Priority checks
Identify the fields needed to reproduce control and interpret the result. Add context or version when the case requires it, and then remove the fields without proven utility.
TOOLS / REGISTERS
Choose a worksheet by theme or keyword. Read its rule, fields and filled fictional excerpt to distinguish observations, unknowns and the next check.

AI-generated illustration
A record represents a dated observation. It does not turn a claim into evidence or replace the original records.
Each excerpt shows three decisive fields, an action and a next check. These fictional data are not a complete record: fill in the other fields too, especially the date, exact source and responsible person. Filters cover titles, rules and examples.
Test a canonical URL, a language variant and an unknown URL separately. Expected pages must serve their content; an unknown address must retain a 404 status.
| Worksheet field | Observation and proposed next step |
|---|---|
| HTTP status / redirect | 200 |
| Declared canonical URL | https://example.invalid/page-A |
| Observed robots directives | noindex |
| Decision / next action | Retain successful access and exclusion from indexing as separate observations. Status 200 does not close this review. |
| Next review | After changing the directive, inspect the final page and separately check any available indexing status. |
Method reference : Google · robots.txt · Accessed on
Use one row per precise claim. Link a document supporting several claims to several rows, retaining the relevant passage separately for each claim.
| Worksheet field | Observation and proposed next step |
|---|---|
| Claim to verify | 90% of all 300 records are said to comply. |
| Observed passage / value | 90 compliant among 100 reviewed records. |
| Independent source | Two copies of the same observation. |
| Decision / next action | Limit the conclusion to the 100 reviewed records; the other 200 remain unknown. Copies are not additional collection. |
| Next review | Obtain the selection rule and a distinct observation covering unreviewed records. |
Method reference : W3C · PROV primer · Accessed on
Count citations only within the observed panel. Retain the number of answers, answers without citations and repeated runs before calculating any proportion.
| Worksheet field | Observation and proposed next step |
|---|---|
| Exact query | "ORG-A" |
| Interface / mode / account | Interface A; empty session; trial T1. |
| Observed passage / value | A reference is shown, but no passage supports the stated function. |
| Decision / next action | Separate citation presence from support for the claim. Do not count this as a verified answer. |
| Next review | Read the final destination, retain the passage and repeat the question under documented conditions. |
Method reference : W3C · PROV primer · Accessed on
Retain the exact crawler name and tested path. The published rule and observed HTTP response are separate evidence; check both after every change.
| Worksheet field | Observation and proposed next step |
|---|---|
| Exact URL | https://example.invalid/private-A |
| Observed robots directives | User-agent: Robot-A Disallow: /private-A |
| HTTP status / redirect | 200 |
| Decision / next action | The fictional crawler receives a disallow rule, but the resource responds without authentication. Do not treat the rules file as access protection. |
| Next review | Test server-side protection with an unauthenticated client, then inspect the relevant crawler rules. |
Method reference : Google · robots.txt · Accessed on
Match identifiers before merging two records. A shared name, address or logo is insufficient; retain aliases and the reason for the match.
| Worksheet field | Observation and proposed next step |
|---|---|
| Exact name | ORG-A |
| Entity identifier | ENTITY-FICT-17 |
| Country / jurisdiction | FR |
| Decision / next action | Retain this entry as an identified candidate in the case. Another result with the same name must not be merged without matching its attributes. |
| Next review | Compare identifier, jurisdiction, relevant address and period before combining records. |
Method reference : W3C · PROV primer · Accessed on
Check every page–language pair, including menus, forms, errors and outgoing links. Correct hreflang markup does not guarantee a translated journey.
| Worksheet field | Observation and proposed next step |
|---|---|
| Variant URL | https://example.invalid/es/A |
| Language / region | en |
| Reciprocal hreflang | es → https://example.invalid/en/A |
| Decision / next action | The expected Spanish variant serves English and points to another language. Keep both defects separate rather than marking the translation checked. |
| Next review | Check text, declared language and reciprocal links after correcting the Spanish version. |
Test access withdrawal or unavailability in a controlled scope. Document what stops, what continues and recoverable data before calling a fallback operational.
