Builders

This map is for readers who build software, products, organizations, institutions, policy standards, interfaces, machines, or methods and want to see where inepistemic makes an epistemic distinction relevant to building. It is not a reading order. Start from a concept already familiar in your work, then follow the linked chapters.

The book's recurring builder question is:

What happens when the thing you build becomes part of somebody else's epistemic environment?

If you want a short path

For a first pass through the builder-facing argument, the most concentrated sequence is:

06 - Abstractions → 09 - Problems → 10 - Coordination → 17 - Arrows → 21 - Error → 22 - Interfaces → 26 - Invention → 29 - Authorities → 30 - Inepistemic

That path moves from abstraction and problem formation, through coordination, transitions, error correction, and interfaces, into invention, authority, and the book's own architecture as a build.

For the fuller argument and the relations among these distinctions, use the longer Reader's Guide: For Builders.

Users, models, and representations

Familiar builder concept	Distinction sharpened in the book	Chapters
User model / persona / segment	Representation / causal explanation. A useful model of a person can support prediction or coordination without containing the singular causal history of the person represented.	01 - Genetics · 02 - Populations · 03 - Societies
Role / permission / classification	Administrative interface / ontology of the person. A category may be necessary for the system to operate without thereby explaining the person to whom it is applied.	03 - Societies · 15 - Personhood · 22 - Interfaces
Personalization	Adaptation / open future. Adapting around past behavior is different from possessing the future problem space of a person who can create new knowledge.	05 - Openness · 18 - Unpredictability
Requirements	Explicit statement / inexplicit knowledge. What someone can state is not necessarily the full knowledge operating in recognition, rejection, skill, expectation, or action. Explicitation can itself create a new representation.	13 - Inexplicit
Capability assessment	Knowledge presently possessed / capacity for knowledge creation. A current competence judgment need not become a completed account of what the person can later learn or explain.	15 - Personhood · 18 - Unpredictability

Architecture, abstraction, and interfaces

Familiar builder concept	Distinction sharpened in the book	Chapters
Abstraction	Useful compression / explanatory loss. An abstraction works by preserving selected structure and hiding the rest. Its success does not make the hidden structure cease to exist.	06 - Abstractions · 22 - Interfaces
Interface	Abstraction assistance / abstraction authority. An interface can make complexity navigable, or it can make one compressed representation feel like the only reality available.	06 - Abstractions · 22 - Interfaces · 24 - AI
Encapsulation	Hidden detail / criticizable dependence. Nobody can reconstruct every underlying layer, but the hidden layer may still need to become a problem when the abstraction fails.	22 - Interfaces · 23 - Understanding
System state	State / explanatory transition. Recording before and after states does not explain why one state became another or which criticism made the transition relevant.	17 - Arrows · 28 - Motion
Extensibility	New values / new dimensions. A system may accept more cases while keeping the governing problem fixed. The stronger form of openness is surviving criticism that changes what the system was for.	18 - Unpredictability · 28 - Motion
Understanding a system	Useful abstraction / exhaustive reconstruction. Understanding need not require possession of every implementation detail if the relevant explanatory structure can be criticized and deeper layers can be reopened when necessary.	23 - Understanding

Problems, optimization, and change

Familiar builder concept	Distinction sharpened in the book	Chapters
Problem statement	Represented problem / epistemic problem. A problem is itself knowledge: it can be badly formed, reconstructed, abandoned, or replaced.	09 - Problems
Optimization	Solution improvement / problem transformation. A system can become extremely effective inside a represented problem while leaving the governing abstraction untouched.	08 - Criticism · 09 - Problems · 23 - Understanding
Critique / review	Rejection / criticism that creates the next problem. Criticism can do more than eliminate a candidate; it can expose a hidden assumption, change a criterion, or create a new problem.	08 - Criticism · 09 - Problems
Prototype	Early solution / instrument for creating the next problem. Externalizing an artifact can reveal failures or possibilities unavailable before the artifact existed.	26 - Invention · 28 - Motion
Change history	Version history / problem genealogy. Preserving states or outputs is different from preserving the criticisms, abandoned alternatives, and changed abstractions that make later states intelligible.	17 - Arrows · 28 - Motion

