open/ms-ai-architect
1
0
Fork 0
Code Issues Releases 1 Activity

fix(ms-ai-architect): idx-26j presisert — 80/20-funnet forutsatte en nevner det selv kaller udefinert

Browse source 
To rettelser i bokfoeringen, ingen endring i korpuset.

1. 80/20-PAASTANDEN MOTSA SEG SELV. Entryen sa samtidig at fottekstens referent
er udefinert OG at 80/20 'matcher hverken 10/5 foer eller 12/5 etter' - et
konkret avvik regnet ut fra én ANTATT nevner. Soesterfila teller samme slags
paastand paa en annen partisjon ('15 (8 verified fra MCP, 7 baseline/code
samples)' folder code samples inn paa baseline-siden og teller ikke
listeposisjoner). Omskrevet: defekten er at linja ikke KAN sjekkes, ikke at et
bestemt tall spriker. Ikke 'fiks' prosentene mot en antatt nevner - avgjoer
referenten foerst.

2. 'Status: GA' PAA LINJE 4 ER ADJUDISERT, IKKE OVERSETT. idx-26g advarer neste
leser mot aa triagere denne klassen som sitat-hygiene, og samme streng staar i
fila denne oekten nettopp stemplet. Den er IKKE falsifisert av stempelet: linje 4
ligger i bold-label-headeren over '## Innhold', og filas foerste Confidence-
stempel er seksjons-terminalt paa :59.

MEN det egentlige funnet er stoerre enn begge filene: feltets referent er
udefinert KORPUS-BREDT. Maalt over skills/*/references/ blander '**Status:**' to
vokabularer som svarer paa ulike spoersmaal - dokumentmodenhet (Established
Practice 43, Gjeldende 18, Reference 7, Komplett 2) og produkttilgjengelighet
(GA 252, Preview 2, 'GA - avvikles 31. mars 2029' 2). Korpuset baerer alt
presedensformen for en fil som spenner begge: 'GA / Preview (varies by feature)'
(2 filer). Denne fila spenner Transparency Notes (GA), Model Cards (praksis) og
scorecard-en (public preview) samtidig.

Konsekvens for idx-26g: header-spoersmaalet er ikke fil-lokalt. Aa avgjoere det
én fil om gangen setter konvensjonen for 389 filer ved uhell.

Suite 1047/1047. Koe: 8 aapne / 10 resolved.
This commit is contained in:
Kjell Tore Guttormsen 2026-08-09 10:24:07 +02:00
commit 0470674801
2 changed files with 22 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

24
docs/r11-pilot-results.md
View file

@ -1472,7 +1472,25 @@ framing of a ratification prompt.
**Booked, not fixed:** `idx-26j` — the same file's footer provenance block, which no entry
has ever measured. `Total MCP calls: 5` enumerates components summing to 6; the line has
no defined referent (original generation run, or current state?); the `80% / 20%` split
matches neither the list's 10/5 before this edit nor its 12/5 after; and this edit itself
introduced a two-dialect compound pair (`model-` on line 101 against the ratified
`modell-` in the rewritten row) that no check can see, because the stamp claims
**cannot be checked at all**, which is the defect rather than any particular mismatch —
read as list positions it fits neither 10/5 before nor 12/5 after, but that reading is an
assumption, and the sibling file counts the same kind of claim on a different partition
(*"15 (8 verified fra MCP, 7 baseline/code samples)"*), so no denominator can be asserted.
This edit also introduced a two-dialect compound pair (`model-` on line 101 against the
ratified `modell-` in the rewritten row) that no check can see, because the stamp claims
verification against source, not orthographic consistency.
**One locator adjudicated out of scope rather than missed.** The header carries
`**Status:** GA` at line 4 — the string idx-26g flags in the sibling as a false status
claim of this same class. It is not falsified by this edit's stamp: line 4 sits in the
bold-label header block above `## Innhold`, and the file's first `Confidence` stamp is
section-terminal at line 59, so no stamp reaches it. The real finding is that the field's
referent is undefined **corpus-wide**. Measured across `skills/*/references/`, `**Status:**`
mixes two vocabularies answering different questions — document maturity (`Established
Practice` 43, `Gjeldende` 18, `Reference` 7, `Komplett` 2) and product availability (`GA`
252, `Preview` 2, `GA — avvikles 31. mars 2029` 2) — and the corpus already carries the
precedent form for a file spanning both: `GA / Preview (varies by feature)`, 2 files. This
file spans Transparency Notes (GA), Model Cards (practice) and the scorecard (public
preview) at once. **So idx-26g's header question is not local to either file — it is a
corpus-wide field-semantics decision, and settling it one file at a time would set the
convention by accident.**

