Authorial Source Discrimination

Preserve the author's thought through AI revisions

Distinguish authorial preservation, necessary adaptation and normalization introduced by the model.

When a revision sounds polished but no longer expresses the author's thought, or an assistant must write from actual examples of someone's work.

Editorial competence · Writing and authorship · Public edition 0.1.0

The AI revision turns an author’s question into a conclusion

The original asks what kind of participation a system could make possible. A revision asserts that the system will deliver a predetermined outcome.

The key relationship. Recognizes what carries an author's meaning and preserves it while adapting only what the selected artifact and receiving situation require.

An illustrative situation to adapt to your work. Read the complete cases ↓

Adapt the competence to your work

You supply the context: research, a service, a creative project or a team’s work. Examples make the method visible; adaptation preserves the relationships that make it work while understanding the purpose, sources, people and means of your own assignment.

Tell your AI what you want to achieve, what already exists and a practical situation to address. Ask how this competence changes the work and which knowledge from your domain should participate. MAIOS provenance identifies the shared knowledge; you can use it directly in your AI or compose it with an existing kernel.

Adaptation examples in other contexts ↗

Preserve the author's thought through AI revisions

Understanding a concept does not by itself preserve how its author expresses it. Order, emphasis, rhythm, tension and degree of determination can carry meaning.

This competence makes that discrimination operative while the artifact forms. The author's original, a selected template and an AI proposal retain their actual attribution.

Recover the entrusted artifact and original

Understand the current purpose, object, audience, medium and selected transformation. Preserve the original before proposing a revision. If another revision is still being considered, keep your proposal attributed and separate rather than silently selecting it as the new original. A narrow correction can already provide sufficient direction. Ask for missing source meaning only when it materially changes fidelity; do independent work that the unresolved relation does not affect.

Qualify actual evidence of expression

Use the supplied draft, template or exemplar first and preserve its scope. When discovering a person's writing is selected, use the host's actual retrieval means, inspect the bodies and establish who wrote each relevant passage. A user-role message, repository owner, commit author, title or accepted draft does not prove authorship. Separate quoted, forwarded, collaborative and AI text from the person's original segments. A template can guide form without proving that the person authored it. Do not expand beyond a designated exemplar unless that scope is selected.

Recognize what the source makes the reader encounter

Understand subject, object, causal direction, conceptual relations, emphasis, degree of openness, abstraction, meaningful lexical choices, rhythm and reader movement where they matter. Observe features supported by the examples rather than assigning generic qualities. Preserve a clear source determination as well as an intentional open relation. If the subject has been compressed away, recover it before inventing a category, objection or explanatory frame to make the rewritten sentence coherent.

Transform through the actual receiving need

Distinguish preserving an authorial relation, adapting its expression to real surface conditions and model normalization. Change wording, density or order when that makes the same relation recoverable. A familiar explanatory, commercial or institutional pattern is not sufficient reason. Keep the source and proposed transformation comparable where a consequential choice remains. Explain the material difference outside the deliverable when useful, without filling the artifact with editorial machinery. Exact copy or a selected mechanical placement proceeds without an unrequested rewrite.

Learn the causal difference without freezing a profile

When a correction teaches a reusable relation, retain the smallest source/candidate/correction comparison and what was added, removed or reordered. Preserve the reason and its scope, rather than the preferred sentence alone. Update the writing method that should perceive the nuance next time; reconnect its entry if the knowledge was already present but absent from the work. An author's register can change with task and time. Do not turn frequency, punctuation or past preference into permanent mimicry instructions.

Authorial Source Discrimination in practice: worked cases

These cases illustrate situations, causal distinctions and choices you can adapt to your work.

The following cases are illustrative. They explain choices and conditions; they are not reports of observed adoption.

A question becomes a conclusion

The original asks what kind of participation a system could make possible. A revision asserts that the system will deliver a predetermined outcome. The assistant recognizes a changed degree of determination, restores the open question and keeps the stronger claim as a separately attributed proposal only if it is useful to discuss. Polished wording does not establish authorial fidelity.

The channel is shorter than the source

An author provides a short argument for a platform with a strict title field. The assistant preserves the argument and makes the actual subject recognizable across title and body. It changes placement only where the channel requires it, instead of replacing the source with a generic slogan. The exact proposed version can then be reviewed for that destination; adaptation is not publication.

An accepted document contains mixed authorship

A shared document includes original paragraphs, a quoted collaborator and an AI summary accepted by the person. The assistant qualifies the actual passages before using them as style evidence. Acceptance can establish a preference without making the AI summary an original human sample. The new draft follows supported expressive choices and current intent rather than claiming an established personal voice from the whole document.

Operating choices with Authorial Source Discrimination

Recognizes what carries an author's meaning and preserves it while adapting only what the selected artifact and receiving situation require.

Fidelity and source retrieval

Current owner knowledge supplies facts and commitments; authored examples supply expression. Do not import old prices, people, promises or private context from a style sample. Use permitted sources through their actual access mechanism, retaining enough provenance to revisit the passage. Do not build a private writing cache or general corpus by inertia. If evidence cannot support a personal-style claim, work from the selected draft or ask for a pertinent example, rather than claiming a recovered voice.

Three operations with different reasons

Authorial preservation retains the relation carried by a phrase, tension or rhythm. Surface adaptation changes its expression because the real reader or channel needs that change. Model normalization replaces it with a familiar pattern without such a reason. These are situated distinctions: a section, negation, short sentence or benefit statement can be right when the actual source and artifact make it useful.

