WWS audit of the NYT Facebook post with the underlying article scored as a separate unit. The standard treats those as different objects when they diverge. The social
Final Article Integrity Score (card): ~32 — Do Not Use as Record
The card is worse because it strips the only on-record institutional sentences.
Routing meaning, per the rubric: reject the package. Do not quote its frame, labels, or implied causality as the file. If the event matters, start from the primary record as if the write-up did not exist. That is not a finding that the calls did not happen.Cotobuzz
Most readers will only consume the Facebook/social media card. Fore the Facebook post-only readers, Dose is a confirmation-bias machine. That is a consumption warning, not a claim about either side’s underlying facts.
WWS already separates the clocks. AIS scores the text. PAS warns about the wrap. BCS asks whether an older file is doing unpaid work inside this object. On this card the answer is Dose: the lede and moral still stand if you strip the new alleged facts. A reader who never opens the article is not receiving a first look. They are receiving a lockup. Cotobuzz
How the card finishes the thought without the record
The inspectable card is: exclusive badge, cropped face, “unusual step,” “unvarnished,” “sobering,” “did not line up,” “current and former officials.” It does not carry Parnell, Cheung, a named commander, a dated call log, or an openable stockpile. Those absences are What’s missing. Dose is the other shrug: the reader already “knows who he is.”
The live house file had already taught the protagonist before 10 September: Vance opposed the war, private assessments were darker than the public victory line, the gap costs him with Trump. The card does not introduce that figure. It doses him. “Unvarnished” and “did not line up” are not new metrics. They are passwords to the old file.
Why that supports confirmation bias specifically on the card
Confirmation bias here is mechanical, not tribal:
If the reader already holds the gulf-story, the card is complete. Portrait plus contrast dek plus anonymous officials is enough to refile “private truth, public spin.”
If the reader already holds the smear-story, the card is also complete. Same objects refile “anonymous officials, Times exclusive, preloaded villain.”
Neither reader needs the Patriot sentence, the access statement, or a definition of “won.” C4 is exactly that test: the feed version still works if the new facts are removed.
A Coat article can still add work after the fold. This body does add alleged mechanism — direct COCOM calls, theater interceptor language, on-record process quotes. Card-only consumption never reaches that layer, so Coat on the article does not rescue Dose on the lockup.
What the warning does not say
WWS refuses party valence as a criterion. “Dose supports confirmation bias” does not mean the calls did not happen, and it does not mean the public statements were true. It means the distributed object can close a prior plot without forcing a trip to records.
Hardened sticker for card-only readers
What’s missing on the card: named commanders, dates, inspectable inventory, the actual public quotes, Parnell, Cheung.
BCS: Dose on the lockup.
Routing for that audience: do not quote the card’s frame. If the event matters, start from the statements and identifiers as if the wrap did not exist. The bias path is the package doing the moral work before the dossier is allowed to speak.
L
Outlet / headline / date / URL: The New York Times — “Behind the Scenes, Vance Gathered Unvarnished Views of the Iran War,” Jonathan Swan, Adam Entous, Maggie Haberman, 10 Sept 2026. Card points at https://nyti.ms/4iqxgPP / article URL https://www.nytimes.com/2026/09/10/us/politics/vance-iran-war-trump-munitions.html.
Units scored: Facebook social card and article body (visible text plus contemporaneous excerpts). Full body is paywalled; scores use the inspectable text, not imagined paragraphs.
Genre: national-security exclusive / live politics investigation. Per WWS, this class usually lands 50–79 and cannot enter Archive Grade on a card.
1. Five-minute red-flag scan
ID
Flag
Card
Article body
Trigger
R1
Undefined load-bearing label
Y
Y
“Unvarnished,” “sobering,” “unusual step,” dek contrast “did not line up.” Body adds the verdict “falsely claim.” No metric for “won,” “demolished,” or “war.”
R2
Anonymous insulation on the decisive fact
Y
Y
Spine is “current and former officials” / “people who spoke to him” / “officials with direct knowledge.”
R3
Missing adverse ground
Y
N (partial)
Card has none. Body includes on-record process statements from Sean Parnell (Pentagon) and Steven Cheung (White House). Those address access and teamwork, not the stockpile or “victory” dispute.
R4
Scope leap
weak
Y
Months of calls become a gulf between private reality and public speech, plus a 2028-campaign shadow in the body.
