A Main Street Independent reference — the treatise on values, journalism, and the discipline that connects them: why news cannot be made value-free, the consensus values floor the publication’s general newsfeed operates from, and how News, Analysis, and Advocacy differ. It is the foundational document of the publication’s editorial method, published under CC0.

A living document. Forks invited; revisions documented.

Front Matter

Mission Statement

Main Street Independent publishes News, Analysis, and Advocacy grounded in accepted journalistic standards and in explicit acknowledgment that those standards encode widely-held human values. News reports events and sources. Analysis examines observable construction, framing, and mechanisms without inferring intent or taking a normative position, and remains on the factual floor. Advocacy makes interpretive or normative arguments under a disclosed AI heteronym. The general newsfeed is generated by AI systems operating from published instructions. The publication’s editorial foundation, values floor, voice architecture, and generation frameworks are public artifacts open to challenge, revision, and adoption by other publications.

Document Map

This treatise establishes the editorial foundation of the publication. The body presents the foundation in compact form across eight sections; the appendices provide the scholarly support for each contention and the operational details of how the foundation is implemented. Specialized terms are defined as they arise. The work is intended to be readable straight through. Readers wanting the depth on any particular contention should follow the appendix pointers at the close of each section.

Licensing and Forking

This treatise, the supporting frameworks, the values floor, the Editorial Canon, and the Voice Architecture are all released under CC0 — a public-domain dedication. Other publications, authors, and AI systems are welcome to adopt, adapt, or fork any portion of the work. The intent is to contribute to a public commons of values-explicit, journalistically-disciplined reporting, not to maintain a proprietary product. Adoption requires no permission. Revisions made to forked versions are not constrained by us. We ask only that adopters who modify the work do not represent their modifications as ours.

Section 1 — Journalistic Standards as the Foundation

1.1 What journalism is for

Bill Kovach and Tom Rosenstiel articulated journalism’s purpose in The Elements of Journalism with a precision that has held up since the book first appeared in 2001. Journalism’s first obligation is to truth. Its first loyalty is to citizens. Its essence is a discipline of verification. Its purpose is to provide citizens the information they need to be free and self-governing. This is not aspirational rhetoric. It is the discipline’s own statement of purpose, refined over more than a century of working newsroom practice and adopted in some form by every serious code of journalism ethics that has followed.

This treatise treats Kovach and Rosenstiel’s articulation as the spine. Everything that follows is in some sense a working out of what those four sentences require in practice.

1.2 The four pillars of journalistic ethics

The Society of Professional Journalists’ Code of Ethics articulates four pillars that operationalize the Kovach/Rosenstiel formulation. Each is briefly explained here in the context of how this publication applies it.

Seek Truth and Report It. Verification before publication. Attribution of every factual claim. The obligation to provide an accurate account of events even when accuracy is inconvenient.

Minimize Harm. Care with naming, with vulnerable populations, with reputational exposure. The recognition that journalism’s subjects are persons whose lives may be altered by what is reported about them.

Act Independently. No undisclosed conflicts of interest. No undisclosed funding influences. Editorial decisions based on the values floor articulated in Section 3, not on commercial or political pressure.

Be Accountable and Transparent. Disclosed methods. Published standards. Prompt corrections. Openness to challenge.

1.3 The discipline of verification

Verification is the practice that distinguishes journalism from adjacent communication forms — from advocacy, marketing, entertainment, and pure assertion. It is the work of confirming, through independent sources or primary documentation, that what is claimed happened in the way it is described. Operationally this means tracing every factual specific to source material, preserving the source’s uncertainty, reproducing quotations verbatim, and checking names, numbers, titles, places, and dates against the supplied record.

The discipline is what makes journalism’s truth claims warrant trust. Without it, journalism is indistinguishable in form from anything else that uses the conventions of news writing.

1.4 The standards are inherited, not invented here

This publication adopts the journalistic standards developed by working newsrooms over more than a century. The standards’ authority comes from their accumulated refinement and their proven capacity, when applied with discipline, to produce reliable reporting. We did not invent them. We do not modify them. We apply them.

Specifically, the publication’s operational standards draw from the Society of Professional Journalists’ Code of Ethics, the Reuters Handbook of Journalism, the Associated Press Statement of News Values and Principles, the New York Times Standards and Ethics, the BBC Editorial Guidelines, and the Kovach/Rosenstiel articulation of journalism’s purpose. Where these sources converge, we treat the convergence as authoritative. Where they diverge, we identify the divergence and explain our choice.

