-dependence, or transfer boundary, or if representative use shows that the same decision values can be preserved with materially less effort.

HCD.2:12 - Relations

• HCD.1: can supply the compatible human capability-demand account and later-Work boundary; programme composition does not alter that demand.
• HCD.3: can supply a qualified human target or non-training return when rival causes change the programme; diagnosis is not mandatory otherwise.
• HCD.4: can supply the current/target claim comparison, protected strengths, priorities, and trade-offs; a profile does not generate programme alternatives.
• HCD.5: selects human-development Methods for the target contribution and evidence limits exposed by a candidate.
• HCD.6: designs representative practice tasks, permitted help, variations, and result criteria needed by the selected Method.
• HCD.7: establishes which provider, source, tool, AI, access, and human-check contributions are actually available.
• HCD.8: compares obtaining or building an exact unavailable contribution and establishes the preparation or operating test the selected route needs; already established usable support can be arranged directly through HCD.7.
• HCD.9 and HCD.10: carry out focused practice and change its variation, spacing, interleaving, challenge, or help while preserving attempt history.
• HCD.11–HCD.13: provide compatible observed performance, transfer, retention, and support-dependence evidence; the programme description does not manufacture those observations.
• HCD.14 and HCD.15: revise affected programme assumptions and parts, or reopen development when later Work or support changes; changes in remaining work or available help can revise the boundary on concurrent learning.
• C.11.CRC: supplies the finite configuration-relative comparison claim. C.11 separately supplies a bounded choice when its preference, belief, outcome, probe, and authority inputs exist.
• PSD and other advice Methods: may recommend a programme to a named recipient from qualified inputs; advice does not compose the missing programme or make the choice for its owner.
• HCD.28: constructs and compares a whole learning product for a future audience, using HCD.5–8 and other applicable contributions. HCD.2 returns the unknown-audience question there and consumes suitable variants only with compatible personal evidence.
• Learning-product and target-domain practices: supply applicable instructional/carrier construction and professional task correctness. HCD.28 coordinates the needed product contributions; HCD.2 returns an exact missing domain question without inventing its answer.
• Provider, clinical, safety, legal, employment, specialist, organization, and release owners: retain their direct competence, participation, authority, and consequence decisions.

HCD.2:End