R5
Middleman supplies the moral
Y
Y
Unnamed officials and the reporter voice deliver the conclusion. No raw commander transcript or inventory table.
R6
Card is the product
Y
N
Your object is headline + photo + compressed lead. Controlling process quotes and munitions identifiers sit below the fold or off the card.
R7
No path to a primary record
Y
Y
Shortlink only on the card. Body gives weapons names and one quantity (“fewer than 100” Patriots in one theater by late spring) but no dossier the reader can open.
Card flags: 6. Default band: Do Not Use as Record. Extract dates and names only.
Article flags: 5. Same default unless primaries independently carry the facts. The Pentagon statement independently carries access, not the assessments.Cotobuzz
2. Layer A — Article Integrity Score (body)
Weights inside each domain are the WWS 1.0 weights.
Domain 1 — Structural Architecture & Semantic Sincerity
Sub
Score
Note
1.1 Definition Bounds (30%)
48
Headline/dek depend on evaluative words. “Falsely claim” is a completed verdict, not a bounded comparison of two quoted statements.
1.2 Complexity (35%)
72
Chronology is usable: spring–summer calls, topics, stockpiles, then public contrast. Extra moral/campaign frames, not a maze.
1.3 Scope creep (35%)
50
Specific calls and one theater’s interceptor count are asked to prove a strategic and political era.
Domain 1 ≈ 57
Domain 2 — Information Routing & Middleman Insulation
Sub
Score
Note
2.1 Sourcing (40%)
40
Remove the unnamed officials and the story of what was said on the calls collapses. Named on-record material covers process only (Hegseth facilitated access; team is “laser focused”). That is the 40–69 band, low end, because the decisive claims are anonymous.
2.2 Middleman insulation (40%)
35
Interpretation arrives before any inspectable record. Commanders are an unseen chorus.
2.3 Uncertainty path (20%)
38
Some “according to.” No update hook. Contested public claims written as settled falsehoods.
Domain 2 ≈ 38
Domain 3 — Metric Verification & Accountability Controls
Sub
Score
Note
3.1 Primary dossier (40%)
45
Identifiers only: Patriots, ATACMS (190 miles), PrSM, “one theater,” “late spring,” “fewer than 100.” No linked inventory, transcript, or notice. Classified stockpiles cannot be treated as a public dossier the reader can open.
3.2 Symmetric scrutiny (40%)
42
Public Trump/Vance lines are framed as false; private anonymous lines are framed as reality. Parnell/Cheung are printed, which is better than the card, but they are not steel-manned on munitions or objectives. Trump’s prior public munitions comments exist in the same season’s record and are not given as a competing operational definition.
3.3 Update trigger (20%)
25
Live war, live talks, no next-fact stamp (inventory audit, after-action, named commander confirmation).
Domain 3 ≈ 40
Raw AIS = (57 + 38 + 40) / 3 ≈ 45
Hard gates (Usable First Look, all four required):
Named actor or linked primary on every decisive claim — fail
Adverse stated ground on the page — partial only (process, not the munitions/victory claim)
No undefined epithet doing causal work in headline/lead — fail
Same-day public competing account not omitted — weak / fail
Fails First Look gates. Archive gates are not in play. Cap would be 69; raw score is already below 50.
Final AIS (article): 45 — Do Not Use as Record
Final AIS (card): ~32 — Do Not Use as Record
The card is worse because it strips the only on-record institutional sentences.
Routing meaning, per the rubric: reject the package. Do not quote its frame, labels, or implied causality as the file. If the event matters, start from the primary record as if the write-up did not exist. That is not a finding that the calls did not happen.Cotobuzz
3. Layer B — Platform & Agency Score (card + house wrap)
PAS is a warning label. It is not averaged into AIS.
Sub
Score
Note
4.1 Engagement packaging (40%)
52
Ordinary-to-loaded news card: “TIMES EXCLUSIVE,” cropped face, contrast dek. Not a moral advertisement, not a flat factual sentence.
4.2 House-narrative autonomy (30%)
42
Swan/Haberman/Entous continuity piece on private-vs-public Trump-war messaging. Fits an existing series. A version in which the calls confirmed the public line is unlikely to ship as this exclusive.
4.3 Reader challenge path (30%)
75
NYT maintains a real corrections desk. The Facebook/X card does not. Score the outlet, flag the card.