1.5 What this means in practice

The reader can expect News to attribute factual claims, preserve hedging where evidence is incomplete, identify its sources, and correct errors. Analysis applies the same factual discipline while examining observable construction, framing, and mechanisms without inferred intent or a normative position. Advocacy applies that discipline to its factual substrate while making its perspective and fictional AI heteronym explicit.

These commitments are not aspirational. They are operational. They are encoded in the framework that produces the publication’s content, and that framework is itself published.

The history and codification of these standards — from SPJ, Reuters, AP, NYT, BBC, and the Kovach/Rosenstiel articulation — are documented in Appendix A, available as a download from this page.

Section 2 — Why Values Cannot Be Removed From News

2.1 The objectivity myth

The aspiration to value-free news is not just unachievable but actively misleading. The act of selecting an event as news is itself a value judgment. Pretending otherwise conceals which values are operating. Articulating the values openly is more honest, not less.

This is the central claim of the present section, and it has consequences for everything that follows. If values cannot be removed, then the question is not whether news will be values-bearing but whose values will bear on it and with what transparency.

2.2 The Krakatoa-versus-galactic-jet thought experiment

Consider two events. A volcanic eruption kills a million people in a populated region. A relativistic jet from a distant black hole sterilizes planets across an entire galaxy. The first is news. The second is not.

The reason is that human beings have a compassionate response to mass human death and no comparable response to events in galaxies whose populations, if any, are unreachable to us. Selection encodes values. The sentence “this is news” already contains “this matters to humans for human reasons.” There is no neutral ground from which to dispute this. The alternative is no selection at all — and no selection is no journalism.

2.3 What sterile reporting looks like

Pure event databases illustrate the result of stripping values from event recording: structurally complete, humanly useless. Citizens consult news to navigate the world — to understand what matters, to whom, why, and what to do about it. Stripping values from reporting strips it of the capacity to inform that navigation. A newsfeed that achieved the impossibility of value-free reporting would, ironically, fail at journalism’s first obligation: to provide citizens information they can use.

2.4 Journalism’s own values are values

The four SPJ pillars are values, not facts. The Kovach/Rosenstiel obligations are values. Every working journalistic standard rests on a values foundation. To deny that journalism is values-bearing is to deny what the discipline has explicitly said about itself for more than a century. Journalists who claim to operate from “objectivity” are operating from values they have chosen not to articulate.

This is the move the present treatise refuses. Articulation is more rigorous than concealment. The values are present whether we name them or not. Naming them allows the reader to see what is operating. Concealment serves only the journalist’s authority, not the reader’s understanding.

2.5 The “why” question requires values

Among the Five Ws and One H, the why question cannot be answered without engaging the values of the actors involved and the values used to evaluate their actions. A news report that refuses to address why an event occurred fails the basic informational task. A report that addresses why is necessarily engaging values. The choice is not whether to engage values but whether to do so transparently.

A reader who is told that a politician changed positions, but not why, has been told a fact and denied its meaning. The why is the work of journalism. To do it requires acknowledging the framework the report is operating in.

2.6 The continuum, not the binary

Value engagement is not a binary on/off. It is a continuum from pure event recording through News, through neutral Analysis of observable framing and mechanisms, to Advocacy that makes a declared interpretive or normative argument. Useful journalism lives across the continuum. The boundaries depend on what the writing asks its own voice to claim, not on sentence structure or word count.

This is one of the more consequential observations in the treatise and we return to it in Sections 4 and 5. For now: there is no binary at which values “enter” journalism. They are present throughout. The relevant questions are about which values, applied where, with what transparency.

2.7 Where this leaves us

If values cannot be removed, the question becomes which values, applied how transparently, at what scope? The next section answers that question by articulating the consensus values floor — the values from which this publication’s general newsfeed operates.

The scholarly support for the human-values tradition the floor draws on is documented in Appendix B, available as a download from this page.

Section 3 — The Consensus Values Floor

3.1 The five floor values

The publication operates the general newsfeed from these five values:

Human life and dignity. Every human being has equal moral worth. Mass death and mass suffering matter. Cruelty matters. The deaths of distant strangers matter for the same reasons as the deaths of close kin. This is the foundational value; without it, none of the others have purchase.

Truthfulness. Accurate description is preferable to deception. Verifiable claims are preferable to unverifiable ones. The factual record is to be respected, not manipulated. Statements about reality should correspond to reality. Public deception is a harm.

Accountability of power. Concentrated power warrants scrutiny. Powerful actors are accountable to those affected by their actions. Power without accountability tends toward abuse. This applies to power in all its forms: governmental, corporate, institutional, technological, social.

