← Back to catalogue
Published

Performance Objective / Review

vr.wm-org-017 · wm-org-017-performance-objective-review

Represent a governed performance case with scoped objectives, evidence, assessment, correction and purpose-limited disclosure.

World Models Society, people and institutions SOC.ORG.PERF

Bundle → Layer → Finding → Questions Filled

7 bundles · 14 layers · 24 findings · 72 questions

Identity and work context Scope-governed identity and work context for one performance case.

Identity Lineage

Model-owned context for identity lineage, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Case identity and authoritative master

A performance case has its own stable identity, distinct from person, period label and system row location. It is the root for scoped review records, not a global talent profile.

  1. Which authoritative namespace and issuer identify this performance case? identity
  2. Which system owns case facts and which identifiers are only aliases? ownership
  3. What distinguishes duplicate case registration from another valid review of the same person? validation

Revision and successor lineage

Revision changes assertions about one case; a new review cycle gets a separate case identity. Corrections link prior assertions without granting indefinite retention of their personal content. Use a policy-authorized minimal redacted assertion when content is erased; unauthorized loss of the predecessor reference is invalid. The retained linkage itself remains subject to policy.

  1. Does this change revise the existing case or establish a new review cycle? lifecycle
  2. When was the corrected claim valid and when did the master learn it? temporal
  3. Which earlier revision is superseded and under whose documented correction authority? provenance

Governing Bindings

Model-owned context for governing bindings, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Governing work context

Exactly one declared reviewing authority and governing work context per case. Employment is required only for employee cases; contractors and founders may use a referenced contract or mandate. Concurrent employment remains external context.

  1. Which employment, contract or founder mandate governs this review? relationship
  2. What authorizes the reviewing organization and reviewer in this work context? authority
  3. How is a contractor reviewed without inventing an employment relationship? exception

Subject and assignment bindings

Reference person, organizational units, positions, memberships and assignments with roles and valid time. No replicated person master, contract salary or credential secret. Multiple contextual roles do not imply multiple people.

  1. Which person and independently owned assignments are referenced by this case? composition
  2. During which intervals did each position or membership apply? temporal
  3. Which neighboring facts can be referenced without copying unnecessary personal data? privacy
Objectives, period and purpose Scope-governed objectives, period and purpose for one performance case.

Period Purpose

Model-owned context for period purpose, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Period and review instrument

The reviewed interval, operating timezone and program version are independent of record creation and rating issuance. A draft may have no assessment; profile and state guards govern required information.

  1. What interval and calendar define the work reviewed by this case? temporal
  2. Which versioned program and startup or enterprise profile applies? classification
  3. Which information is required now for draft, open assessment or issued outcome? state

Purpose and permitted use

Declare why the review exists before gathering evidence. Development, period appraisal, recognition and external reference are distinct purposes; downstream use requires an independent authority and compatibility decision.

  1. What explicit purpose governs this performance case and evidence collection? requirement
  2. Which reader purposes are compatible with that declaration? access
  3. Who decides whether a proposed new use is permissible and with what evidence? decision

Objective Plan

Model-owned context for objective plan, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Objective and success criteria

Objective items describe intended outcomes and qualitative or quantitative acceptance criteria. They may reference independently owned organizational goals; they do not become competency definitions or pay instructions. Record changes in opportunity, available work time, leave and dependency conditions where purpose-authorized, without copying health or accommodation details into a general review.

  1. What intended outcome does this objective express? definition
  2. What evidence distinguishes met, unmet and not-evaluable for this objective? measurement
  3. Which dependencies or resource assumptions limit responsibility for the outcome? constraint

Objective changes and shared alignment

An objective revision records who changed expectations and when. Shared team outcomes are linked and contribution is asserted separately; they are not multiplied into private achievements.

  1. Which team or organization objective does this item support without duplicating its master? relationship
  2. What changed the agreed objective or its success criterion during the period? event
  3. Who acknowledged the revised expectation and its effective date? authority
Evidence and attribution Scope-governed evidence and attribution for one performance case.

Evidence Quality

Model-owned context for evidence quality, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Evidence origin and observation

The case records an evidence citation and assessment of its relevance; the original document, log or work product keeps its own identity and master. Record method, timing and unit of analysis.

