For a public publication record, use the publication date fields rather than a Crossref index or deposit timestamp. Created, update/deposit and index dates describe metadata events: first deposit, redeposit, or reindexing after member, Crossref or external changes. Publication, print publication, online publication and acceptance dates are separate fields relating to editorial or publication events. Carry the publication date only when that field is supplied, and compare it with the publisher's own record. A recent index update does not imply the work is new or that anyone is currently working or employed.
Find public social profile clues with our TraceSeek app.
Name the field before you read anything into it
Start by identifying exactly which date field you are looking at. Crossref distinguishes created dates (first metadata deposit), update or deposit dates (a deposit or redeposit), and index dates (including reindexing caused by member, Crossref or external updates). These describe what happened to a metadata record. In contrast, publication, print publication, online publication and acceptance dates are separate fields relating to editorial or publication events. Writing "indexed" beside a timestamp, then reading it as "published," mixes two different timelines. [Crossref: metadata and publication date fields]
Separate the metadata timeline from the work timeline
A metadata record can change long after a paper or book was published: a member corrects a field, Crossref adjusts a record, or an external update triggers reindexing. None of that moves the work's publication date. Acceptance is also its own editorial event and is not an index date either. Keep three columns in front of you: field name, date shown, and what event that field describes. Only the publication-related fields belong on the work's timeline. [Crossref: metadata and publication date fields]
| observation | limit | next step |
|---|---|---|
| A recent deposit or update date | Shows a metadata change, not a publication event | Check whether a publication date field is also present |
| A recent index date | Can reflect reindexing after member, Crossref or external updates | Do not move the work's publication date |
| An acceptance date | Editorial event, separate from publication | Record it in its own row, not as the publication date |
| No publication date field present | The record does not supply it | Stop and compare the publisher record; do not substitute an index date |
Carry the publication date only when it is supplied
When a publication date field appears in the record, carry that value forward and label it as such. If the field is absent, ambiguous, or only an index or deposit date is visible, do not fill the gap with the nearest timestamp. Record the precise limit: which field was missing and what you could not establish. An unsupported inference does not become safer because it is written confidently. In the hypothetical case of a record whose index date is this week and whose publication date field is empty, the honest entry is "publication date not supplied," not "published this week." [Crossref: metadata and publication date fields]
Compare against the publisher record
Crossref metadata is one view; the publisher's own record is another. Put the two side by side and compare the publication date specifically. If the publisher shows a publication date the metadata record omits, note the source of each value rather than merging them. If the publisher page is unreachable, login-gated, or shows a different work, record that mismatch as a limit and stop. Do not seek private records, and do not assume that a difference proves error, retraction or a new edition. [Crossref: metadata and publication date fields]
Do not read a recent index update as current work or employment
A fresh index date tells you the metadata record was touched, nothing more. It does not show that the authors are currently working on the topic, that a project is active, or that anyone holds a particular role or employer. Treat those as unsupported inferences and leave them out. If a reader needs current activity, the route is a direct, permitted public statement, not a timestamp on a bibliographic record. [Crossref: metadata and publication date fields]
Close with a short date ledger
Finish by writing one line per date you actually saw: the field name, the date, the event it describes, and where you saw it. Mark any missing publication date explicitly. This ledger forces the separation between metadata events and publication events to stay visible after you stop looking at the record.
- List the field name exactly as shown before interpreting it.
- Keep created, deposit/update and index dates on a separate line from publication dates.
- Flag an empty publication date as a limit, not as a value to estimate.
- Note the publisher record comparison result, including when no comparison was possible.
When the source is absent or unclear
If the record cannot be opened, the page requires private access, or the field naming is ambiguous, write that exact limit and stop. Do not substitute a guess, a nearby timestamp, or an inference from how recent the record looks. State what would settle the question, such as a publication date field in the publisher's public record, and leave the entry unresolved until it appears.
A couple more questions
The index date changed this week. Does that mean a new article was published?
No. Index dates can reflect reindexing after member, Crossref or external updates to the metadata record. Check whether a publication date field is supplied and compare it with the publisher record before treating anything as newly published.
Which field should I use as the publication date?
Use the publication date field, or the print or online publication fields where those are what the record supplies. Acceptance is a separate editorial event. Carry the value only when the field is present; if it is absent, record the gap rather than substituting an index or deposit date.
Can a recent metadata update show that an author is still working in the field?
No. A metadata timestamp describes a record event. It does not establish current work, employment or ongoing activity, so leave that inference out and rely only on explicit, permitted public statements if the question matters.
Sources & editorial notes
Official references checked on October 24, 2026. Checklists, comparison tables and example wording are DateClues editorial tools, not official tests or validated risk scores.
General information. No guide or tool can certify a person’s identity, intentions or safety. Read our editorial standards.