--- schema: "swft.publication/v1" id: "software-factory-methods" title: "Methods and corrections" description: "How SWFT labels evidence, selects AI engineering sources, reviews automated discovery, keeps answer copy visible, and corrects the record." summary: "SWFT labels each important claim so readers can see whether it was observed directly, found in a study, reported by the source, or concluded by SWFT. AI-drafted pages require a dated check by a second AI editorial-review agent (Codex) before admission." canonical: "https://swft.io/methods" author: "SWFT Editorial" author_type: "Organization" published: "2026-08-31" modified: "2026-09-02" kind: "policy" section: "About" tags: ["editorial method", "AI engineering sources", "evidence labels", "corrections policy", "answer engine optimization"] evidence_labels: ["INFERENCE", "OBS"] source_ids: ["ai-engineer-worlds-fair-2026", "dora-2025-errata", "hraness-home", "hraness-x", "latent-space-factory-loops", "smol-news-harness-scan"] authorship_disclosure: "AI-drafted from the cited public sources and independently checked by a second AI editorial-review agent (Codex) for source fit, claim boundaries, overlap, and reader utility. SWFT Editorial is responsible for corrections." --- # Methods and corrections How SWFT shows what it saw, what a source reported, what research found, and which conclusions are its own. > **Authorship:** AI-drafted from the cited public sources and independently checked by a second AI editorial-review agent (Codex) for source fit, claim boundaries, overlap, and reader utility. SWFT Editorial is responsible for corrections. SWFT wants readers to understand both the claim and how much confidence to place in it. Every article shows its sources, the kind of evidence behind each important statement, where the work came from, and the date of its source window. ## What do SWFT's evidence labels mean? ### Observed artifact (OBS) SWFT inspected a public artifact or dated event, such as a page, transcript, repository, release, or document. “Observed” proves what the artifact showed on its recorded access date. It does not prove every claim inside that artifact. ### Independent study (STUDY) The claim comes from research with a described method, sample, evaluation, or analysis. SWFT keeps the population and limits visible. One benchmark or study does not become a universal rule. ### First-party report (SELF-REPORT) A person or organization is describing its own product, system, customers, or results. First-party accounts can contain valuable operating detail. SWFT attributes them and does not present them as independent comparisons. ### Analysis (INFERENCE) This is SWFT's own definition, classification, recommendation, or conclusion drawn from named evidence. Readers can inspect the support and disagree with the interpretation. The short codes stay in Markdown and data exports so machines and repeat readers can identify the category reliably. The visible labels use everyday language. ## How does SWFT choose sources? SWFT prefers primary research and corrections, official documentation or artifacts, and detailed first-party operating accounts. Independent reporting follows. Secondary summaries help us discover a topic, but they should not carry a factual claim when the underlying source is available. [AI Engineer's 2026 program](https://ai.engineer/worldsfair/2026), [Latent Space](https://www.latent.space/p/aiewf-daily-dispatch-loops), and [Smol News](https://news.smol.ai/issues/26-08-17-not-much) are useful radar for current vocabulary and debates. SWFT follows those signals to the talk, paper, repository, release, specification, or operator account behind a claim whenever possible. Commercial affiliation does not disqualify a source. It changes what the source can establish. Product capabilities and company metrics remain first-party reports until independent evidence supports a stronger statement. ## How does SWFT choose company cases? A company enters the comparison when public first-party evidence describes a repeatable path from incoming work through agent action, verification, and a handoff or decision. General use of a coding assistant, one impressive demo, or a product announcement is not enough by itself. SWFT prefers cases with several dated sources, concrete operating detail, visible artifacts, or reported outcomes. A case can still qualify when the implementation is private, but the access limit and missing evidence must be stated. Inclusion is not a ranking, endorsement, or claim that the archive is complete. It means the public record is detailed enough to teach and compare without inventing the missing parts. ## What rules apply to claims? - Numbers name the organization, population, date, and method available in the source. - Benchmarks explain what tasks and scoring rules they cover. - Changing maps and product descriptions show their source-window date. - Quotations stay short unless SWFT has permission to republish the work. - “Unknown” stays unknown. Missing public evidence is not proof that something does not exist. - A model-generated summary never replaces the underlying source. ## How does SWFT use AI in editorial work? SWFT Editorial guides, comparisons, references, and