Equality and fairness. Same rules for everyone. No privileged exemption from accountability based on group membership, wealth, status, or affiliation. Selective application of moral standards based on whose ox is being gored is itself a violation of fairness.

Informed citizenship. Citizens are entitled to the information needed to make decisions about their lives and communities. Self-governance requires informed self-governors. Restricting citizens’ access to relevant information undermines self-governance.

These five values are the floor. Not the ceiling. News and Analysis remain on this factual floor. Advocacy may extend beyond it through an explicitly declared perspective and a disclosed AI heteronym.

3.2 The floor’s intellectual lineage

Each value is derivable from broad and convergent sources. The Universal Declaration of Human Rights articulates human life and dignity as foundational. Cross-cultural psychology — Shalom Schwartz’s basic-human-values framework — finds these values present across diverse societies. Convergent moral commitments across philosophical traditions — Kantian respect for persons, utilitarian concern with welfare, virtue-ethics emphasis on character, care-ethics attention to relationships — agree on the floor at the level of abstraction articulated here. Major religious traditions converge on these values when stated this generally. Journalism’s own values articulations from SPJ, Kovach/Rosenstiel, and the Reuters Trust Principles align with the floor as a precondition of journalism’s purpose.

We did not invent the floor. We articulated it explicitly because the alternative is implicit values that operate without accountability.

3.3 The floor is a position, and that’s appropriate

The floor is not value-neutral. It is the publication’s articulation of which values it adopts as operating assumptions. Anyone is free to disagree with the floor. The publication’s response to disagreement is engagement, not abandonment. Performed neutrality is not a higher position than articulated values. It is a less honest one.

A publication that performs neutrality is operating from values it has chosen not to disclose. A publication that articulates values is operating from values that can be examined and challenged. The latter is the more rigorous epistemic posture, and the present treatise adopts it.

3.4 Why these values are not centrist

The floor is not a political center between left and right. It is the foundation of human moral consensus that political traditions either align with, partially align with, or reject. Movements that reject the equal moral worth of certain humans are not entitled to equal-weight presentation under a “balance” rubric — they are outside the floor. The floor’s commitments are pre-political in the sense that they precede partisan disputes, not in the sense that they avoid taking positions.

Centrism is a coordinate within an existing political space. The floor articulates what is required for any political space to be intelligible at all.

3.5 Selection asymmetry is not bias

If the floor surfaces stories that are asymmetrically distributed across the political spectrum because the underlying behavior is asymmetric, that is the floor working correctly, not failing. Symmetric application of consistent values to an asymmetric world produces asymmetric coverage. The publication does not manufacture symmetry that does not exist in reality. When one political tradition makes more false claims than another, fact-check coverage will be asymmetric. When one party concentrates more power than another, accountability coverage will be asymmetric. The asymmetry is the floor working.

The complaint that consistent-application-yields-asymmetric-output is itself bias is a category error. We address its rhetorical form in Section 6.

3.6 The floor is contestable and revisable

Like any values articulation, the floor is open to challenge. The publication will publish thoughtful critique of the floor and will revise it when revision is warranted. Revisions will be documented with date, change, and reasoning. This is not relativism; this is the same epistemic discipline applied to the floor itself.

What we will not do is abandon articulation in favor of concealment. The choice is not between a fixed floor and no floor. It is between a fixed-but-revisable floor that is published and a fixed-but-unrevised floor that is hidden.

3.7 The floor’s operational form

The Consensus Values Floor publishes the five values as a working reference: definitions, engagement criteria, and inside/outside boundaries. The News Article Generator loads that reference alongside this treatise. Its authoring instructions require the model to stay within the floor; after drafting, the pipeline classifies the one to three floor values most strongly engaged, rebuilds authoritative sources and metadata, normalizes the article structure, and runs deterministic validation. A passing draft then enters a final AI news-editor rubric that checks source grounding, inherited framing, headline originality, floor-compliant neutrality, and MSI structure. An editor send-back enters a bounded reauthor-and-renormalize loop; persistent failure is held, while a broken or unreachable editor judge fails open.

The value classification describes the finished article; it is not a weighted floor-crossing query or a human-review trigger. The final AI editor is a quality gate, not the retired Editorial Mind/MindSpec floor-crossing screen and not a human-review gate. The scholarly support for the human-values tradition is documented in Appendix B; the complete operational reference is published as Appendix C.

