constraining

Type: types/tag-readme.md

Constraining narrows the space of valid interpretations an artifact admits, from conventions to deterministic code; relaxing deliberately widens that space again. Assign this tag when an artifact explains, applies, tests, or decides when to use or reverse that narrowing. Merely choosing an output or improving task accuracy is insufficient unless the interpretation space is at issue. LLM reliability covers output deviations and their correction more broadly; constraining is one possible response. A child of learning-theory.

Definition and spectrum

• constraining — definition and spectrum: definitions, conventions, structured sections, schemas, validators, and deterministic code
• program refinement — the checked narrowing relation behind constraining; an executor's output is a candidate until checked
• codification — the move from natural language to a symbolic artifact whose consumers are held to a reference that fixes what it permits
• prompting changes failure diagnosis with the semantic reference — the foundation: two accounts of prompting, under which an instruction permits several behaviors and the executor produces one

Instances and techniques

• Selecting an LLM output fixes a result, not its interpretation — boundary case: selection fixes result identity; it counts as constraining only when adoption also narrows which interpretations remain operative
• Constraining during deployment can produce learning outside weights — versioned constraining beats weight updates on inspectability and rollback
• spec mining as codification — observe behavior, extract deterministic rules, grow the calculator surface monotonically
• error messages that teach are a constraining technique — in agent systems the error channel is an instruction channel
• methodology enforcement is constraining — review gates and validation as constraining applied to the KB's own methodology

Deciding and reversing

• codify-versus-LLM decision heuristics — four lenses on the codify-vs-LLM decision, with evidence they come apart at the edges
• specification strategy should follow where understanding lives — when to commit: before execution, during execution, or after repeated observation
• progressive constraining commits only after patterns stabilize — the default discipline: codify after the pattern proves the need
• unified calling conventions enable bidirectional refactoring — how to commit reversibly: neural and symbolic components behind the same callable interface
• codification and relaxing navigate the bitter lesson boundary — why reversibility matters: codification is a bet that may need relaxing
• operational signals that a component is a relaxing candidate — five testable signals for detecting when to reverse codification
• open-ended knowledge work shares one epistemic layer across roles — why conversion needs the derivable/decided split recorded: only a decided resolution is a bet, and only a recorded bet can name its relaxing trigger in advance

Related Tags

• deploy-time-learning — the post-release change constraining absorbs; the verifiability gradient locates constrained artifacts