  1. Which observed work product or attestation supports the claimed result? evidence
  2. Who produced the evidence and by what collection method? provenance
  3. When did the observation occur and when was it recorded for review? temporal

Evidence quality and proxy exclusion

Evidence may be admitted, contested or excluded with reasons. Raw activity or surveillance proxies are not a sole or automatic basis for assessment. Missing evidence is unknown, not zero contribution.

  1. Is this evidence direct, self-reported, derived, contested or excluded? quality
  2. Would the proposed rating rely only on activity counts or surveillance proxies? constraint
  3. What non-proxy evidence supports the assessment and what uncertainty remains? validation

Credit Conflict

Model-owned context for credit conflict, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Attribution and shared credit claims

Attribution binds evidence to a subject, role and work context. Shares may be unknown; any fraction requires a denominator and method. Contribution does not establish IP ownership or automatically allocate revenue.

  1. Who claims which contribution in which employment or assignment context? relationship
  2. If a numerical share is proposed, what denominator and allocation method justify it? measurement
  3. Which attestation supports or contests the contribution claim? evidence

Conflicting evidence and unresolved claims

Keep competing source-qualified claims visible. A human may issue a qualified review with recorded uncertainty, but cannot label it uncontested or silently discard an equally authoritative contrary source.

  1. Which claims conflict about the same result or contribution? evidence
  2. Who adjudicated the conflict and what reasons were preserved? decision
  3. How is an issued outcome qualified when the dispute remains unresolved? exception
Assessment and scales Scope-governed assessment and scales for one performance case.

Reviewer And Mode

Model-owned context for reviewer and mode, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Reviewer assignment and recusal

Reviewer authority is scoped to the case, items and interval. Manager, peer, customer and appeal officer roles are not interchangeable. A role assignment does not prove an assessment was performed.

  1. Which assignment grants this rater authority over these review items? authority
  2. What evidence is this rater permitted to read for the declared purpose? access
  3. What conflict of interest requires recusal or another accountable reviewer? exception

Assistance versus accountable assessment

Agents may organize evidence and draft questions. A human author issues review outcomes under explicit authority. AI assistance is disclosed; no recommendation executes hiring, dismissal, promotion or pay.

  1. Was this assessment drafted by a person or prepared with agent assistance? process
  2. Which accountable human accepted the evidence and issued the review? authority
  3. What prevents an agent-produced suggestion from executing an employment decision? constraint

Instrument And Assertion

Model-owned context for instrument and assertion, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Contextual scale version

A narrative-only review requires no scale. When a numeric or ordinal rating is used, pin exactly one applicable version, anchors, scope and method. Capability levels and grades are referenced concepts, not performance scores.

  1. Does this profile use a rating scale or narrative-only assessment? classification
  2. Which immutable scale version defines the meaning of this rating value? measurement
  3. What explicit mapping is needed before comparing two incompatible scale versions? interoperability

Assessment assertion and issue guard

Assessments are 0..n in a case, with at most one current issued outcome for the chosen release profile. Prior outcomes remain referencable subject to retention policy. A period rating is never a permanent person property.

  1. Is this assertion draft, issued or superseded? state
  2. Which admissible evidence and unresolved conflicts support this outcome? evidence
  3. Which period, purpose and scale version bound the issued assessment? temporal
Calibration and recognition Scope-governed calibration and recognition for one performance case.

Calibration Outcome

Model-owned context for calibration outcome, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Calibration acts and session history

Calibration is an accountable before/after act with reasons. Allow repeated sessions and preserve dissent. A session can reference several case outcomes without owning their identities or exporting their full dossiers. A multi-case Calibration Session is a peer object with its own population and policy. Only the case-local calibration outcome belongs to this aggregate. For post-issuance calibration, the after assertion supersedes the before assertion and the before status becomes superseded.

  1. Which prior assessment was changed by this calibration act? event
  2. What criterion, human panel and reason explain the adjustment? decision
  3. Which case-specific slices may the panel read without receiving unrelated employee records? privacy

Outcome and downstream effect boundary

Issuance records feedback or a qualified review, not an executable personnel instruction. Reward, staffing, compensation and legal decisions retain separate accountable actors and authorities.