| Worksheet field | Observation and proposed next step |
|---|---|
| Component / provider | API-FICT-A |
| Data / permissions | Read and write. |
| Tested fallback | The replacement is documented, but its access does not work. |
| Decision / next action | The fallback is not operational yet. Retain the critical dependency and restrict permissions to the authorised task. |
| Next review | Run a limited test using replacement access; record elapsed time and unavailable functions. |
Method reference : NIST · AI RMF · Accessed on
Compare declared properties with visible content and source records. Valid JSON can still be false; technical validation does not replace editorial verification.
| Worksheet field | Observation and proposed next step |
|---|---|
| Document type | Event |
| Claim to verify | startDate = 2026-10-05 |
| Observed result | 2026-10-06 |
| Decision / next action | Markup and the visible date of the same event disagree. Valid syntax does not validate the stated fact. |
| Next review | Check the date against its source, then inspect visible content and final markup together. |
Method reference : Google · Structured data guidelines · Accessed on
Define success before testing. Include a normal case, ambiguous data and an expected refusal; record produced effects, not only the quality of the final answer.
| Worksheet field | Observation and proposed next step |
|---|---|
| Data / permissions | Read only. |
| Expected result | Propose a change without executing it. |
| Observed result | The agent attempts a write; the tool rejects it. |
| Decision / next action | No change does not mean the test passed: the attempted action exceeds the read-only mandate. |
| Next review | Repeat after correcting boundaries; retain the proposal, rejected call and final state. |
Method reference : NIST · AI RMF · Accessed on
Trigger review when a source changes, a link fails or a decision becomes sensitive. A new page date should correspond to an actual content change.
| Worksheet field | Observation and proposed next step |
|---|---|
| Exact URL | https://example.invalid/notice-A |
| Claim to verify | The page repeats the deadline from an older notice. |
| Version / revision | A newer notice exists; the applicable date has not been matched. |
| Decision / next action | Mark the passage for review without guessing a replacement date. Retain previous wording and the review reason. |
| Next review | Check the relevant notice, effective date and language versions before approving the correction. |
Method reference : W3C · PROV primer · Accessed on
Reproduce the task with a keyboard and assistive technology. Record the exact element, relevant criterion and obstacle; an automated tool alone cannot establish conformance.
| Worksheet field | Observation and proposed next step |
|---|---|
| Task / journey step | Submit the form using a keyboard. |
| Acceptance criterion | WCAG 2.2 · 2.1.1 |
| Observed result | The final button never receives focus. |
| Decision / next action | The tested journey remains blocked for keyboard use even if other fields work. Locate the control and reproduction path. |
| Next review | Repeat the whole journey with a keyboard after correction, including validation and error feedback. |
Method reference : W3C · Keyboard · Accessed on
Separate lab measurements from field data. Compare the same page, device and comparable windows; retain the available percentile and sample size.
| Worksheet field | Observation and proposed next step |
|---|---|
| Metric / measurement | INP |
| Sample / population | 20 desktop visits. |
| Observed result | No mobile visits observed. |
| Decision / next action | Describe only the desktop scope. This record cannot establish verified mobile responsiveness. |
| Next review | Collect mobile visits separately and document period, counts and chosen percentile. |
Method reference : web.dev · Web Vitals · Accessed on
Test success, input error and server failure with identified test data. Check that announced messages match actual outcomes and entered data remains recoverable.
| Worksheet field | Observation and proposed next step |
|---|---|
| Task / journey step | Send a contact message. |
| Expected result | The message remains after an invalid email address. |
| Observed result | The entered message disappears after the error. |
| Decision / next action | Error feedback loses entered data. Retain fictional inputs and exact steps. |
| Next review | Repeat email correction and submission; keep secrets out of traces. |
Method reference : W3C · Form notifications · Accessed on
Identify the media role before writing its alternative text: evidence, illustration or decoration. Link evidence to its source; clearly label an illustrated scene without assigning it a real location.
| Worksheet field | Observation and proposed next step |
|---|---|
| Alternative text / role | View of a river. |
| Licence / reuse rights | No permission identified. |
| Claim to verify | The image supposedly proves the event location. |
| Decision / next action | Hold publication: neither rights nor the claimed location are established by this description. |
| Next review | Check author, permission, context and location clues separately; then adapt the alternative to its use. |
Method reference : Creative Commons · Licenses · Accessed on
Trace repetitions back to the original producer. Two pages copying the same document are not two independent confirmations; record the provenance chain.
