View source for Calibration Recommendations

{{Development Notice
|status = Active Development
}}
= Calibration Recommendations =

'''Recommend realignment of the instrument. Do not seize the definition. Do not play judge.'''

'''Calibration Recommendations''' is a '''documentation pattern''' (Seed / Self-Assessed) for packaging post-run recommendations after a calibration procedure — especially [[Abstract Definition Mechanics (Seed)]] and related map-calibration work.

The object of recommendation is the '''conceptual mechanism''' (and its documented **material** gaps), not the truth, morality, or historical validity of ideas, policies, institutions, or actors that later use that mechanism.

{{GameModule
| type = Documentation Pattern
| category = [[:Category:Calibration Procedures|Calibration Procedures]]
| Calibration Type = Metrological
| Application Layer = Framework / Reporting
| Version = 0.2
| Maturity = Seed
| Last Updated = 2026-07-27
| description = Pattern for gated post-run calibration recommendations: standing-honest headers, protocol vs IR operator rules vs redesign advice, material omission, severity levels, and explicit non-recommendations — without ownership or moral judgment.
}}

== Purpose ==

Make recommendations after a calibration run:

* '''inspectable''' — what is recommended, why, and at what severity
* '''bounded''' — instrument realignment only
* '''gated''' — redesign advice only when a material gap blocks stated purpose
* '''non-captured''' — no ownership theater, no moral theater
* '''reusable''' — comparable shape across reports
* '''standing-honest''' — presentation does not outrun Self-Assessed / experimental status

== When to use ==

Use this pattern when:

* a run is multi-pass or high-stakes
* multiple children, material omissions, or interconnect rules must be stated clearly
* readers need **IR operator rules** before deeper theory talk

Skip or shorten when a single-line recommendation in the report is enough.

== Metrological stance ==

Recommendations:

* point at map gaps, mechanism grades, boundaries, and realignment options
* leave adoption to definition authors, maintainers, or voluntary operators
* do '''not''' raise standing of the procedure or of the abstract under test
* do '''not''' determine moral worth, historical guilt, political legitimacy, or which ideology is “correct”
* claim '''procedural transparency and re-runnability''', not consequence-neutrality (live terms may have asymmetric downstream effects)

'''The recommendation object is the mechanism and its documentation — not the ideas that ride on the word.'''

== Three layers (do not mix) ==

{| class="wikitable"
! Layer
! Role
! Example
|-
| '''Protocol'''
| Required to '''run''' the calibration procedure (e.g. ADM)
| Grade mechanisms A/B/C; material omission check; case friction; no moral verdicts
|-
| '''IR operator rules'''
| Rules for '''using this compiled map''' in analysis or debate
| Child-lock on *this* object; forbid silent 4→1.2 *for this IR*
|-
| '''Recommendations'''
| Optional advice to '''improve the instrument text or structure'''
| Owner may add transition criteria; consider splitting senses
|}

Protocol ≠ universal laws of language.  
IR operator rules bind to a '''dated report'''.  
Recommendations never equal ownership.

== Recommendation gate ==

Issue a '''change recommendation''' (redesign / realignment advice) only if '''both''' hold:

1. A gap is '''material''' — demonstrably required for the instrument’s '''stated purpose''' to be checkable (one sentence: *Purpose P requires checkability of X; text is silent on X*), '''and'''
2. The fix is framed as owner-may-add text, split, bound, or exception — '''not''' a filled preferred theory

Otherwise:

* '''Document the gap only''' (or ordinary underspecification / open question)
* '''No redesign recommendation'''

High priority requires the materiality sentence. Ranking by political convenience fails inversion.

== Severity scale ==

When a recommendation passes the gate, tag severity:

{| class="wikitable"
! Level
! Meaning
|-
| '''L1'''
| Clarification of existing text
|-
| '''L2'''
| Boundary / scope note
|-
| '''L3'''
| Child split / rename
|-
| '''L4'''
| Structural revision
|-
| '''L5'''
| Calibration failure — cannot minimally realign without conceptual laundering
|}

== Evidence strength ==

* '''One friction case''' → local recommendation only  
* '''Broader wording''' → more cases or explicit underclaim  
* Do not let recommendation volume substitute for standing

== Required sections ==

Fill only sections that apply. Prefer underclaim.

=== 0. Header ===

* Object under test (name + dated text / version)
* Procedure used (e.g. ADM v0.3.2)
* Pattern version (this page, e.g. v0.2)
* Cycle depth (single pass / Cycle 1–3)
* Standing of the run (default: Self-Assessed)
* Explicit notes:
  * recommendations ≠ ownership
  * recommendations ≠ moral verdict
  * '''this instance is not a universal methodology'''
  * IR operator rules bind to this report, not to language in general

=== 1. IR operator rules (use this map) ===

Short rules for anyone using '''this compiled IR''' in debate or analysis — not eternal laws of speech.

