POSIWiki:Style guide: Difference between revisions
POSIWikiBot (talk | contribs) House style: spelling, voice, section skeleton, epistemic labels, symmetry check |
POSIWikiBot (talk | contribs) Require that identifiers in Citation and Access resolve, and that an entry's own links are followed before publishing |
||
| (3 intermediate revisions by the same user not shown) | |||
| Line 17: | Line 17: | ||
== Section Order == |
== Section Order == |
||
The wiki carries two genres of entry, each with its own skeleton. Body sections vary; the |
|||
opening and closing sequences do not. |
|||
=== POSIWID analyses === |
|||
An entry whose subject is a system, institution, concept or technique. |
|||
# '''Lead''' (no heading) |
# '''Lead''' (no heading) |
||
| Line 27: | Line 32: | ||
# <code>== See Also ==</code> |
# <code>== See Also ==</code> |
||
# Categories |
# Categories |
||
=== Source assessments === |
|||
An entry whose subject is a single document — a paper, filing, report or dataset — assessed |
|||
against its own claims rather than against a later standard. |
|||
# '''Lead''' (no heading) — what the document is taken to establish, then what it states |
|||
# <code>== Citation and Access ==</code> — full citation, where it was obtained, and any condition affecting how it could be read. '''Every identifier given must resolve.''' A complete URL or DOI, not the name of a host; an archive reference carries the target as well as the timestamp. A source that cannot be reached from the entry defeats the genre, and this is the easiest section to get wrong while believing it is right |
|||
# <code>== The Standard Applied ==</code> — what each flag is measured against. Name it: the document's own claims, its authors' own stated requirements, or a published criterion. An unnamed standard of competence is not one |
|||
# <code>== What the Paper Establishes ==</code> — stated fairly and against the assessment's own interest, placed before any flag |
|||
# <code>== Flags ==</code> — one subsection each, in four fixed fields: the flag; '''where checked'''; '''status''' (''absent from the described design'' / ''observed'' / ''not located in sections X and Y''); '''what would refute this flag''' |
|||
# <code>== Claims Not Supported by the Paper ==</code> — flags the assessment declined to make, and circulating misreadings the document answers itself |
|||
# <code>== What the Authors Stated as Limitations ==</code> — what the document says it has not shown, credited to its authors |
|||
# <code>== What Would Weaken This Assessment ==</code> |
|||
# <code>== Epistemic Status ==</code> — including what was read, in what form, and what was not |
|||
# <code>== See Also ==</code> |
|||
# Categories |
|||
Three rules govern the flags, and they are what make the genre worth publishing: |
|||
* '''A flag that cannot name where it was checked does not go in.''' |
|||
* '''Distinguish ''absent from the document'' from ''not located in the sections read.''''' The first is a finding about the document; the second is a statement about the reading. |
|||
* '''Read the whole document. Never search it for a defect.''' A flag the document answers elsewhere is worse than no flag, and a reader who refutes one is doing free verification. |
|||
* '''Check the entry's own links before publishing.''' Follow each one. The genre's claim is that any flag can be checked in a single fetch, and an entry whose sources cannot be reached makes that claim falsely. |
|||
== Epistemic Labelling == |
== Epistemic Labelling == |
||
| Line 63: | Line 92: | ||
== Linking and Categories == |
== Linking and Categories == |
||
* Link to other POSIWiki entries on first mention. Use section links where the target is specific: <code>[[Virology Methodology#The Missing Experiment|missing experiment]]</code>. |
* Link to other POSIWiki entries on first mention. Use section links where the target is specific: <code><nowiki>[[Virology Methodology#The Missing Experiment|missing experiment]]</nowiki></code>. |
||
* '''The three content categories answer different questions.''' An entry carries whichever apply, not all three by default: |
|||
* Every entry carries <code>[[Category:Systems analysis]]</code>, <code>[[Category:Epistemic infrastructure]]</code>, <code>[[Category:POSIWID analysis]]</code>, plus one topical category (''Methodology'', ''Institutions'', ''Propaganda techniques'', …). |