PAS ≈ 56
Warning (40–59): House frame or engagement card is doing real work. Open the links first. Do not let the card substitute for the body, and do not let the body substitute for records.
4. Printout
AIS 45 / PAS 56
Band: Do Not Use as Record
What’s missing (plain list):
Named commanders, dates of calls, theaters
Any sentence a named officer will own about Patriots, ATACMS, or PrSM
The actual Trump and Vance public quotes next to the private paraphrase, not the word “falsely”
A definition of “won,” “war,” and “no problems with supplies”
An inspectable stockpile or readiness product
An update trigger (what fact would change the piece)
Routing sentence: File Parnell’s statement that Hegseth “facilitated direct access to combatant commanders,” Cheung’s coordination line, Trump’s public munitions and “not a war / small potatoes” remarks from the same season, and Vance’s public downplay. Treat the content of the calls, the “fewer than 100 Patriots,” and the “gulf” as unverified until a named officer, transcript, or unclassified inventory carries them. The access is on the record. The unvarnished views are not. NYTimes
5. Separation the standard requires
WWS is explicit: a weak package is not proof the event is false.
Can be extracted without adopting the frame
Vance is reported to have called combatant commanders in the Middle East, Europe, and Asia in spring–summer 2026.
Pentagon spokesman Sean Parnell said the secretary facilitated that access for the vice president and other senior White House leaders.
White House communications director Steven Cheung issued a coordination statement.
The piece names Trump, Hegseth, Gen. Dan Caine, Rubio, and Susie Wiles as knowing of the calls, still via anonymous “officials with direct knowledge.”
Public record in the same window includes Trump rejecting munitions-shortage claims and later calling the conflict “small potatoes,” and Vance treating it as not a “war.”
Cannot be filed from this package
That commanders gave “sobering” assessments that “did not line up” with public statements
Theater interceptor counts and “lowest they had ever seen”
Vance’s private “deep concerns” and the implied charge that he and Trump were saying one thing while knowing another
Causal story that private military truth was suppressed by a public victory narrative
That is the Working Standard doing what it says it does: route the reader, archive later, and refuse to let a megaphone card become the dossier.
WWS audit of the NYT social card, with the underlying article scored as a separate unit. The standard treats those as different objects when they diverge.
Cotobuzz
Outlet / headline / date / URL: The New York Times — “Behind the Scenes, Vance Gathered Unvarnished Views of the Iran War,” Jonathan Swan, Adam Entous, Maggie Haberman, 10 Sept 2026. Card points at https://nyti.ms/4iqxgPP / article URL https://www.nytimes.com/2026/09/10/us/politics/vance-iran-war-trump-munitions.html.
Units scored: social card (what you attached) and article body (visible text plus contemporaneous excerpts). Full body is paywalled; scores use the inspectable text, not imagined paragraphs.
Genre: national-security exclusive / live politics investigation. Per WWS, this class usually lands 50–79 and cannot enter Archive Grade on a card.
1. Five-minute red-flag scan
ID
Flag
Card
Article body
Trigger
R1
Undefined load-bearing label
Y
Y
“Unvarnished,” “sobering,” “unusual step,” dek contrast “did not line up.” Body adds the verdict “falsely claim.” No metric for “won,” “demolished,” or “war.”
R2
Anonymous insulation on the decisive fact
Y
Y
Spine is “current and former officials” / “people who spoke to him” / “officials with direct knowledge.”
R3
Missing adverse ground
Y
N (partial)
Card has none. Body includes on-record process statements from Sean Parnell (Pentagon) and Steven Cheung (White House). Those address access and teamwork, not the stockpile or “victory” dispute.
R4
Scope leap
weak
Y
Months of calls become a gulf between private reality and public speech, plus a 2028-campaign shadow in the body.
R5
Middleman supplies the moral
Y
Y
Unnamed officials and the reporter voice deliver the conclusion. No raw commander transcript or inventory table.
R6
Card is the product
Y
N
Your object is headline + photo + compressed lead. Controlling process quotes and munitions identifiers sit below the fold or off the card.
R7
No path to a primary record
Y
Y
Shortlink only on the card. Body gives weapons names and one quantity (“fewer than 100” Patriots in one theater by late spring) but no dossier the reader can open.
Card flags: 6. Default band: Do Not Use as Record. Extract dates and names only.
Article flags: 5. Same default unless primaries independently carry the facts. The Pentagon statement independently carries access, not the assessments.