Typical form (content is object-specific):

* Name the child / sense before arguing
* Forbid silent upgrades across senses or registers **for this map**
* Treat register-tagged senses as '''first-class jobs''', not lesser meanings; do not upgrade them into technical mechanism claims without a tag
* Tag **material** omission only with a purpose-checkability sentence
* Attach empirical evaluation to described institutions or programs under a locked child — not to the prestige noun alone

=== 2. Per-instrument or per-child sheets ===

For each major child or load-bearing claim cluster:

* Recommendations that **passed the gate**, with '''severity (L1–L5)''' and evidence note
* Gaps **documented only** (gate failed or underspecification)
* Realignment target (what “better map” would state — without inventing payload as already present)
* Explicit '''not recommended''' moves (laundering, prestige retention with payload swap, silent register upgrade)

=== 3. Structural recommendations ===

Cluster-level or interconnect rules (still gated where they demand text changes):

* co-application matrix highlights
* false-friend / collapse-move list
* shared-term scope locks (undefined key nouns)

=== 4. Meta / procedure recommendations ===

Only when the run teaches something about the '''procedure''' or reporting pattern — not about politics.

Examples:

* promote a recurring **material** gap type into procedure vocabulary
* multi-cycle depth for prestige terms
* stop rules when evidence classes are exhausted

=== 5. Explicit non-recommendations ===

State what this package does '''not''' authorize, including:

* rewriting the source definition by calibrator fiat
* declaring the domain “settled”
* moral or historical verdicts from map calibration alone
* filling omissions with preferred theory under the guise of clarification
* raising instrument grade or standing by recommendation volume
* consequence-neutrality or “no side is affected”
* treating IR operator rules as a ban on ordinary colloquial speech
* selling this instance as a general standard for all abstracts or all scholarship

=== 6. Priority order ===

If time is scarce: ordered list of what to apply first (IR rules before optional redesign advice).

=== 7. One-line summary ===

Single sentence capturing the discipline of the package.

== Optional sections ==

* Link to full Calibration Report / cycle artifacts
* Dependency list (TDP, Semantic Drift, ADM, child pages)
* Follow-on work: locked-child empirical pilot (separate procedure object)
* Recipient note: recommendations address maintainers and IR operators; adoption remains voluntary (including when “owner” is diffuse)

== Relationship to other instruments ==

* '''[[Abstract Definition Mechanics (Seed)]]''' — primary consumer; recommendation gate and material omission align with ADM §3 and §7
* '''Calibration Report''' — recommendations may be embedded or attached; this page is the '''shape''' standard
* '''[[Semantic Drift]]''' — may appear as a tagged condition; Drift is not itself a recommendation engine
* '''[[Term Decomposition Protocol]]''' — lower layer; term-level recommendations stay term-scoped unless escalated

== Anti-patterns ==

* Essay-length moral conclusions labeled “recommendations”
* “Therefore the ideology is false/true” as a calibration output
* Silent merge of colloquial register into technical senses
* Inventing omitted thresholds and presenting them as already in the instrument text
* Omission-maximalism (infinite wishlist without purpose-checkability sentences)
* Recommendation spam that substitutes for floors and underclaim
* Imperative tone that outruns Self-Assessed standing
* Conflating protocol, IR operator rules, and redesign recommendations

== Success criteria for this pattern ==

The pattern is working when recommendation packages are:

* '''Comparable''' across different abstracts
* '''Actionable''' for operators without seizing definitions
* '''Gated''' — redesign advice only when materiality + non-laundering hold
* '''Separated''' from judgment disciplines
* '''Honest''' about material omissions, documented-only gaps, non-recommendations, and standing

== Permanent Beta ==

This pattern is subject to the same metrological discipline as the procedures it supports. Improve by documented use, not by rhetorical expansion.

'''See the Game. Refuse the Game. Build Better.'''

[[Category:Calibration Procedures]]
[[Category:Meta & Framework]]

{{Resource
| Title = Calibration Recommendations
| URL = https://www.thesovereigngames.com/wiki/Calibration_Recommendations
| Description = Seed documentation pattern (v0.2) for gated post-run calibration recommendations: material omission, recommendation gate, protocol vs IR operator rules vs redesign advice, severity L1–L5, standing hygiene, and explicit non-recommendations — mechanism-focused, without ownership or moral judgment.
| Category = Calibration Procedures
}}

{{Admin Page Status
<!-- Do Not Remove commented out Reference Options -->
| categorization = Done
| calibration_review = Self-Assessment
| instrument_grade = Experimental
| validation = Low
| review_date = 2026-07-27
| reviewed_by =
| priority = Supporting
| review_threshold = 90
| has_backlinks = Not checked
| outbound_links_valid = Not checked
| in_outline = Not checked
| in_category_outline = Not checked
| templates_complete = Needs review
| formatting_standard = Meets standard
| symmetry_check = Not applicable
| self_report_flagged = No
| terminology_consistent = Needs work
| standing_check = Self-assessed only
| drift_report_status = None open
| depends_on = Abstract Definition Mechanics (Seed); Semantic Drift
| cited_by =
}}