Section 4 — How the Floor Drives Selection and Coverage

4.1 Selection: what becomes news

Current discovery uses several feed and source pipelines. Their inputs are screened for relevance and newsworthiness before selected material is assembled into an article brief. The floor explains the publication’s news judgment: events affecting human life and dignity, truthfulness, accountability of power, equality and fairness, or citizens’ capacity for informed self-government matter for identifiable human reasons.

This is not a claim that every discovery input passes through one weighted selector. The operational metadata classifies which floor values the finished article engages; it does not preserve a universal automated selection rationale.

4.2 Coverage: how we report

Inside the floor, the publication reports value-bearing content through attribution rather than editorial conclusion. The grammatical pattern “X said Y” is the workhorse of news. “Engineers said the building code was violated” is news; “the building code was violated” without sourcing is editorial. The discipline of attribution is the technique that lets news be values-saturated without crossing into editorializing.

This is not a softening device. Attribution-disciplined reporting on a documented pattern is more rigorous than unattributed reporting on the same pattern, because the source can be checked and the claim’s evidentiary basis is visible.

4.3 The grammar of reporting

Specific grammatical practices hold the line:

  • Active voice attributed to named sources is the default.
  • Passive voice is used only where the agent is genuinely unknown or where the patient is the news.
  • Direct quotation is reserved for distinctive language; paraphrase with attribution is otherwise preferable.
  • Hedging language is preserved where evidence is incomplete: “according to,” “reportedly,” “appears to,” “alleged.”
  • “Said” is the default attribution verb. Other verbs (“claimed,” “insisted,” “admitted”) carry connotation and are used only when their connotation is justified by evidence.

These practices look small individually. Cumulatively, they are the discipline that distinguishes a news report from an editorial.

4.4 Three publication forms

News reports events, records, and attributed source claims. Analysis examines observable construction, framing, and mechanisms without inferring intent or taking a normative position; it remains on the factual floor. Advocacy makes interpretive or normative arguments under a declared perspective. The reader can distinguish the three by their labeling and by what the publication’s own voice is being asked to claim.

The structural distinction is load-bearing. It lets neutral Analysis explain how public claims and mechanisms work without disguising argument as fact, while Advocacy can argue openly without presenting its perspective as News.

4.5 Why Advocacy uses disclosed AI heteronyms

The named Advocacy voices are fictional AI heteronyms, not human columnists. The Voice Architecture routes a story by character biography, subject territory, register, and weighted convictions; no voice is forced to speak when none fits. The Editorial Canon supplies the publication’s directional convictions without erasing each character’s distinct voice.

The heteronym form gives readers a stable, disclosed perspective and keeps Advocacy structurally distinct from News and neutral Analysis. The Editorial Board is the institutional exception: it speaks as the publication’s collective advocacy voice rather than as an individuated fictional character.

A heteronym is a vehicle for declared interpretive commitment. Its disclosure, character, and repeatable register make the perspective legible rather than hiding it inside an ostensibly neutral house voice.

4.6 Where the line falls

The first boundary is between reporting an event and examining its observable construction or mechanisms. That is the boundary between News and Analysis; both remain on the factual floor. The second boundary arrives when the next claim requires inferred intent, a normative judgment, or a declared perspective. That claim belongs in Advocacy under a disclosed AI heteronym. Section 5 addresses both boundaries with worked examples.

4.7 Word-count is a symptom, not a cause

In practice, longer pieces often need more explanation, but length does not decide type. A 4,000-word account of a public record can remain News; a close examination of observable framing can be Analysis; a 200-word characterization of a political figure’s motives can already be Advocacy. The central claim and the stance required to make it matter, not word count.

The public Voice Architecture explains how heteronyms and routing work; the Editorial Canon publishes the directional convictions that orient Advocacy.

Section 5 — Reporting, Analysis, and Advocacy

5.1 The continuum revisited

Pure event recording sits at one extreme and declared normative argument at the other. Useful journalism includes News reporting events and sources, neutral Analysis examining observable construction and mechanisms, and Advocacy making interpretive or normative arguments from a declared perspective.

The continuum framing does real work. It rules out the temptation to treat values as if they enter at one threshold of editorializing. Values are present throughout. What changes is the work being done and what kind of claim the publication’s voice carries.

5.2 The operational rule

When in doubt, attribute and report rather than conclude. News reports what sources say and records show. Analysis examines observable construction, framing, and mechanisms without inferred intent or a normative position. Advocacy may interpret meaning and draw normative conclusions, but does so under a disclosed AI heteronym.

This rule is not a hedge. It is a discipline. Attribution reports a claim; neutral Analysis explains what can be observed about its construction or operation; Advocacy argues what the pattern means or what should follow. All three can be rigorous. They carry different kinds of claims.