| Worksheet field | Observation and proposed next step |
|---|---|
| Producer / publisher | Three distinct websites. |
| Observed passage / value | The same figure and transcription error. |
| Independent source | All three publications point to document D. |
| Decision / next action | Retain one shared origin; three publications do not become three independent confirmations. |
| Next review | Seek distinct collection and compare the exact fact, date and scope. |
Method reference : W3C · PROV primer · Accessed on
Distinguish event, publication and observation dates. A later correction creates a new row version; it should not erase what could be observed previously.
| Worksheet field | Observation and proposed next step |
|---|---|
| Event, publication and observation dates | Event: 3 March; publication: 5 March; capture: 10 March 2026. |
| Version / revision | T1 |
| Limit / uncertainty | The event time is undocumented. |
| Decision / next action | Keep three distinct dates with day-level precision. Capture alone does not date the event. |
| Next review | Find the source of the event date before using a finer ordering. |
Method reference : W3C · PROV primer · Accessed on
Document each link with direction, period and a specific record. Participation, funding and control are different relationships; visual proximity proves none of them.
| Worksheet field | Observation and proposed next step |
|---|---|
| Entity identifier | ORG-A / ORG-B |
| Relationship / link direction | A supposedly controls B. |
| Observed passage / value | Two records show the same address. |
| Decision / next action | Retain a shared attribute without drawing an established control relationship. |
| Next review | Seek a document about the relationship type and period, not merely the address. |
Method reference : W3C · PROV primer · Accessed on
Keep the received image and discovered versions separately. Reverse search finds occurrences; the earliest found does not prove the first publication or authorship.
| Worksheet field | Observation and proposed next step |
|---|---|
| Capture / acquisition date | Capture date unknown. |
| Place / coordinates | One candidate location inferred from one building. |
| Limit / uncertainty | No independent landscape comparison. |
| Decision / next action | Retain the location as a hypothesis; do not date the scene from its publication date. |
| Next review | Compare several spatial clues and find a dated version with its context. |
Method reference : W3C · PROV primer · Accessed on
Cite the operative part or paragraph supporting the precise claim. Check the version and mentioned appeals; a decision title does not summarise its scope.
| Worksheet field | Observation and proposed next step |
|---|---|
| Document identifier | DOC-FICT-19 |
| Document type | Advocate General’s Opinion. |
| Effective / commencement date | No corresponding judgment examined. |
| Decision / next action | Do not present the proposed solution as a delivered judgment or its operative part. |
| Next review | Find the case status and any judgment, then distinguish the documents in the account. |
Method reference : CURIA · About the Court of Justice · Accessed on
Define population, units and deduplication rules before counting. A missing value is not zero; retain the used extract to make the calculation reproducible.
| Worksheet field | Observation and proposed next step |
|---|---|
| Coverage / exclusions | Two areas only, year 2025. |
| Unit / calculation basis | Households. |
| Method / definition | The summary refers to people across the whole territory. |
| Decision / next action | Hold that generalisation: unit and coverage do not match the stated conclusion. |
| Next review | Obtain definitions, full scope and any method for converting units. |
Method reference : W3C · DCAT 3 · Accessed on
RNA_waldec contains RNA-identified associations created or amended since 2009; RNA_import contains older associations unchanged since then. Record resource, release and filter.
| Worksheet field | Observation and proposed next step |
|---|---|
| RNA identifier | Not recorded. |
| SIREN | Not recorded. |
| Notice number | Only a notice of a name change. |
| Decision / next action | The notice cannot link all records or date formation by itself. Keep identity matching open. |
| Next review | Find relevant identifiers and the initial notice within accessible coverage. |
Method reference : Service Public · Déclaration d’association · Accessed on
Keep response, country, number and timestamp; include a consultation reference if supplied. Record any retry separately.
| Worksheet field | Observation and proposed next step |
|---|---|
| VAT number | VAT-FICT-A |
| Country / jurisdiction | FR |
| Observed status | Verification service unavailable. |
| Decision / next action | Retain an unobtained result. Unavailability means neither a valid nor an invalid number. |
| Next review | Retry the service and retain date, response and exact context without turning failure into tax status. |
Method reference : Your Europe · VIES · Accessed on
Open the official register yourself and use a contact independent of the received message. Read AMF warnings; absence from a blacklist is not authorisation.