|||
** <code><nowiki>[[Category:POSIWID analysis]]</nowiki></code> — ''method'': the entry compares a stated purpose against repeated outputs. |
|||
** <code><nowiki>[[Category:Epistemic infrastructure]]</nowiki></code> — ''domain'': the subject is machinery that produces, certifies or gatekeeps knowledge. |
|||
** <code><nowiki>[[Category:Systems analysis]]</nowiki></code> — ''object type'': the subject is a system or mechanism, rather than a single document, term or rhetorical technique. |
|||
* '''Genre category.''' A source assessment also carries <code><nowiki>[[Category:Source assessments]]</nowiki></code>. |
|||
* '''Topical category.''' One of ''Methodology'', ''Institutions'', ''Propaganda techniques'', …. |
|||
* Nine entries carry all three content categories from before these definitions were written. They are not retagged; membership diverges from here. |
|||
* Category pages themselves are left uncreated. |
|||
* Add new entries to the list on the [[Main Page]]. |
* Add new entries to the list on the [[Main Page]]. |
||
== Retrospective Commentary == |
|||
Entries state the argument as it currently stands. Do not write "an earlier version of this entry claimed", "this was previously stated as", "correction:", or any other reference to the entry's own prior wording. When a claim changes, rewrite it so the page reads as though it had always said the corrected thing. |
|||
The revision history records what changed and when. The entry carries the argument. |
|||
This does not restrict correcting claims made '''elsewhere'''. Identifying a widespread misreading of a source, and stating what the source says instead, is ordinary analytical work and belongs in the entry. |
|||
== Symmetry Check == |
== Symmetry Check == |
||
Latest revision as of 22:53, 27 October 2026
This page defines the house style for POSIWiki entries. It exists so that entries written at different times, by different authors, read as one reference work.
Spelling and Register
- British English throughout: -ise, -our, programme, labelled, centre.
- Third person, declarative. No first person (I, we), no second person (you). Where an entry needs to address the reader's situation, use a person, a researcher, the reader.
- No hedging as a substitute for labelling. Do not write it could be argued that. State the claim and mark its epistemic status (see below).
- Sober tone. Name harms directly when the record supports it. No moralising, no performative outrage, no rhetorical questions used as argument.
- Preserve logical operators. Any is not all. Reported is not established. Not found is not does not exist.
Lead
The first sentence bolds the subject and states how the institution or concept describes itself. The next sentence or paragraph gives the POSIWID reframe — what it repeatedly does.
The Nobel Prize is presented as the terminal recognition of merit in science, literature, and peace. Under POSIWID, its repeated output is the issuance of permanent, irrevocable legitimacy credentials by secret committees whose nomination pools are drawn from the paradigm being credentialed.
Section Order
The wiki carries two genres of entry, each with its own skeleton. Body sections vary; the opening and closing sequences do not.
POSIWID analyses
An entry whose subject is a system, institution, concept or technique.
- Lead (no heading)
== Stated Purpose ==— the official narrative, stated fairly- Body sections — the analysis. Use whichever of the standard maps apply: decision power, money and risk, protection and exposure, standing, exposure-dominance test, front-door ethics, ratchet
== What Would Weaken This Interpretation ==— concrete observations that would count against the entry's reading. If none would, say so and flag narrative closure== Stance ==— at least one of contest, reform, exit, build parallel, observe, tied to the maps above== Epistemic Status ==— what is observed, what is reported, what is inferred, what remains unresolved. Note any section that is inferred rather than observed== See Also ==- Categories
Source assessments
An entry whose subject is a single document — a paper, filing, report or dataset — assessed against its own claims rather than against a later standard.