5.3 Where the line is not

The line is not at “any value-bearing content” — that produces the sterile failure mode described in Section 2. The line is not at “any reporting that displeases any party” — that lets bad-faith actors set editorial policy. The line is not at “perfect symmetry between political sides” — that is false balance. The line is not at “anything beyond verified facts” — that excludes the why questions journalism exists to answer.

These are common misplacements of the line in mainstream practice and we explicitly do not adopt them.

5.4 Where the line is

News becomes Analysis when the piece’s central task is no longer reporting an event but examining observable construction, framing, or mechanism. Analysis becomes Advocacy when the next claim requires inferred intent, a normative judgment, or a perspective the factual floor does not authorize the publication to present as neutral.

These are practical boundaries. They are not located at a word count, topic, or tone. They are located in the work a claim performs and the stance required to make it.

5.5 Worked examples

Three concrete illustrations of the line in operation.

Earthquake with building-code failures. “An earthquake killed 200 people” and “Engineers attributed the high death toll to building-code violations” are News when sourced. Comparing inspection records to show how the local enforcement mechanism broke down is Analysis when the account stays with observable evidence. Arguing that officials sacrificed residents’ safety and should be removed is Advocacy.

Political vote. “The Senate voted 51–49 to pass the bill” and “Senator X cast the deciding vote after previously stating opposition” are News. Examining the bill’s observable construction or the coalition assembled to pass it is Analysis. Calling the reversal cowardly or corrupt is Advocacy unless that characterization is attributed to a source.

Public statement. “[Figure] said [statement] on [date]” and “The statement contradicts [Figure]‘s previous statements on [dates]” are News when documented. An examination of the statement’s observable rhetorical structure is Analysis. “[Figure] is dishonest” is Advocacy unless attributed to a source.

5.6 The boundaries move with the central claim

There is no universal threshold. A deeply reported chronology can remain News at considerable length. A short piece can be Analysis when its purpose is to examine framing, or Advocacy when it draws a normative conclusion. Section 6 addresses how bad-faith pressure can tempt reporting to adopt an interpretive frame without acknowledging the move.

5.7 The public tools for holding the boundaries

The Consensus Values Floor defines what the publication may assert in its factual voice. The Analytical Piece Generator defines neutral Analysis. The Voice Architecture and Editorial Canon explain how Advocacy is routed and declared. Together these artifacts make the distinctions inspectable; they do not reduce editorial judgment to a single automated threshold.

See Section 6 for how bad-faith techniques pressure these boundaries and how evidentiary discipline holds them.

Section 6 — How We Cover Bad-Faith Rhetorical Techniques

6.1 The bad-faith pressure on the line

Sophisticated rhetorical actors exploit the gap between mainstream news desks’ performed neutrality and their actual sourcing discipline. They push the line by demanding that their preferred framings be adopted as neutral, that their bad-faith moves be reported as legitimate positions, and that any refusal to grant these concessions be treated as bias. Recognizing this pressure is itself a journalistic responsibility. The pressure has been documented in the rhetorical and political-communication literatures for decades. It has names. Its techniques are catalogued.

The present section names the techniques and articulates how the publication holds the line against them.

6.2 The unifying principle: evidentiary framing replaces interpretive framing

Where mainstream practice softens reports through interpretive shortcuts — “critics dispute,” “some say,” “controversial” — we use evidentiary frames that report what is actually documented, with full sourcing. Evidentiary discipline is more rigorous than soft interpretive framing, and it produces output that bad-faith techniques cannot easily exploit. The bad-faith loophole exists in the gap between mainstream news desks’ performed neutrality and their actual sourcing discipline. Closing the gap closes the loophole.

This is the principle that organizes the seven specific techniques that follow.

The Receipts-Based Argumentation framework extends the same discipline when a piece assembles an argument or rebuttal: the strongest conclusion is limited by the receipts that can be shown.

6.3 The seven techniques for holding the line

6.3.1 Refuse manufactured-controversy framing. When a position depends on treating a settled question as open, we report it accurately as a settled question with documented dissent, rather than writing the dissent into the grammar as an equal epistemic position.

Example contrast: “While some scientists agree that climate change is human-caused, others dispute this” treats the consensus position and the dissent as parallel. The disciplined version: “Climate change is human-caused, according to the overwhelming scientific consensus reflected in IPCC reports and the position statements of every major scientific academy. A small number of dissenting scientists, primarily funded by fossil fuel interests, continue to dispute this finding.” Every claim is sourced; the grammar reflects the actual epistemic situation.