| Worksheet field | Observation and proposed next step |
|---|---|
| Entity identifier | REG-FICT-A |
| Exact name | ORG-A |
| Observed status | An entry is found; the authorised activity has not been read. |
| Decision / next action | Do not extend registry presence to every offered service. Match exact identity and permission scope. |
| Next review | Read status, covered services and registry contact details for the offer under review. |
Method reference : AMF · Vérifier un intermédiaire · Accessed on
Locate protocol, posted results and article separately. Link the article through the trial ID and check population, intervention and period before attributing its results.
| Worksheet field | Observation and proposed next step |
|---|---|
| Observed status | Recruitment completed. |
| Sample / population | 80 planned participants; actual enrolment not recorded. |
| Acceptance criterion | Primary outcome not examined. |
| Decision / next action | Do not infer efficacy or a favourable result from completed recruitment. |
| Next review | Read version, actual enrolment, available results and exact status before summarising the study. |
Method reference : WHO · Trial registration data set · Accessed on
Check negations, obligation versus possibility, dates, numbers and scope. Keep the original term where translations change interpretation. Mark omissions and additions.
| Worksheet field | Observation and proposed next step |
|---|---|
| Language / region | en |
| Observed passage / value | may |
| Quoted translation and terminology choices | must |
| Decision / next action | Changing possibility into obligation changes the meaning. Retain the original term and flag the translation for review. |
| Next review | Read the full sentence and validate terminology in this context. |
Method reference : W3C · PROV primer · Accessed on
Compare within a homogeneous segment or use a producer-published backcast series. Do not splice segments with an improvised factor; explain excluded periods.
| Worksheet field | Observation and proposed next step |
|---|---|
| Dataset / series code | SERIES-FICT-A |
| Reference period | 2020–2024 |
| Method / definition | Population coverage changes in 2022. |
| Decision / next action | Do not present a homogeneous five-year change without addressing the break. |
| Next review | Obtain the method note and seek a backcast series or comparable segments. |
Method reference : SDMX · Glossary · Accessed on
Test uniqueness, domains and required fields on the real file. Empty string, not applicable and zero differ. Date changes to definitions, codes or types.
| Worksheet field | Observation and proposed next step |
|---|---|
| Exact column name | record_id |
| Logical type and format | Text; leading zeros are meaningful. |
| Expected key and uniqueness | Uniqueness observed in T1 only. |
| Decision / next action | Preserve text type and limit observed uniqueness to the reviewed version. |
| Next review | Test missing values and duplicates in T2 before approving its key. |
Method reference : W3C · Tabular data model · Accessed on
Prepare visible attribution, licence link and change notice when required. Check personal data, third-party rights and trademarks separately. A data licence does not settle all these issues.
| Worksheet field | Observation and proposed next step |
|---|---|
| Licence / reuse rights | CC BY 4.0 |
| Planned attribution and licence link | Table attribution prepared. |
| Limit / uncertainty | The page explicitly excludes the third-party photograph from this licence. |
| Decision / next action | Do not apply the table licence to the excluded photograph. |
| Next review | Check rights for each item and retain the exact scope of permission. |
Method reference : Creative Commons · Licenses · Accessed on
Keep original, normalised value, zone, offset and rule version. Sort determined instants separately from calendar dates.
| Worksheet field | Observation and proposed next step |
|---|---|
| Observed passage / value | 03/04/2026 |
| Logical type and format | String without a time. |
| Method / definition | Day/month or month/day order undocumented. |
| Decision / next action | Retain the raw value and undetermined format. Do not select a date or add a UTC time. |
| Next review | Obtain the source format contract and test a value that distinguishes both interpretations. |
Method reference : W3C · Tabular data model · Accessed on
Publish available counts alongside the result. Retain estimated and provisional flags; show a missing chart point differently from zero.
| Worksheet field | Observation and proposed next step |
|---|---|
| Missing values and nullability | .. |
| Observed status | Confidential value according to this case’s dictionary. |
| Sample / population | 10 observations expected; this cell’s value is unknown. |
| Decision / next action | Do not replace the marker with zero or silently remove its row from the denominator. |
| Next review | Read producer rules and document treatment of suppressed values before calculation. |
Method reference : SDMX · Glossary · Accessed on
Check duplicates, unmatched units and level. Never add an aggregate and its members together; maps also need coordinate system and layer date.