  1. What exact review outcome has been issued and to whom? decision
  2. Which downstream actions are explicitly not authorized by that outcome? constraint
  3. How may a separate decision cite this review under its own purpose and authority? relationship

Recognition

Model-owned context for recognition, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Contribution acknowledgement and reward recommendation

Case-local acknowledgement and reward recommendation are separate from grading. Independently issued awards, public recognition and monetary rewards remain externally mastered and are referenced. Recognition here is a business concept distinct from the recognition/observation facet. Neither a note nor recommendation changes an assessment or payroll.

  1. What contribution does this recognition acknowledge? event
  2. Does recognition refer to an objective, work product or separate contribution? relationship
  3. How is a recognition note prevented from changing the period grade or salary? constraint
Contest, correction and disclosure Scope-governed contest, correction and disclosure for one performance case.

Contest Correction

Model-owned context for contest correction, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Fact correction, opinion contest and process appeal

Distinguish fact correction, attribution correction, opinion disagreement, process contest and proxy exclusion. A disagreement alone does not prove a factual error; a supplementary statement must remain attached and visible according to policy.

  1. Is the subject contesting a fact, attribution, opinion, process or inadmissible proxy? classification
  2. Which independent route and decision-maker handles this contest? process
  3. What supplementary statement and new evidence accompany the appeal? evidence

Correction, supersession and recipient notice

An accepted correction appends a successor assertion, records effective and knowledge time, and evaluates who must receive correction notice. Notice is a policy hook; no universal legal deadline is asserted.

  1. Which authorized correction supersedes the contested assessment? lifecycle
  2. How are original knowledge time and corrected valid time preserved? temporal
  3. Which previous recipients require a purpose-limited correction notification? access

Disclosure

Model-owned context for disclosure, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Disclosure and employer change

Each Projection describes one canonical object. A multi-case packet is a context pack of separately authorized projections. A new employer receives no former review master by default; any limited excerpt needs explicit purpose, recipient, expiry and authority. Pin the source revision. Disclosing an assessment must preserve applicable dispute qualifications and authorized supplementary statements; exception handling is deferred and the reference validator rejects omission of applicable dispute context. Do not give a one-sided impression by selectively removing an active contest.

  1. Which fields of this one case may be disclosed for the requested purpose? privacy
  2. What authorizes the receiving employer to use the excerpt and for how long? authority
  3. Which source organization retains mastership after the subject changes employer? ownership
Governance and interoperability Scope-governed governance and interoperability for one performance case.

Profile Retention

Model-owned context for profile retention, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Retention, local profiles and exception policy

A startup may use narrative review and omit calibration. An enterprise may require contextual scales, recusal and independent appeals. Retention, legal holds and erasure are declared local policies; history preservation never grants indefinite retention.

  1. What policy governs retention, restriction, erasure and legal hold for this case? retention
  2. Which startup or jurisdictional profile relaxes optional ceremony while preserving identity and authority? exception
  3. Which minimized audit metadata may remain after content is lawfully removed? security

Validation Crosswalk

Model-owned context for validation crosswalk, referencing neighboring masters. Design choices are proposals grounded in source precedents, not asserted standard clauses.

Interoperability and bounded validation

Crosswalk HR Open concepts only to their verified scope. SFIA/ESCO skill identity and ISO organizational reporting remain neighbors. The companion schema and synthetic fixtures prove selected reference rules, not production IAM, HRIS interoperability or legal compliance.

  1. Which verified source field or concept maps to this model and with what loss? interoperability
  2. Which positive and negative fixtures demonstrate the declared reference-profile rules? validation
  3. Which legal, licensed-source, regional or operational claims remain explicitly unverified? quality

Classifiers Filled

Family
World Models
Category
Society, people and institutions
Entry kind
aggregate
Navigation path
NAV.SOC.ORG.PERF
Domain
SOC.ORG.PERF
Industry
Cross-industry
Tags
performanceobjectivereviewsoc.org.perf

What it is Filled

A performance case is an independently identified, purpose-qualified aggregate for one subject and one governing work context. It owns review-specific objective agreements, evidence assessments and attribution claims, reviewer bindings, period assessments, calibration history, contest/correction and disclosure records. Person, organization, employment, membership, assignment, contracts, original evidence artifacts and downstream employment/pay decisions remain external masters. Internal record kinds retain stable IDs and separate lifecycles; grouping them is not identity equivalence. Startup narrative-only and contractor profiles are permitted. Recognition and scales are distinct record kinds with a future separation hold.