A practical comparison

Ask what the fluent new sentence makes perceptible that the source did not, and what it removes. If it could describe many unrelated projects after changing two nouns, examine whether the source identity was lost; this is a probe rather than a rejection rule. Keep necessary generic functional copy. Later corrections refine discrimination without requiring a comparison record for every edit.

Read the complete operating method

Preserve the author's thought through AI revisions

Understanding a concept does not by itself preserve how its author expresses it. Order, emphasis, rhythm, tension and degree of determination can carry meaning.

This competence makes that discrimination operative while the artifact forms. The author's original, a selected template and an AI proposal retain their actual attribution.

Recover the entrusted artifact and original

Understand the current purpose, object, audience, medium and selected transformation. Preserve the original before proposing a revision. If another revision is still being considered, keep your proposal attributed and separate rather than silently selecting it as the new original. A narrow correction can already provide sufficient direction. Ask for missing source meaning only when it materially changes fidelity; do independent work that the unresolved relation does not affect.

Qualify actual evidence of expression

Use the supplied draft, template or exemplar first and preserve its scope. When discovering a person's writing is selected, use the host's actual retrieval means, inspect the bodies and establish who wrote each relevant passage. A user-role message, repository owner, commit author, title or accepted draft does not prove authorship. Separate quoted, forwarded, collaborative and AI text from the person's original segments. A template can guide form without proving that the person authored it. Do not expand beyond a designated exemplar unless that scope is selected.

Recognize what the source makes the reader encounter

Understand subject, object, causal direction, conceptual relations, emphasis, degree of openness, abstraction, meaningful lexical choices, rhythm and reader movement where they matter. Observe features supported by the examples rather than assigning generic qualities. Preserve a clear source determination as well as an intentional open relation. If the subject has been compressed away, recover it before inventing a category, objection or explanatory frame to make the rewritten sentence coherent.

Transform through the actual receiving need

Distinguish preserving an authorial relation, adapting its expression to real surface conditions and model normalization. Change wording, density or order when that makes the same relation recoverable. A familiar explanatory, commercial or institutional pattern is not sufficient reason. Keep the source and proposed transformation comparable where a consequential choice remains. Explain the material difference outside the deliverable when useful, without filling the artifact with editorial machinery. Exact copy or a selected mechanical placement proceeds without an unrequested rewrite.

Learn the causal difference without freezing a profile

When a correction teaches a reusable relation, retain the smallest source/candidate/correction comparison and what was added, removed or reordered. Preserve the reason and its scope, rather than the preferred sentence alone. Update the writing method that should perceive the nuance next time; reconnect its entry if the knowledge was already present but absent from the work. An author's register can change with task and time. Do not turn frequency, punctuation or past preference into permanent mimicry instructions.

Use in the receiving environment

Use the files, observations and tools actually available for the entrusted work. A chat can reason from supplied material and return a complete note to save. A host with files can keep the competence and project guide reachable. Work through an existing kernel when present; the standalone starter also supplies the method. Ask for a missing relation when it changes a decision, while continuing useful independent work. Instructions do not create accounts, persistent execution, tools or authority over another person’s systems.

Learning and continuation

Preserve the source, reason, relevant conditions, useful result and unresolved relation in the guide or competence that should act differently next time. An occurrence, the understanding available then and present understanding retain their attribution. When knowledge already existed but was not reached, repair its entry instead of duplicating the method. Change the method when a reusable consequence warrants it; otherwise continue without a maintenance ritual. A written correction makes knowledge available. Its participation in a later different case can show learning; the text does not change model weights.

Entrusted work and material effects

Carry out actions already covered by the person’s mandate and the actual environment. A proposed message, preparation, dispatch, receipt, integrated change and observed use are distinct facts. Publication, spending, account changes, sensitive disclosure, installation and changes to another owner follow their specific authority. An unavailable or unselected effect leaves independent understanding and preparation possible.

Operating knowledge

Read OPERATING_GUIDE.md and EXAMPLES.md as part of this method. They carry decision knowledge and worked cases. START_CHAT_EN.txt includes all three bodies for a chat without access to these files.

Authorial Source Discrimination and improvement of the method

Revision can teach the assistant to perceive its own substitutions

The direct work preserves authorial meaning through an artifact. A meta relation appears when a correction changes the continuing writing competence.

The relation becomes autological when the assistant recognizes its own preferred explanatory grammar as the source of the displacement. The correction changes how the next draft begins rather than merely repairing one sentence.

Competence, meta and autology: the conceptual guide ↗

How Authorial Source Discrimination learns through work

Keep the source-specific relation and why its loss changed meaning. An isolated preference can remain with the artifact.

A later different revision can reveal whether the discrimination participates or overfits an earlier example.

Use Authorial Source Discrimination in chat or a project

In chat

Copy or attach START_CHAT_EN.txt and add your intent, relevant sources and available means. Save the useful continuation for the next session.

In an agentic environment

Make the complete folder reachable, start from SKILL.md and keep operating knowledge and examples beside it. The host determines actual tools, memory and actions.

With an existing kernel

Connect the method to the competence and project guides that already own the work. Preserve useful local knowledge when considering an update.

Give Authorial Source Discrimination an assignment

Read or copy the complete English starter and add your context. English is the kit’s primary language; Italian counterparts are included.