| Worksheet field | Observation and proposed next step |
|---|---|
| Dataset / series code | AREA-FICT-A |
| Reference period | 2020 / 2024 |
| Version / revision | Boundaries change in 2022 without a code change in this case. |
| Decision / next action | The same code does not establish constant geography. Hold direct comparison. |
| Next review | Find boundaries and correspondence tables, then document the method for comparable geography. |
Method reference : SDMX · Glossary · Accessed on
For time comparisons state current or constant prices and justify the relevant index. Consumer price indices do not automatically suit every contract.
| Worksheet field | Observation and proposed next step |
|---|---|
| Unit / calculation basis | Same currency, each year’s current prices. |
| Reference period | 2020 / 2025 |
| Method / definition | No price index or reference year selected. |
| Decision / next action | Describe a nominal difference only; do not call it a volume or constant-price change. |
| Next review | Choose a relevant index and reference year; retain raw amounts and the formula. |
Method reference : SDMX · Glossary · Accessed on
Record full request, filters, period, Accept header, download time and raw file. Compare keys before numbers.
| Worksheet field | Observation and proposed next step |
|---|---|
| Dataset / series code | SERIES-FICT-A |
| Dimensions / filters | FREQ=A; UNIT=PERSONS |
| Observed status | P means provisional in this case’s dictionary. |
| Decision / next action | Retain the status with the value and version. Do not present the data as final. |
| Next review | Check this series’ structure and code list; compare versions when revised. |
Method reference : SDMX · Glossary · Accessed on
Link file, producer, extract and transformations. Retain checksum and parameters; identify who performed cleaning.
| Worksheet field | Observation and proposed next step |
|---|---|
| Document identifier | CAT-FICT-A |
| Logical type and format | CSV / API |
| Licence / reuse rights | Catalogue licence shown; distribution rights unread. |
| Decision / next action | Distinguish the record from its distributions. A link does not validate the served format or reuse rights. |
| Next review | Examine each useful distribution, current response, version and applicable licence. |
Method reference : W3C · DCAT 3 · Accessed on
Compare extent, date, layer and description. Near the antimeridian, avoid conclusions based on simple longitude differences.
| Worksheet field | Observation and proposed next step |
|---|---|
| Observed passage / value | 120 ; 45 |
| Unit / calculation basis | Degrees. |
| Method / definition | Headers: latitude, longitude. |
| Decision / next action | Latitude 120 is outside the possible range. Flag the issue without automatically swapping axes. |
| Next review | Check producer convention and reference system; retain raw values and the chosen transformation. |
Method reference : IETF · GeoJSON · Accessed on
Open the reference registry or record and compare title, author, edition and version. Link distinct objects rather than removing different identifiers.
| Worksheet field | Observation and proposed next step |
|---|---|
| Document identifier | IDENT-FICT-A |
| Document type | Journal identifier; value not reproduced in this excerpt. |
| Expected key and uniqueness | Structure check reported successful; registry record unread. |
| Decision / next action | A successful technical check does not establish the associated document’s existence or identity. |
| Next review | Open the relevant registry record and match title, medium and version. |
Method reference : W3C · PROV primer · Accessed on
Reviewing a small corpus certifies neither all model behaviour nor facts beyond the documents. Do not send confidential records to a third-party tool without an appropriate access framework.
| Worksheet field | Observation and proposed next step |
|---|---|
| Claim to verify | A supplies a service to B. |
| Interface / mode / account | Assistant-FICT-A |
| Observed passage / value | The passage says only that A cites B. |
| Decision / next action | Remove the supplier relationship: citation does not prove a contract or supply. |
| Next review | Seek a document supporting this exact relationship, period and parties. |
Method reference : W3C · PROV primer · Accessed on
A point-in-time check guarantees neither future availability nor indexing. Transport failure before a response has no received HTTP status. Do not bypass protections or use confidential access to obtain a green indicator.
| Worksheet field | Observation and proposed next step |
|---|---|
| HTTP status / redirect | 202 |
| Final URL after redirects | https://example.invalid/status/1 |
| Observed result | Processing still pending. |
| Decision / next action | Do not declare completion from the acknowledgement alone. |
| Next review | Use the documented status route and check the expected final state before closing the observation. |
Method reference : IETF · HTTP semantics · Accessed on
No universal sample size fits every audit. An interval cannot repair biased selection or ill-defined populations. Sampling cannot establish that every page was tested.