Cotobuzz
2. Layer A — Article Integrity Score (body)
Weights inside each domain are the WWS 1.0 weights.
Domain 1 — Structural Architecture & Semantic Sincerity
Sub
Score
Note
1.1 Definition Bounds (30%)
48
Headline/dek depend on evaluative words. “Falsely claim” is a completed verdict, not a bounded comparison of two quoted statements.
1.2 Complexity (35%)
72
Chronology is usable: spring–summer calls, topics, stockpiles, then public contrast. Extra moral/campaign frames, not a maze.
1.3 Scope creep (35%)
50
Specific calls and one theater’s interceptor count are asked to prove a strategic and political era.
Domain 1 ≈ 57
Domain 2 — Information Routing & Middleman Insulation
Sub
Score
Note
2.1 Sourcing (40%)
40
Remove the unnamed officials and the story of what was said on the calls collapses. Named on-record material covers process only (Hegseth facilitated access; team is “laser focused”). That is the 40–69 band, low end, because the decisive claims are anonymous.
2.2 Middleman insulation (40%)
35
Interpretation arrives before any inspectable record. Commanders are an unseen chorus.
2.3 Uncertainty path (20%)
38
Some “according to.” No update hook. Contested public claims written as settled falsehoods.
Domain 2 ≈ 38
Domain 3 — Metric Verification & Accountability Controls
Sub
Score
Note
3.1 Primary dossier (40%)
45
Identifiers only: Patriots, ATACMS (190 miles), PrSM, “one theater,” “late spring,” “fewer than 100.” No linked inventory, transcript, or notice. Classified stockpiles cannot be treated as a public dossier the reader can open.
3.2 Symmetric scrutiny (40%)
42
Public Trump/Vance lines are framed as false; private anonymous lines are framed as reality. Parnell/Cheung are printed, which is better than the card, but they are not steel-manned on munitions or objectives. Trump’s prior public munitions comments exist in the same season’s record and are not given as a competing operational definition.
3.3 Update trigger (20%)
25
Live war, live talks, no next-fact stamp (inventory audit, after-action, named commander confirmation).
Domain 3 ≈ 40
Raw AIS = (57 + 38 + 40) / 3 ≈ 45
Hard gates (Usable First Look, all four required):
Named actor or linked primary on every decisive claim — fail
Adverse stated ground on the page — partial only (process, not the munitions/victory claim)
No undefined epithet doing causal work in headline/lead — fail
Same-day public competing account not omitted — weak / fail
Fails First Look gates. Archive gates are not in play. Cap would be 69; raw score is already below 50.
Final AIS (article): 45 — Do Not Use as Record
Final AIS (card): ~32 — Do Not Use as Record
The card is worse because it strips the only on-record institutional sentences.
Routing meaning, per the rubric: reject the package. Do not quote its frame, labels, or implied causality as the file. If the event matters, start from the primary record as if the write-up did not exist. That is not a finding that the calls did not happen.
Cotobuzz
3. Layer B — Platform & Agency Score (card + house wrap)
PAS is a warning label. It is not averaged into AIS.
Sub
Score
Note
4.1 Engagement packaging (40%)
52
Ordinary-to-loaded news card: “TIMES EXCLUSIVE,” cropped face, contrast dek. Not a moral advertisement, not a flat factual sentence.
4.2 House-narrative autonomy (30%)
42
Swan/Haberman/Entous continuity piece on private-vs-public Trump-war messaging. Fits an existing series. A version in which the calls confirmed the public line is unlikely to ship as this exclusive.
4.3 Reader challenge path (30%)
75
NYT maintains a real corrections desk. The Facebook/X card does not. Score the outlet, flag the card.
PAS ≈ 56
Warning (40–59): House frame or engagement card is doing real work. Open the links first. Do not let the card substitute for the body, and do not let the body substitute for records.
4. Printout
AIS 45 / PAS 56
Band: Do Not Use as Record
What’s missing (plain list):
Named commanders, dates of calls, theaters
Any sentence a named officer will own about Patriots, ATACMS, or PrSM
The actual Trump and Vance public quotes next to the private paraphrase, not the word “falsely”
A definition of “won,” “war,” and “no problems with supplies”
An inspectable stockpile or readiness product
An update trigger (what fact would change the piece)
Routing sentence: File Parnell’s statement that Hegseth “facilitated direct access to combatant commanders,” Cheung’s coordination line, Trump’s public munitions and “not a war / small potatoes” remarks from the same season, and Vance’s public downplay. Treat the content of the calls, the “fewer than 100 Patriots,” and the “gulf” as unverified until a named officer, transcript, or unclassified inventory carries them. The access is on the record. The unvarnished views are not.