Other tagged notes

• A methodology governs its own extension only as far as it settles the meta-decisions it raises - A retained methodology governs the consequential extension decisions it supplies or imports; actor competence can carry the process further without making those choices settled by the method
• A theory's prototype standing is its revision cost: external binding plus lost investment - A theory's prototype standing is its expected revision cost — external binding plus the investment a revision discards — so natural-language versus symbolic form determines neither component and acceptance status is a separate axis
• Ad hoc prompts extend the system without schema changes - Any system with an LLM agent layer can absorb new requirements through natural language prompts without changing the deterministic base
• Adaptation signals choose pressure; artifact analysis chooses the retained surface - Maps agentic-adaptation signals onto artifact-analysis axes so KB learning records which retained surface changes, what authority it gains, and how to review it
• Alexander's patterns connect to knowledge system design at multiple levels - Maps Alexander's Context/Problem/Forces/Solution pattern to typed document contracts and his generative process to incremental codification, while marking the looser 'centers' analogy
• Bidirectional codification as a comparative test - Proposal: promote bidirectional codification from design guidance to a comparative conjecture, tested by evolving the same task stream under natural-language-only, symbolic-only, one-way promotion, and bidirectional codify-and-relax regimes
• Changing requirements conflate genuine change with disambiguation failure - Separates world change from late discovery that downstream work chose the wrong interpretation of an underspecified requirement; short iterations mainly limit propagation of the latter
• Code complements the weight–prompt pair with independently executed symbolic operations - Distinguishes model interpretation from runtime-assigned execution, including model-generated code and the same code artifact used as prompt context or as an executable operation
• Commitment, not derivation, creates new ground truth - Derivation — claims recoverable from the source, nothing added — leaves the source as ground truth; what adds unentailed resolutions becomes ground truth at commit, repaired by supersession
• Constraining and extraction can trade generality for reliability, speed, or cost - Constraining narrows interpretation and extraction produces focused use-shaped artifacts; both can trade generality for reliability, speed, or cost when task fit is good
• Enforcement without structured recovery is incomplete - The enforcement gradient covers detection and blocking but has no recovery column — recovery strategies (corrective → fallback → escalation) are the missing layer, and oracle strength determines which are viable at each level
• Exact implementation does not validate a requirement against its objective - An artifact can exactly implement a requirement while the requirement remains a conjectured proxy for a declared objective; assess each named path separately, and attribute failure to the link without erasing local correctness
• Improvements outside the admitted formal language need a pre-formal stage somewhere - An improvement whose concepts have no expression in a loop's admitted formal language is reached only through a pre-formal stage, inside the loop or fixed at design time in the choice of language; translation relocates that stage
• Inspectable artifact, not supervision, defeats the blackbox problem - Readable scripts and validators let humans or agents inspect and revise individual rules in generated systems; neither symbolic membership nor human supervision alone supplies that inspection route
• Legal drafting solves the same problem as context engineering - Legal drafting parallels context engineering because both write ambiguous natural-language specifications for judgment-based interpreters, but law develops constraining more than codification
• LLM-executed methodologies are metacircular interpreters, not compilers - Self-hosting LLM methodologies are closer to metacircular interpreters than compilers: agents re-interpret natural-language rules each session, while stable paths codify into validators and commands
• Methodology with incomplete coverage and its live theory fallback form a two-layer execution system - In open or incompletely covered domains, the theory-derived fast path and live theory fallback co-execute while methodology-native content follows a separate maintenance regime
• Mocking the tool boundary makes instruction behavior testable in isolation - Controlled tool responses and captured effects let instruction tests expose protocol violations that a plausible final artifact can hide; coverage depends on mock fidelity.
• Model-resolved indirection adds interpretation work to LLM execution - A reference adds model-side interpretation only when the model must resolve it; upstream literalization is worthwhile when binding, token, authority, and regeneration costs favor it
• Moving the interpretation–enforcement boundary requires cross-form coverage - Moving responsibility between model-interpreted rules and formal enforcement crosses natural-language and symbolic forms, so governing the transfer requires coverage of both and their mapping
• Natural-language meaning is coordination among learned approximations, not approximation of a target - Untested hypothesis extending Lampinen's learned-approximation mechanism to natural language: no external semantics exists to approximate, so meaning is coordination held by weak checks; recasts constraining and deviation diagnosis
• Oracle strength spectrum - Exploratory framework — oracle strength, how cheaply correctness can be verified, as the gradient underlying the exact-spec/proxy-theory distinction, with an oracle-hardening pipeline
• Preferential codification concentrates less predictable work at the agent boundary - Explains the negative-selection mechanism by which preferential codification changes the composition of work retained at an agent boundary
• Process structure and output structure are independent levers - Distinguishes constraints on reasoning steps from constraints on result shape and identifies the evidence needed to separate their effects
• Programming practices for prompts need evidence matched to the question - Maps typing, testing, code extraction and versioning to the questions they answer; distinguishes requirement adequacy, conformance, predictive fidelity and execution variation.
• Reflective coverage is graded across representational forms - Reflective coverage is stated per represented form and operation profile; control of an external dependency does not make that dependency part of the system's reflective coverage
• Retire, Redact, Supersede, And Relax Memory - Memory systems need lifecycle operations for redaction, decay, supersession, retirement, relaxation, and temporal validity
• Technical constraints turn KB objective-function choice from philosophy into engineering - Three technical constraints and the codification lever make KB objective-function choice testable engineering, not philosophy; goals set the loss, local contracts specialize it, and oracle strength differs by objective
• The verifiability gradient - Compare verification by the requirement checked, reset conditions, retry cost and strength of feedback; prompts and code have no automatic ordering on these dimensions
• Traditional software can bracket executor conformance; LLM systems cannot - Wrongness is a relation to a norm, never intrinsic to a computation; classical stacks bracket the executor-conformance norm so every failure resolves to the spec, and LLM systems cannot, which is what generates the three-source deviation taxonomy
• Writing styles are strategies for managing underspecification - Maps descriptive, prescriptive, prohibitive, explanatory, and conditional context-file styles to distinct ways of narrowing agent interpretation, each trading constraint against generality