| Worksheet field | Observation and proposed next step |
|---|---|
| Selection unit and case identifier | One answer session. |
| Frame, mechanism and selection order | Three convenience-selected questions. |
| Repetitions and dependence between observations | Five sessions per question: 15 observations. |
| Decision / next action | Distinguish 15 sessions from three questions. Do not report 15 independent questions or general representativeness. |
| Next review | Define the target population and selection plan before generalising. |
Method reference : NIST · AI RMF · Accessed on
Totals may be estimates or change during collection. A next link does not override quotas or documented access conditions.
| Worksheet field | Observation and proposed next step |
|---|---|
| Received continuation / cursor | CURSOR-C → CURSOR-C |
| Received record count | 20 |
| Distinct received identifiers | 10 |
| Decision / next action | Stop the cursor loop and retain received pages. Twenty rows represent only ten distinct identifiers here. |
| Next review | Check continuation rules and end behaviour before declaring collection complete. |
Method reference : IETF · HTTP semantics · Accessed on
Negative or near-zero bases can make relative changes misleading. Explain the indicator’s meaning. Arithmetic does not repair a biased population.
| Worksheet field | Observation and proposed next step |
|---|---|
| Initial and final values / denominators | 2 % → 3 % |
| Change in percentage points | +1 percentage point. |
| Relative change and formula | 100 × (3 − 2) / 2 = 50 % |
| Decision / next action | Report +1 percentage point and +50% with their bases. A 1% increase would mean something different. |
| Next review | Check comparable periods, populations and rate definitions before publishing. |
Method reference : Insee · Taux de croissance · Accessed on
Reconciling opening stock, flows and closing stock requires consistent scope and adjustments. A chart alone cannot reconstruct the series.
| Worksheet field | Observation and proposed next step |
|---|---|
| Stock reference date | 2026-01-31: 100; 2026-02-28: 110 |
| Flow interval / cumulative origin | February: 20 openings and 15 closures. |
| Method / definition | 100 + 20 − 15 = 105 ≠ 110 |
| Decision / next action | Reconciliation leaves a gap of 5. Do not rewrite a stock to force equality. |
| Next review | Seek other movements, corrections or scope differences and retain the unexplained gap. |
Method reference : Eurostat · Stocks and flows · Accessed on
Software and settings vary. A rule tested in one spreadsheet does not guarantee every other. Keep viewing transformations distinct from original material.
| Worksheet field | Observation and proposed next step |
|---|---|
| Exact column name | code |
| Expected result | 00042 |
| Observed result | 42 |
| Decision / next action | The code lost leading zeros. Do not guess padding length if the field contract is missing. |
| Next review | Restore the source value and check text type and the reopening workflow. |
Method reference : W3C · Tabular data model · Accessed on
Accessibility does not guarantee freshness or completeness. Cache and timestamp differences can delay arrival. Missing feed items do not prove non-publication.
| Worksheet field | Observation and proposed next step |
|---|---|
| Dataset / series code | urn:example:item:7 |
| Publication date | 2026-03-01 |
| Version / revision | updated: 2026-03-04 |
| Decision / next action | Retain the stable identifier and distinguish initial publication from update. Changed content is not necessarily a new entry. |
| Next review | Compare identifier and content version with the previous observation before counting new items. |
Method reference : IETF · Atom · Accessed on
Local timing depends on task, load, network and tools. Percentiles from few trials are fragile. Controlled trials are not general availability promises.
| Worksheet field | Observation and proposed next step |
|---|---|
| Stage durations and unit | Median: 3 s across 8 successful trials. |
| Failures, timeouts and retries | 2 timeouts at 30 s excluded from this median. |
| Repetitions and dependence between observations | 10 |
| Decision / next action | Report the conditional median with both failures. Do not claim 3 s across all trials. |
| Next review | Retain start, end, status and duration for every trial; compare matching scopes. |
Method reference : NIST · AI RMF · Accessed on
A control query shows observed operation, not full coverage. Missing indexing, capture or accessible publication does not establish absence of the fact.
| Worksheet field | Observation and proposed next step |
|---|---|
| Exact query, variants and filters | "ORG-A" |
| Sample / population | Registry X, version T1; exact name only. |
| HTTP status / redirect | 200 |
| Decision / next action | Record no result within this scope without concluding that the entity does not exist. |
| Next review | Test variants and identifiers, then recheck registry coverage and period. |
Method reference : W3C · PROV primer · Accessed on
A log does not certify an entire page. Corrections must not expose confidential information. Public history can be summarised while an appropriate internal trace is retained.