company cases are AI-drafted from cited public sources. Their institutional byline and authorship disclosure appear in HTML, Markdown, feeds, and structured metadata. Ben Guo's personal byline is used only for his authenticated imported essay. Each admitted page has a dated review record naming the second AI editorial-review agent (Codex), the scope of that check, the evidence type, the closest overlap, the risk if a claim is wrong, and the maintenance trigger. SWFT does not imply that a human or site owner reviewed a page, and an AI editorial check is not qualified clinical, legal, or financial review. AI drafting does not change the evidence standard. Every factual claim still needs the appropriate source boundary, first-party reports stay labeled, and SWFT Editorial remains responsible for corrections. ## Can automation publish an SWFT article? No. Discovery automation may find, normalize, and group possible sources, but it cannot create a publication record or change a public route. A separately recorded editorial admission must identify the reader job, claim boundary, evidence fit, overlap decision, reviewer, risk, and maintenance plan before a page can enter the registry. One admitted record drives the HTML page, Markdown version, metadata, structured data, feeds, and source list. A question section is added only when it resolves a follow-up decision the article body does not already answer, and the same wording then powers visible and structured representations. SWFT does not maintain hidden answer-engine copy. ## Corrections ### How do I request a correction? [Send @hraness a public message on X](https://x.com/hraness) and include the SWFT page, the disputed sentence or record, and the strongest source you have. The SWFT repository is private, so a GitHub issue link would not be a usable public correction form. SWFT will reproduce the issue, decide whether it concerns fact, attribution, interpretation, link, or presentation, and correct the canonical record plus every derived surface. A meaningful correction changes the modified date. If the earlier version could materially mislead, the page receives a visible note. Spelling and presentation fixes that do not change meaning can remain silent. ## Current corrections No material corrections have been recorded as of September 2, 2026. The [DORA 2025 errata](https://dora.dev/research/2025/errata/) is one useful model: keep corrections close to the work they qualify and preserve the history. Corrections are part of a trustworthy publication, not something to hide. ## How we know - **Observed artifact (OBS)** DORA maintains a dated, version-specific errata page for its 2025 report. Sources: [DORA Research: 2025 Errata](https://dora.dev/research/2025/errata/). - **Observed artifact (OBS)** AI Engineer and Smol News publish dated public surfaces that SWFT uses as discovery radar for current terminology and source leads. Sources: [AI Engineer World's Fair 2026](https://ai.engineer/worldsfair/2026); [AINews: harness-level evaluation and secure execution](https://news.smol.ai/issues/26-08-17-not-much). - **Analysis (INFERENCE)** SWFT's four-label system, company-case selection rules, and correction workflow are the publication's own editorial policy. Sources: [DORA Research: 2025 Errata](https://dora.dev/research/2025/errata/); [AIEWF daily dispatch: loops and software factories](https://www.latent.space/p/aiewf-daily-dispatch-loops); [Hraness](https://hraness.com/). ## Sources - **Observed artifact (OBS)** [AI Engineer World's Fair 2026](https://ai.engineer/worldsfair/2026) — AI Engineer; accessed 2026-09-01. The official 2026 program groups software factories, harness engineering, context engineering, evals, sandboxes, memory, and agentic engineering into dedicated tracks. - **Observed artifact (OBS)** [DORA Research: 2025 Errata](https://dora.dev/research/2025/errata/) — DORA; updated 2025-11-24; accessed 2026-08-31. The official correction record for the 2025 report and its current versioning practice. - **Observed artifact (OBS)** [Hraness](https://hraness.com/) — Ben Guo; accessed 2026-08-31. The author's home publication and the provenance link for SWFT. - **Observed artifact (OBS)** [@hraness on X](https://x.com/hraness) — Ben Guo; accessed 2026-09-01. The explicit public contact channel for SWFT correction requests. - **Observed artifact (OBS)** [AIEWF daily dispatch: loops and software factories](https://www.latent.space/p/aiewf-daily-dispatch-loops) — Latent Space; published 2026-07-01; accessed 2026-09-01. A dated field dispatch connecting software-factory practice to the complete loop from signals and prioritization through execution, review, deployment, and learning. - **Observed artifact (OBS)** [AINews: harness-level evaluation and secure execution](https://news.smol.ai/issues/26-08-17-not-much) — Smol News / Latent Space; published 2026-08-17; accessed 2026-09-01. A secondary trend scan used for vocabulary discovery. SWFT follows its links to primary sources before making product claims. ## Read next - [About SWFT](/about) - [The software factory map](/map) - [Inside Uber's AI software factory: measuring cost and quality at scale](/companies/uber-software-factory)