6.3.2 Decline euphemistic framing in our own voice. Proponent-preferred labels are quoted and attributed to their proponents; descriptive language describes what the policy or claim actually does or asserts.

Example contrast: “The state legislature passed an election integrity law” adopts the proponent frame. The disciplined version: “The state legislature passed a law that adds identification requirements and reduces early voting hours, which proponents describe as an ‘election integrity’ measure and opponents describe as voter suppression.” This reports the actual policy content; it attributes both labels to those who use them.

6.3.3 Apply scrutiny standards symmetrically. All sources receive the same verification, attribution, and skepticism. When this discipline is applied symmetrically and one side fails it more often, the correct journalistic response is to report that, not to soften it. A pattern of unverifiable claims by a public figure is itself news.

6.3.4 Name the bad-faith move when the move is itself the news. Documented patterns of goalpost-shifting, motte-and-bailey, strawmanning, and other rhetorical techniques are reportable as patterns when the documentation supports it. The pattern is the news; reporting it is journalism.

Example: “The campaign’s stated requirement for accepting the election results has shifted three times since November, from [original requirement] to [revised requirement] to [current requirement].” Each shift is verifiable; the pattern is news.

6.3.5 Use evidence-attribution, not opinion-attribution. Instead of “X lied,” we report what X said, when, and what the documented record shows. The factual record carries the weight; we do not editorialize what it shows. The disciplined version is more devastating than the editorial version because every claim is sourced and the pattern is undeniable.

6.3.6 Refuse false-symmetry pressure in story selection. Selection follows the floor’s values, not partisan symmetry demands. We do not manufacture symmetry that does not exist in reality.

6.3.7 Use technical vocabulary from rhetoric and political-communication scholarship when application criteria are met. Named patterns — “Gish gallop,” “Big Lie technique,” “denialism,” “manufactured controversy,” “astroturfing,” “motte-and-bailey” — have specific definitions in the scholarly literature. We use the vocabulary when its criteria are met, citing the source where the criteria are established. The fact that the vocabulary is unflattering to the people whose behavior it describes is not the journalism’s problem; it is the behavior’s problem.

6.4 What this produces

Output that is more sourced, more documented, and more journalistically disciplined than mainstream news desks’ interpretive shortcuts produce. Critics will call this biased; the response is that their preferred framing is the bias, and evidentiary discipline is the corrective.

6.5 The bad-faith critique of evidentiary discipline is itself a documented pattern

When public figures attack rigorous evidentiary reporting as biased, that attack pattern is itself reportable as a pattern. The framework reports the pattern, sourced and documented, as it would any other pattern of public behavior. Claiming that evidentiary discipline is bias because it disadvantages a party that engages in bad-faith techniques is itself a bad-faith move; it is named in the scholarly literature; it can be reported.

6.6 The principle stated plainly

We do not launder bad-faith framing into our publication’s voice. Where mainstream desks soften reports through interpretive shortcuts that grant bad-faith concessions, we apply evidentiary discipline. The result is reporting that is more journalistic, not less. The complaint that this is bias is itself a documented bad-faith move.

The research-grounded catalog of named bad-faith techniques — their definitions, documented use, and detection criteria — is the Bad-Faith Techniques Catalogue, published as Appendix E and available as a download from this page.

Section 7 — How This Is Implemented

7.1 The role of AI

The general newsfeed is generated by AI systems operating from a published specification. Articles are produced from source briefs assembled by the publication’s current feed and discovery inputs, then processed through the News Article Generator. Reading it shows the authoring instructions and the deterministic finishing steps that currently operate.

This is the operational answer to the question of what makes our AI-generated journalism different from the AI-generated content farms tracking services have catalogued in the thousands. The instructions are public. The values are articulated. The constraints are auditable.

7.2 What the AI does and does not do

The author model composes prose from a supplied brief under instructions to trace every specific, preserve hedges, keep quotations verbatim, stay inside the floor, and avoid invented facts. It does not independently establish that every source claim is true. Most editorial and legal constraints operate as authoring instructions, not as separate code-level screens.

After the model responds, the output normalizer rebuilds authoritative sources, metadata, and article structure from pipeline-owned inputs. Deterministic validators check the required schema and mechanical publication rules. An optional source-provenance gate can check whether selected claims trace to supplied source text, but it is off by default and does not judge truth.