Coordination, organizations, and authority

Familiar builder concept	Distinction sharpened in the book	Chapters
Coordination	Shared interface / shared mind. People can coordinate successfully without converging on one internal model or becoming a collective epistemic subject.	04 - Solitude · 10 - Coordination
Standardization	Thin coordination / explanatory convergence. A coordination layer can specify what must be shared without requiring the participants' internal problem worlds to become alike.	10 - Coordination
Legibility	Administrative visibility / epistemic adequacy. A representation may make a system easier to manage without becoming a sufficient explanation of the people or processes represented.	03 - Societies · 22 - Interfaces
Governance	Decision structure / epistemic sovereignty. A governing mechanism may be necessary without becoming permanently authoritative over future problems and criticism.	21 - Error · 29 - Authorities
Authority	Status or expertise / tentative relation solving a problem. Authority can be treated as a revisable relation whose governing explanation must itself remain criticizable.	29 - Authorities
Default / metric / workflow	Governing mechanism / removable ruler. People are not the only things that can rule. A metric, model, standard, default, or legacy assumption can also acquire authority when alternatives become practically unreachable.	22 - Interfaces · 29 - Authorities

Reliability, safety, and correction

Familiar builder concept	Distinction sharpened in the book	Chapters
Reliability	Preventing failure / preserving an error-correction surface. Error correction can depend on people, criticism, records, alternatives, institutions, interfaces, reversibility, and time rather than on any component being infallible.	21 - Error
Safety control	Constraint / corrigibility of the constraint. A control can reduce known hazards while still needing mechanisms through which errors in the control itself can become visible.	21 - Error · 29 - Authorities
Access control	Epistemic access / execution authority. Broad access to knowledge, modeling, criticism, and alternatives does not require equally broad authority to perform irreversible actions.	20 - Resources · 21 - Error
Audit trail	Record / explanatory history. A complete record of actions can still omit which criticism, assumption, or changed problem made those actions intelligible.	12 - Justification · 17 - Arrows · 28 - Motion
Resilience	Return to prior state / preservation of epistemic movement. Earlier failed conjectures and abandoned criteria can remain causally useful; recovery may require more than restoring the latest known-good state.	21 - Error · 28 - Motion

Values, resources, and invention

Familiar builder concept	Distinction sharpened in the book	Chapters
Goal / value / policy principle	Compressed explanation / irreversible compression. A value can carry prior reasoning compactly, but becomes epistemically dangerous when the explanation beneath it can no longer be reopened.	19 - Values
Resource allocation	Capacity / epistemic reach. Resources change which problems can practically be pursued without determining which problem should become significant.	20 - Resources
Opportunity cost	Lost benefit / lost epistemic trajectory. Restricting a resource can remove not only a known outcome but experiments, criticisms, skills, and successor problems that never become available.	20 - Resources
Recombination	Inventory / invention. Having all the components, precedents, dependencies, and enabling technologies does not by itself explain the new relation or abstraction under which they became a solution.	18 - Unpredictability · 26 - Invention
Contribution attribution	Global weight / problem-indexed causal work. Different people, tools, standards, resources, and abstractions can be causally indispensable without being reducible to one percentage scale. Ask what changed because this knowledge was present.	27 - Weights
AI-assisted building	Faster externalization / abstraction authority. AI can make candidate structures objective earlier, moving criticism forward, while also influencing the framing before the problem has stabilized.	24 - AI · 26 - Invention

The build itself

Familiar builder concept	Distinction sharpened in the book	Chapters
Implementation	Executable commitment / proof. Making an epistemic commitment explicit in architecture can expose ambiguity and failure without proving the epistemology embodied in the implementation.	30 - Inepistemic
Iteration	Artifact revision / recursive inquiry. The deeper recursive object is not merely the artifact but the changing problem, criticism, and knower moving through successive builds.	26 - Invention · 28 - Motion · 30 - Inepistemic
Successful architecture	Usability / finality. The recurring builder problem is how an abstraction can remain useful and stable without becoming epistemically sovereign over the next problem.	22 - Interfaces · 29 - Authorities · 30 - Inepistemic