NYTimes
5. Separation the standard requires
WWS is explicit: a weak package is not proof the event is false.
Can be extracted without adopting the frame
Vance is reported to have called combatant commanders in the Middle East, Europe, and Asia in spring–summer 2026.
Pentagon spokesman Sean Parnell said the secretary facilitated that access for the vice president and other senior White House leaders.
White House communications director Steven Cheung issued a coordination statement.
The piece names Trump, Hegseth, Gen. Dan Caine, Rubio, and Susie Wiles as knowing of the calls, still via anonymous “officials with direct knowledge.”
Public record in the same window includes Trump rejecting munitions-shortage claims and later calling the conflict “small potatoes,” and Vance treating it as not a “war.”
Cannot be filed from this package
That commanders gave “sobering” assessments that “did not line up” with public statements
Theater interceptor counts and “lowest they had ever seen”
Vance’s private “deep concerns” and the implied charge that he and Trump were saying one thing while knowing another
Causal story that private military truth was suppressed by a public victory narrative
That is the Working Standard doing what it says it does: route the reader, archive later, and refuse to let a megaphone card become the dossier.
Layer C warning: Dose on the card. Coat on the article. None on the on-record statements.
That is the BCS sticker. It does not change AIS 45. It hardens the wrap warning and blocks Archive Grade on the package.
Run gate (need 2 of 4; all 4 fire)
C1 — Same named person and class in this house file inside a live window: Vance as the war-skeptic, Trump as the public-victory voice, munitions as the hidden constraint. That file was already live in the April war-origin story, the Haberman/Swan book excerpts, and Haberman’s June Meet the Press turn.
Mediaite
C2 — The card assumes the prior character. “Unvarnished” and “did not line up with what he, President Trump and other top officials were saying publicly” only lands if you already know the private-vs-public plot.
C3 — On the card, yes: strip the new call details and the moral still ships. On the body, mixed: the COCOM calls and “fewer than 100 Patriots” are new alleged facts.
C4 — The feed object still works as portrait + verb + old nickname. You do not need Parnell, Cheung, or a missile count.
Coat sentence (required, or it is only a hunch.
This desk already taught that Vance opposed the Iran war, that private assessments were darker than the public victory line, and that the gap costs him with Trump. The September card does not introduce him. It doses a known figure through a new door: he called the commanders himself.
What the warning does, per 1.1
Does not cap AIS.
Does harden What’s missing on the card.
Does block Archive Grade on the consumed package.
A clean article under a Dose card stays at most Usable First Look as a consumed object. This article is not clean; AIS is already 45.
This is not Course. Reprints of one exclusive are not a multi-desk swarm.
Public sticker vs auditor line
Public: What’s missing — named commanders, dated calls, inspectable stockpile, the actual public quotes beside the word “falsely,” Parnell/Cheung not on the card.
Auditor: BCS: Dose on lockup (Vance/Iran-skepticism beat from spring–summer), Coat on article, None on the Pentagon/WH statements.
Full printout stays unblended: AIS 45 / PAS 56 / BCS: Dose on card, Coat on article.
- #WWS_Audit
- #ArticleIntegrityScore
- #AIS_32
- #DoNotUseAsRecord
- #ObjectDivergence
- #SocialCardVsArticle
- #FiveMinuteScan
- #RedFlagAudit
- #DoseOnTheLockup
- #ConfirmationBiasMachine
- #MechanicalBias
- #ConsumptionWarning
- #OldFileUnpaidWork
- #PriorPlotClosure
- #TheCardIsTheProduct
- #WhatsMissing
- #AnonymousInsulation
- #LoadBearingLabels
- #MissingAdverseGround
- #MiddlemanMoral
- #NoPrimaryRecord
- #BelowTheFold
- #JonathanSwana
- #AdamEntous
- #MaggieHaberman
- #VanceIranWar
- #UnvarnishedViews
- #PrivateTruthPublicSpin
- #SeanParnell
- #StevenCheung
No comments:
Post a Comment