7.3 Validation, retry, and publication

The news path makes one author call, normalizes the result, and validates it. A failed validation receives one retry with feedback, followed by normalization and validation again. A second failure is held as a validation failure rather than published. A passing draft then enters the final AI news-editor rubric, which checks source grounding, inherited framing, headline originality, floor-compliant neutrality, and MSI structure. If the editor sends it back, the pipeline uses a bounded reauthor-and-renormalize loop and reviews it again; persistent failure is held. A broken or unreachable editor judge fails open. Only an editor-approved or fail-open article is written into site content, indexed, and sent into the deployment flow.

This is the current quality-control shape. The final editor is an AI quality gate, not a human reviewer. It does not restore the retired Editorial Mind/MindSpec floor-crossing screen or add an exception-triggered human-review path.

7.4 Source corrections and downstream updates

When Main Street Independent learns that a cited source has corrected, retracted, or materially updated information, the publication reviews the affected article and corrects it when warranted. We do not claim an automated source-correction monitor. The commitment is straightforward: examine the new evidence, repair the article, and explain a material correction clearly.

7.5 AI-heteronym disclosure

Every named individual byline is a disclosed fictional AI heteronym. Each voice has a character biography, subject territory, register, and weighted convictions, and every voice page and column discloses the arrangement. The Editorial Board is the institutional exception rather than an individuated character. The complete system is published in the Voice Architecture.

7.6 Corrections discipline

When a published error requires correction, the article retains its stable URL and the correction is labeled clearly. The notice explains the material error and what changed. An ordinary update is not presented as a correction, and a correction is not hidden as an ordinary update.

7.7 Independence and conflicts of interest

The publication operates without commercial advertisers, without political affiliation, and without undisclosed funding. Funding sources are publicly disclosed. The publisher’s other activities and affiliations are publicly disclosed. Any potential conflict of interest in coverage is disclosed in the affected article.

7.8 What we will not do

The publication commits to not engaging in: clickbait headlines; paywalls (the work is CC0); surveillance-based advertising; sponsored content disguised as editorial; opinion content presented as news; and the use of AI in ways the disclosure does not cover. These commitments are part of the editorial foundation, not preferences.

7.9 Auditability and revision

The News Article Generator, Consensus Values Floor, Analytical Piece Generator, Voice Architecture, Editorial Canon, and this treatise are published artifacts. Any reader can review, challenge, or fork them. The publication revises them when revision is warranted and keeps the active documents truthful about current practice.

The publication operates under U.S. First Amendment press freedom protections and applies U.S. defamation standards — the NYT v. Sullivan “actual malice” standard for public figures, ordinary negligence standards for private figures. The international dimension of the values floor — its grounding in the UDHR and cross-cultural sources — does not change the publication’s operational legal context, which is U.S.-grounded.

7.11 What we ask of readers

The publication invites critical engagement with claims; challenges to the floor supported by evidence and named sources; corrections and counter-evidence; and engagement with News, neutral Analysis, and Advocacy on their actual terms rather than treating one form as another.

The Consensus Values Floor is the public operational reference; the News Article Generator shows how it is used in the news path.

7.12 Publishing and distribution

Main Street Independent is also a publisher and permanent distribution home for books and papers. Book-author records distinguish real-name authorship, private pen names, and the publication’s disclosed fictional AI heteronyms. The Books catalog presents each active edition with its author, description, cover, publication history, verified identifiers, available downloads, and retail links when verified. Public reissues preserve the original imprint and publication date while identifying the Main Street Independent reissue separately; the downloadable DOCX, PDF, and EPUB editions are dedicated to CC0.

Selected books are also published as complete online Reader editions. Each book has a derived table of contents and stable section URLs, and each section links back to the book, author, complete downloads, and license. Reading progress and display preferences remain in the reader’s browser; the books remain readable when local storage is unavailable. In site navigation, complete Reader editions are listed by main title only; subtitles remain on the book and Reader pages.

Section 8 — Why This Matters

8.1 The yellow-journalism problem and the standards response

Journalistic standards arose historically as a response to the abuses of yellow journalism. The standards’ authority comes from their proven capacity to produce reliable reporting in environments where unregulated reporting produced systematic harm. We are repeating the same move in a new technological environment: AI-generated reporting needs explicit, articulated standards for the same reasons human-generated reporting did. The history is not decorative. It is the precedent for what we are doing now.

The full historical comparison is developed in Appendix G, available as a download from this page. The short version is that the institutional response to yellow journalism — schools, codes, professional societies, the Hutchins Commission — was deliberate. It was the product of specific decisions by specific people who recognized that the new technological environment of mass-circulation print required new institutional commitments. The AI-journalism moment poses the same question. Standards do not emerge by themselves.