2
scripts/kb-eval/data/g7-review-queue.json
View file

@ -233,7 +233,7 @@
"class": "multi-locator",
"status": "open",
"raised": "2026-08-09",
"summary": "Surfaced while closing idx-26h in this file; measured, not inferred. The footer provenance block makes three counting/provenance claims and no G7 entry has ever measured it. First: \"Total MCP calls: 5 (microsoft_docs_search: 3, microsoft_docs_fetch: 2, microsoft_code_sample_search: 1)\" — the enumerated components sum to 6, not 5. The line is internally inconsistent as written, independently of anything this session did. Second: the line has no defined referent. If it records the original generation run it is a historical fact that should not move, and if it describes the file current provenance it is now wrong, because idx-26h added two sources fetched live 2026-08-09. The file does not say which, so no reader and no check can adjudicate it. Third: \"Confidence: 80% Verified (MCP), 20% Baseline\" does not match the list own split — 10 verified / 5 baseline (67/33) before idx-26h, 12/5 (71/29) after. This one is PRE-EXISTING, not created by idx-26h, and it is unclear whether the percentages are meant to describe source counts or content proportion; that ambiguity is itself the defect. Note for the reviewer: this footer block form is likely replicated across the corpus, so the fix should be decided as a form, not as one file. FOURTH LOCATOR, introduced by idx-26h and named here rather than silently repaired: the section now spells the same compound two ways eight lines apart — line 101 \"model- og data-innsikter\" (pre-existing) against the rewritten role row \"modell- og data-innsikter\". The row spelling is the operator-ratified string, so it was NOT quietly aligned; the pair is the idx-26d mixed-dialect class reappearing, and which spelling wins is a form decision for the file, not a typo fix. The stamp is not falsified by it — the stamp claims verification against source, not orthographic consistency — which is exactly why no check will ever catch it.",
"summary": "Surfaced while closing idx-26h in this file; measured, not inferred. The footer provenance block makes three counting/provenance claims and no G7 entry has ever measured it. First: \"Total MCP calls: 5 (microsoft_docs_search: 3, microsoft_docs_fetch: 2, microsoft_code_sample_search: 1)\" — the enumerated components sum to 6, not 5. The line is internally inconsistent as written, independently of anything this session did. Second: the line has no defined referent. If it records the original generation run it is a historical fact that should not move, and if it describes the file current provenance it is now wrong, because idx-26h added two sources fetched live 2026-08-09. The file does not say which, so no reader and no check can adjudicate it. Third: \"Confidence: 80% Verified (MCP), 20% Baseline\" cannot be checked at all, and THAT is the defect rather than any particular mismatch. Read as list positions it is 10/5 before idx-26h and 12/5 after (67/33 and 71/29), matching neither — but that reading is an assumption, and the sibling file counts the same kind of claim differently (\"15 (8 verified fra MCP, 7 baseline/code samples)\" folds code samples into the baseline side and does not count list positions at all). Whether the percentages describe source counts, a different partition, or content proportion is undefined, so no denominator can be asserted. PRE-EXISTING, not created by idx-26h. Do not \"fix\" the percentages against an assumed denominator - decide the referent first. Note for the reviewer: this footer block form is likely replicated across the corpus, so the fix should be decided as a form, not as one file. FOURTH LOCATOR, introduced by idx-26h and named here rather than silently repaired: the section now spells the same compound two ways eight lines apart — line 101 \"model- og data-innsikter\" (pre-existing) against the rewritten role row \"modell- og data-innsikter\". The row spelling is the operator-ratified string, so it was NOT quietly aligned; the pair is the idx-26d mixed-dialect class reappearing, and which spelling wins is a form decision for the file, not a typo fix. The stamp is not falsified by it — the stamp claims verification against source, not orthographic consistency — which is exactly why no check will ever catch it. FIFTH LOCATOR, adjudicated as OUT OF SCOPE here rather than overlooked - recorded because idx-26g warns the next reviewer not to triage this class as citation hygiene. The header carries **Status:** GA at line 4. It is NOT falsified by idx-26h stamp: line 4 sits in the bold-label header block above the --- and ## Innhold, and the first Confidence stamp is at line 59, section-terminal for section 1, so no stamp in this file reaches line 4. But the field referent is undefined CORPUS-WIDE, which is the real finding: measured across skills/*/references/, **Status:** mixes two vocabularies that answer different questions - document maturity (Established Practice 43, Gjeldende 18, Reference 7, Komplett 2) and product availability (GA 252, Preview 2, \"GA - avvikles 31. mars 2029\" 2). This file covers Transparency Notes (GA), Model Cards (practice) and the scorecard (public preview) at once, and the corpus ALREADY carries the precedent form for exactly that case: \"GA / Preview (varies by feature)\", 2 files. So the question idx-26g raises for the sibling is not local to either file - it is a corpus-wide field-semantics decision, and fixing it one file at a time would set the convention by accident.",
"evidence": "skills/ms-ai-governance/references/responsible-ai/transparency-documentation-standards.md footer; idx-26h resolution; docs/r11-pilot-results.md",
"anchors": [
"**Total MCP calls:** 5 (microsoft_docs_search: 3, microsoft_docs_fetch: 2, microsoft_code_sample_search: 1)",