Compare sources

Compare public sources against your question

Two services repeating the same document are not two independent pieces of evidence. Compare the research object, data origin and period first.

Archive consultation with a microfilm reader and historical documents.

AI-generated illustration

READING / ANNOTATED CASE

Three disagreements that need different responses

Before deciding which source is right, check that both claims concern the same object, period and definition. These examples are fictional.

Fictional example · no observation of a service or client.

On small screens, scroll the table horizontally.

Three disagreements that need different responses
Apparent mismatchPossible explanationDecisive check
A: 12 establishments · B: 10 enterprisesDifferent statistical units.Read definitions and identifiers; an enterprise may have several establishments.
A: open in June · B: closed in AugustObservations from different dates.Reconstruct the timeline and find the effective change date.
A: amount 100 · B: amount 120Price, unit, tax or scope may explain the gap.Check currency, date, unit and tax inclusion before comparing.
When is it a contradiction?

When incompatible claims concern the same object and property in a comparable context. If period or definition explains the gap, describe that difference instead of arbitrarily declaring a source wrong.

What if the disagreement remains unexplained?

Keep both wordings, passages, versions and dates. Do not average them to obtain a reassuring number. State the uncertainty and the check that could resolve it: history, producer methodology or another independent observation.

Choose 2 to 4 items to compare

Checked choices remain visible while searching. Remove a choice to select another when the limit of four is reached.

Compare sources
Compare sourcesICANN LookupWayback Machine
Operator or reference institutionICANNInternet Archive
Result typeRegistry recordArchived document
Result unitDomain registration data, not editorial identityCapture of one URL at a specified timestamp
Time reference to retainAccess date and displayed update date; current state does not reconstruct all history.Capture timestamp differs from publication and access dates.
What this source helps you findQuery public domain, IP network and ASN registration data through RDAP.Find an archived web page, compare captures and retain the URL actually opened with its timestamp. Also check images and links needed for the analysis.
Limit to retain in your conclusionSome data may be redacted; registration does not prove who operates a website.A capture can be incomplete or combine resources from different dates. Archiving does not validate page claims; absence from replay does not automatically mean an element did not exist at the time.
Useful search anchors
  • Domain / IP resource / ASN
  • Exact URL
  • Reference period
Fields to record when citing a result
  • Domain / IP resource / ASN
  • Producer / publisher
  • Access date and time zone
  • Observed status
  • Exact URL
  • Capture / acquisition date
  • Coverage / exclusions
Evidence record to retainExact name, dated RDAP response, statuses, queried server and any links from an official site.Original URL, capture URL, capture time and any missing or incomplete page elements.
Check before drawing a conclusionRedacted fields, intermediaries and hosts often prevent identification of the actual operator.An archive shows a retrieved version at a time, not content accuracy or its creation date.
CoverageINTINT
Official reference page

Official reference page

Open the official service

Official reference page

Open the official service
Read this source in practiceDoes domain data identify a website's author?Does a capture date the first publication?
Additional documented context

Edit selection

Access, download and API conditions depend on the service. Consult the official documentation and the resource terms.

Choose and cross-check sources

Assess independence

Identify the producer of the underlying document. A portal, API and search engine may distribute the same data. To confirm a fact, seek a document with a different origin; to check a transcription, compare the original with its reuse.

Compare dates and versions

The consultation date does not replace the effective, publication or acquisition date. A current and an archived result may both be correct for different periods. Record the status and version of each document.

Interpret a missing result

Keep the query, name variants and filters. No result may reflect limited coverage, a delay, restrictive searching or a wrong identifier. It does not establish that a fact does not exist. Make the finding reproducible: write “no result for this query, within this scope, on this date”, then specify the next check: another identifier, a wider period or a complementary register.

DEEPER READING / PRACTICE

Compare what the documents actually establish

Two sources give different information on the same topic. Before concluding to a contradiction, it is necessary to check if they describe the same object and the same period.

Updated on .

Priority checks

Align the definition, date, territory, unit and status of the document. Separate method difference, update, correction and contradiction not yet solved.

Interpretation pitfalls

Different numbers may be compatible if they do not refer to the same scope. A recent publication does not always replace an old measure when it describes another period.

Working record to retain

The comparative matrix retains each value or proposition with its context. It indicates the comparable elements, the explainable differences and the questions that require a new source.

EXAMPLE / READING THE NUMBERS

Count origins rather than repetitions

Fictional data illustrating the method; not client results or measurements of this website.

Count origins rather than repetitions
ObservationCountInterpretation
Articles consulted66 publications identified.
Articles repeating one release55 repetitions: one origin.
Other documentary origin11 publication based on a distinct record.
Show the interpretation limit

What these numbers do not prove

Six articles may represent only two origins. Two origins alone establish neither independence nor the truth of every claim: check their records and methods.