In scope

  • Case identity, governing work context, purpose and period.
  • Objectives, evidence quality, attribution and conflicting claims.
  • Optional contextual scales, human assessment and calibration.
  • Typed appeals, corrections, retention hooks and one-object projections.

Out of scope

  • Person, employment, position, membership or contract master data.
  • Automatic hiring, promotion, dismissal or payroll execution.
  • Intrinsic or portable lifelong person scores and single-proxy productivity scoring.
  • Certified standard/legal compliance or a production HRIS implementation.

Why it exists Filled

Represent a governed performance case with scoped objectives, evidence, assessment, correction and purpose-limited disclosure.

Distinguishing features Filled

  • A scoped review case for one subject in one work context, not a lifelong score for a person.
  • Separates objectives, evidence, assessment, calibration and contest as distinct records.
  • Differs from employment and pay records: decisions based on it live elsewhere.
  • Supports non-employee contexts such as contractors and founders without forcing a rating scale.

What robots and AI may and may not do Filled

Must not

  • Issue a rating or outcome without an accountable human author.
  • Use activity or surveillance data as the sole basis for assessment.
  • Treat missing evidence as zero contribution.
  • Make or trigger hiring, promotion, dismissal or pay decisions.
  • Disclose reviews beyond the stated purpose.

Only with a human decision

  • Issuing review outcomes.
  • Calibration decisions.
  • Deciding appeals.

May

  • Help draft objectives for agreement.
  • Collect and cite evidence with its quality and attribution.
  • Summarize evidence for a human reviewer.
  • Record appeals and corrections.

Moral aspects Filled

  • Performance reviews affect livelihoods and must be fair, explained and contestable.
  • Bias in evaluations can follow gender, ethnicity, disability or caring responsibilities.
  • Monitoring-based metrics intrude on privacy and may misrepresent contribution.

Who is affected

  • Reviewed workers
  • Managers and reviewers
  • Teams whose shared work is attributed

Owners Filled

Steward

Declared Dimension purpose and accountable owner.

Roles

Subject
Provide evidence and supplementary statements; read permitted own-case information.
Review owner
Declare governing purpose, work scope, master system and applicable instrument.
Human reviewer
Assess scoped evidence and remain accountable for issued outcome.
Calibrator
Record justified before/after acts with role-limited access.
Appeal authority
Distinguish factual, attribution, opinion and process disputes; record independent disposition.
Data steward
Maintain provenance, corrections, policy-qualified access and retention.
Agent
Draft questions and evidence packets within delegated read/write scope; never execute personnel decisions.

Links to other meta-models Filled

references

  • vr.wm-per-001 - Person identity is externally mastered; reference minimal authorized identity only.
  • vr.wm-org-001 - Organization identity and reviewing authority owner.
  • vr.wm-org-005 - Employment applies conditionally to the employee profile; contractors do not create fictitious employments.
  • vr.wm-org-006 - Membership, when relevant, keeps its own relationship identity.
  • vr.wm-org-016 - Assignments define work binding; this case evaluates scoped work without owning assignments.
  • vr.wm-org-004 - Position and capability requirements are referenced, never copied into period scores.
  • vr.wm-rec-001 - Original evidence artifacts keep their document identity and retention master.
  • vr.wm-eco-006 - Contractor agreements remain outside case ownership.

neighbor

  • vr.wm-per-001 - Person identity is externally mastered; reference minimal authorized identity only.
  • vr.wm-org-001 - Organization identity and reviewing authority owner.
  • vr.wm-org-005 - Employment applies conditionally to the employee profile; contractors do not create fictitious employments.
  • vr.wm-org-006 - Membership, when relevant, keeps its own relationship identity.
  • vr.wm-org-016 - Assignments define work binding; this case evaluates scoped work without owning assignments.
  • vr.wm-org-004 - Position and capability requirements are referenced, never copied into period scores.
  • vr.wm-rec-001 - Original evidence artifacts keep their document identity and retention master.
  • vr.wm-eco-006 - Contractor agreements remain outside case ownership.

parent

  • WM-ORG-016

What else AI and robots need to interact with it Filled

Identity and identifiers required Filled

  • Authoritative master-system identifier with issuer and namespace.
  • Stable registry-qualified local identifier when no upstream master exists.
  • Immutable revision identity and source digest as integrity evidence, not the sole person identifier.