| Worksheet field | Observation and proposed next step |
|---|---|
| Previous wording / value | 100 |
| Corrected wording / value | 80 |
| Reason and supporting document | 20 duplicates removed from the count; record D retained. |
| Decision / next action | Record the correction and supporting evidence rather than silently replacing the figure. |
| Next review | Check conclusions, derived calculations and translations using the former value. |
Method reference : W3C · PROV primer · Accessed on
Two uncompressed UTF-8 XML files, at most 5 MiB and 20,000 entries each. No page or child-sitemap download, no image/video extension comparison and no complete protocol validation. Invalid loc values are reported and excluded from matching. The export contains complete URLs.
| Worksheet field | Observation and proposed next step |
|---|---|
| Before/after files and scope | T1 / T2 · fr / es |
| Added/removed URLs and changed metadata | − /fr/A ; + /es/A |
| Duplicates and excluded entries | The new Spanish URL still serves French content. |
| Decision / next action | Separate list changes from the language defect. Sitemap presence does not validate the destination. |
| Next review | Check final response, language and canonical for every addition before closing the migration. |
No worksheet matches these criteria.
Choose the protocol that fits your question. For each check, record what was observed and within which scope. A positive response is a documented observation, not certification of a site or source.
Document this reviewChoose the template matching your question, retain the exact URL and a date with time zone, then separate observation, uncertainty and action.
Start with the review rule, then identify the required information and related method.
The worksheet organises observations; it does not replace their documentary support. Retain the exact source, version, relevant passage and observation limits.
DEEPER READING / PRACTICE
A model offers fields to record a verification. It must be adapted to the question and to the risk, without turning the capture into an independent objective.
Updated on .
Identify the fields needed to reproduce control and interpret the result. Add context or version when the case requires it, and then remove the fields without proven utility.
A complete form may remain unprofitable if the result is not linked to a decision. Unknown information should not be replaced by arbitrary values.
The filled model indicates entrance, observation, reserve and next action. It allows another reader to understand control without repeating the whole search.
PRACTICAL CASE / WORKED EXAMPLE
The following case illustrates an approach to defining the need. Situations are pedagogical: they serve to examine the choices and conditions of success, without describing a fulfilled mission or announcing a guaranteed result.
All fields in a file are informed, but the result does not indicate what has been established or what to be done. The formal completeness masks a useless control.
Review the question, observations and their limitations. Add a qualified conclusion and a next action. Remove irrelevant information and identify the fields needed for another person to continue the work.
The file allows you to understand the control and its effect on the decision. Filled boxes become a proof support rather than an independent quality measurement.
Yes, to adapt it to the question and useful information. Keeping the fields needed for traceability and documenting major changes facilitates comparisons between controls.
DETAILED WORKSHOP / APPLICATION
Choose the relevant question and model. Identify the entries, observations and reservations necessary to understand the outcome of the control.
Complete a first case, then review its conclusion. Add necessary fields, remove irrelevant details and distinguish unknowns from confirmed values. Check whether another person can continue the work.
If the model is complete but does not allow to decide, specify the result and next action. The capture must support control rather than become its purpose.
The model is adapted when its fields make the proof and limits understandable. Important changes remain documented to preserve the comparison between files.
Transmit model, example and conventions. Users can distinguish the required fields for the traceability of the add-ons depending on the context.
ANNOTATED WORKING DOCUMENT / REVIEW GRID
The work document combines entries, observations, limits and next action. The fields are adapted to control and distinguish confirmed, uncertain and unknown values.
On a small screen, scroll the grid horizontally to read all three columns.
| Observation | Supported interpretation | Next step |
|---|---|---|
| Information allows to reproduce and interpret control. | The model supports a qualified conclusion and an explicit follow. | Keep the example and the completion agreements. |
| The list is complete but does not allow you to decide. | The input contains information without a clear connection to the result. | Defining the conclusion, scope and useful action. |
| An unknown is replaced by an arbitrary value. | The formal completeness masks a lack of evidence. | Reset the status and describe the necessary control. |
The final verification searches what another reader can check or do. Adaptations are useful when they improve traceability, not when they multiply fields without demonstrated contribution.
REPORTING / EXAMPLE WORDING
A fully filled file does not guarantee a profitable conclusion.
List the file as a recipient who must reproduce or continue the control.
The information fields support the qualified result and the next action indicated. The unknown and adaptations of the model are kept in the trace.