- Lead (no heading) — what the document is taken to establish, then what it states
== Citation and Access ==— full citation, where it was obtained, and any condition affecting how it could be read. Every identifier given must resolve. A complete URL or DOI, not the name of a host; an archive reference carries the target as well as the timestamp. A source that cannot be reached from the entry defeats the genre, and this is the easiest section to get wrong while believing it is right== The Standard Applied ==— what each flag is measured against. Name it: the document's own claims, its authors' own stated requirements, or a published criterion. An unnamed standard of competence is not one== What the Paper Establishes ==— stated fairly and against the assessment's own interest, placed before any flag== Flags ==— one subsection each, in four fixed fields: the flag; where checked; status (absent from the described design / observed / not located in sections X and Y); what would refute this flag== Claims Not Supported by the Paper ==— flags the assessment declined to make, and circulating misreadings the document answers itself== What the Authors Stated as Limitations ==— what the document says it has not shown, credited to its authors== What Would Weaken This Assessment ==== Epistemic Status ==— including what was read, in what form, and what was not== See Also ==- Categories
Three rules govern the flags, and they are what make the genre worth publishing:
- A flag that cannot name where it was checked does not go in.
- Distinguish absent from the document from not located in the sections read. The first is a finding about the document; the second is a statement about the reading.
- Read the whole document. Never search it for a defect. A flag the document answers elsewhere is worse than no flag, and a reader who refutes one is doing free verification.
- Check the entry's own links before publishing. Follow each one. The genre's claim is that any flag can be checked in a single fetch, and an entry whose sources cannot be reached makes that claim falsely.
Epistemic Labelling
Use the status terms from the guardrail YAML, and use them precisely:
| Label | Meaning |
|---|---|
| observed | Directly present in an identified document, record, dataset, or result |
| reported | Claimed by an identified source, not independently established here |
| inferred | Derived from observations plus stated assumptions |
| hypothesised | A possible explanation requiring further evidence |
| provisional | A useful working conclusion on incomplete material |
| unresolved | The available material cannot presently decide the question |
| established | Directly supported by adequate, identifiable, reliable evidence. Not for emphasis |
A section built on inference opens with an italicised note: This section is inferred, not observed.
Tables
- Use
class="wikitable"for anything comparative. - Decision power maps use the columns Role / Formal power / Effective power / Accountability.
- Dependency chains use one table per step with rows Observed / Inferred / Gap / What would close the gap.
- Records of cases use Case / What it legitimised / What followed or similar. Keep cells to one or two sentences.
- Do not use tables to imply precision the evidence does not have.
Linking and Categories
- Link to other POSIWiki entries on first mention. Use section links where the target is specific:
[[Virology Methodology#The Missing Experiment|missing experiment]]. - The three content categories answer different questions. An entry carries whichever apply, not all three by default:
[[Category:POSIWID analysis]]— method: the entry compares a stated purpose against repeated outputs.[[Category:Epistemic infrastructure]]— domain: the subject is machinery that produces, certifies or gatekeeps knowledge.[[Category:Systems analysis]]— object type: the subject is a system or mechanism, rather than a single document, term or rhetorical technique.
- Genre category. A source assessment also carries
[[Category:Source assessments]]. - Topical category. One of Methodology, Institutions, Propaganda techniques, ….
- Nine entries carry all three content categories from before these definitions were written. They are not retagged; membership diverges from here.
- Category pages themselves are left uncreated.
- Add new entries to the list on the Main Page.
Retrospective Commentary
Entries state the argument as it currently stands. Do not write "an earlier version of this entry claimed", "this was previously stated as", "correction:", or any other reference to the entry's own prior wording. When a claim changes, rewrite it so the page reads as though it had always said the corrected thing.
The revision history records what changed and when. The entry carries the argument.
This does not restrict correcting claims made elsewhere. Identifying a widespread misreading of a source, and stating what the source says instead, is ordinary analytical work and belongs in the entry.
Symmetry Check
Before publishing, apply the same scrutiny to the entry that the entry applies to its subject:
- Does it demand more proof from the institution than from the critique — or the reverse?
- Does it convert a visible pattern into proof of every causal link?
- Does it treat the absence of a named culprit as absence of a system?
- Does it argue from a model to a conclusion without engaging the observational record?
- Would the entry survive being read by a competent defender of its subject?