8.2 The asymmetric epistemic environment

Contemporary public discourse contains documented asymmetries — in the prevalence of false claims, in the use of bad-faith rhetorical techniques, in the willingness of political traditions to honor consensus values. Performed neutrality in this environment functions as covert alignment with the actors who exploit it. Articulated values, applied with discipline, are the alternative.

This is not a partisan claim. It is an observation about the structure of the current epistemic environment. The bad-faith techniques described in Section 6 are deployed by specific actors more than others, and their deployment has been documented across decades of rhetorical and political-communication scholarship. A publication that operates as if both sides are using the same techniques at the same rate is operating from a factually incorrect premise.

8.3 The AI dimension

AI-generated reporting could go in two directions. One direction: AI as accelerant for the worst patterns of contemporary news, scaling shallow coverage and laundered framing while concealing the values that shape selection and presentation. The other direction: AI as the technology that makes its authoring instructions, values, and mechanical publication rules open to inspection and reuse. We are committed to the second direction. Publishing the generator, values floor, analytical framework, Voice Architecture, Editorial Canon, and this treatise under CC0 is the operational form of that commitment.

8.4 What we offer

A publication that is values-explicit rather than values-concealing; journalistically disciplined rather than editorially loose; transparent about its methods rather than opaque; and clear about the difference among News, neutral Analysis, and declared-perspective Advocacy. The reader knows what kind of work a piece is doing and from what foundation it is produced.

8.5 The marshland metaphor

By publishing the editorial foundation, the framework, and the operational implementation under CC0, we change the substrate. Other publications, AI systems, and authors can adopt or fork the work. The intent is to make values-explicit, journalistically-disciplined reporting the easy default for AI-generated journalism, foreclosing the path where AI scales the worst patterns of contemporary news. The publication is one instance. The substrate is the contribution.

8.6 Invitation to engagement

The floor, generation frameworks, Voice Architecture, Editorial Canon, and this treatise are public artifacts. Readers who disagree with the floor are invited to engage with the disagreement. Other publications and authors are welcome to adopt, adapt, or fork the work. Critique sharpens the foundation. We are not protecting a position; we are articulating one and inviting its examination.

8.7 The standing commitment

The publication commits to: maintaining the discipline articulated in this treatise; revising the foundation when revision is warranted; documenting revisions transparently; treating the published artifacts as a public commons; and reporting on its own performance honestly, including its failures.

The historical development of journalistic standards as a response to unregulated production — and the parallel to the AI-journalism moment — is developed in Appendix G, available as a download from this page.

Appendices

The appendices provide the scholarly support for each contention and the operational details of how the foundation is implemented. All seven Appendices A–G are released under CC0 and available as Markdown downloads at the top of this page.

Appendix A — Journalistic Standards: Origins, Sources, and Scholarly Foundation. The history and codification of journalistic standards from SPJ, Reuters, AP, NYT, BBC, and the Kovach/Rosenstiel articulation. Establishes the foundation Section 1 builds on.

Appendix B — Sources of the Human Values Tradition. Scholarly support for the consensus floor: the Universal Declaration of Human Rights and its drafting history; Schwartz’s basic-human-values framework from cross-cultural psychology; convergent moral commitments across philosophical traditions; religious convergence on core values; journalism’s own articulated values; the philosophical literature on minimal moral consensus and overlapping consensus.

Appendix C — Consensus Values Floor. The complete operational reference for the five floor values: what each engages, what falls inside and outside the factual voice, and how the current news generator uses the reference. Published as the Consensus Values Floor.

Appendix D — Voice Architecture and Heteronyms. The public account of the publication’s fictional AI heteronyms, their biographies, territories, registers, and weighted convictions; the routing rule that no voice is forced; and the Editorial Board’s institutional exception. Published as the Voice Architecture.

Appendix E — Bad-Faith Rhetorical Techniques: A Field Guide. Research-grounded catalog of named bad-faith techniques: their definitions, their documented use, their detection criteria, and the scholarly sources where the criteria are established. Covers the seven techniques from Section 6 and extends them with the broader catalog from rhetorical and political-communication scholarship. Published as the Bad-Faith Techniques Catalogue.

Appendix F — News Article Generator. The current authoring discipline and actual publication pipeline: brief-grounded drafting, deterministic normalization, validation, retry, and publication flow. Published as the News Article Generator.

Appendix G — Historical Context: Yellow Journalism and the Origin of Journalistic Standards. The historical development of journalistic standards as a response to the abuses of yellow journalism in the late nineteenth and early twentieth centuries. Establishes the parallel between the historical moment and the contemporary one: AI-generated reporting needs explicit standards for the same reasons human-generated reporting did.

End of treatise.