Direct properties not applicable Not applicable

Not applicable

Institutional or informational subject: no invented physical properties.

Recognition optional Filled

  • A review case has a subject, work context, period, objectives and an issuing authority.
  • Confused with an employment record, a 360 feedback survey, productivity dashboards and a disciplinary case.

Capabilities and actions required Filled

  • Open a scoped review case: Proposed governed operation; this specification is not a deployed HRIS action executor.
  • Agree or revise expectations: Proposed governed operation; this specification is not a deployed HRIS action executor.
  • Cite and assess evidence: Proposed governed operation; this specification is not a deployed HRIS action executor.
  • Prepare an assessment: Proposed governed operation; this specification is not a deployed HRIS action executor.
  • Issue a review outcome: Proposed governed operation; this specification is not a deployed HRIS action executor.
  • Record a calibration act: Proposed governed operation; this specification is not a deployed HRIS action executor.
  • Resolve or qualify a contest: Proposed governed operation; this specification is not a deployed HRIS action executor.
  • Produce a permitted one-case projection: Proposed governed operation; this specification is not a deployed HRIS action executor.

Hazards and failure modes required Filled

  • Unfair outcomes from biased or proxy evidence.
  • Leak of reviews to unauthorized colleagues.
  • Automated decisions without human review.

Standards and interfaces required Filled

  • HR Open Standards.
  • ISO 30414 human capital reporting.
  • 1EdTech CASE for competency frameworks.
  • ISO 8601 date and time for review periods.

Context of use required Filled

  • OPM is a US public-sector precedent, not universal practice.
  • Legal requirements are jurisdiction- and purpose-specific; unread sources remain unverified.

Sources Filled

  1. Employee Performance Management specification overview - HR Open Standards Consortium
  2. The Organization Ontology - W3C
  3. PROV-O: The PROV Ontology - W3C
  4. Competency assessment LEDA - SFIA Foundation
  5. The SPACE of Developer Productivity - Microsoft Research; ACM Queue authors
  6. Performance Management Cycle - US Office of Personnel Management
  7. Whole-object description profile - Vercy
  8. Human capital reporting and disclosure - ISO

Open questions

  • Specialist labor/privacy review and jurisdiction-specific retention, appeal and employee-representation profiles remain open.
  • Full licensed HR Open EPM XSD and ISO 30414 text/field mappings are not verified.
  • Production IAM, evidence adapters, retention enforcement and native Dimension instance bindings are not implemented by the companion example validator.
  • Public-service, collective-bargaining, platform-worker and organizational group-assessment profiles require separate research.
  • Shared rating-scale, calibration-session and recognition models may need promotion to separate canonical packages after cross-contour research.
  • Leave/accommodation fairness, multilingual profiles and appeal escalation policies are explicitly designed hooks, not fully validated operating profiles.
  • ESCO alignment was considered but deferred; no pinned ESCO schema or concept set is asserted.
  • Reference profile disclosure exceptions, full unlinking on erasure and pre-issuance calibration need separate operating profiles.
  • Claude no-tools audit read a frozen semantic text summary. Full JSON/code and fixture results were validated locally by Codex, not attested by Claude.
  • Specialist labor/privacy review and jurisdiction-specific retention, appeal and employee-representation profiles remain open.
  • Full licensed HR Open EPM XSD and ISO 30414 text/field mappings are not verified.
  • Production IAM, evidence adapters, retention enforcement and native Dimension instance bindings are not implemented by the companion example validator.
  • Public-service, collective-bargaining, platform-worker and organizational group-assessment profiles require separate research.
  • Shared rating-scale, calibration-session and recognition models may need promotion to separate canonical packages after cross-contour research.
  • Leave/accommodation fairness, multilingual profiles and appeal escalation policies are explicitly designed hooks, not fully validated operating profiles.
  • ESCO alignment was considered but deferred; no pinned ESCO schema or concept set is asserted.
  • Reference profile disclosure exceptions, full unlinking on erasure and pre-issuance calibration need separate operating profiles.
  • Claude no-tools audit read a frozen semantic text summary. Full JSON/code and fixture results were validated locally by Codex, not attested by Claude.

Machine files

Provenance

world-models research · reviewable-draft

Built from: models/wm-org-017-performance-objective-review/spec.yaml, ver-cy/world-models/card-supplements/wm-org-017-performance-objective-review.json