eduKateSG · Why Science?
Cite the object you actually used
Learn how Zenodo records connect files, metadata, DOIs, versions, licences and communities so research objects remain findable and reproducible.
Research outputs are more than journal articles. Data, software, posters, reports, workflows and other objects also need stable records if other people are to find, cite and inspect them. Zenodo provides repository records that connect metadata, files and persistent identifiers.
Zenodo documentation explains that a record begins as a draft and becomes public when published. Publication registers a DOI when appropriate; published files and the persistent identifier are fixed, while metadata can be updated. New file versions become linked records with their own version identifiers.
Did you know? A concept-level DOI can guide readers to the latest version while a version-specific DOI identifies the exact release used. Choosing between them is a scientific decision: cite the specific version for reproducibility and use the broader concept link when referring to the evolving work as a whole.
This guide separates landing page, file, DOI, version, licence and community membership. Each has a different role in discovery and reuse, and none alone proves the scientific claim inside the uploaded object.
Chapter index
Chapters 1–9: record lifecycle and persistent identifiers
- 1. A record packages files and metadata
- 2. Draft and published states differ
- 3. DOIs provide persistent identity
- 4. Files become fixed after publication
- 5. New files call for a new version
- 6. Concept and version identifiers answer different questions
- 7. Metadata can be corrected
- 8. Resource type shapes expectations
- 9. Licences define permitted reuse
Chapters 10–17: versions, communities, reuse and reproducibility
- 10. Creators and contributors support attribution
- 11. Related identifiers connect outputs
- 12. Communities curate collections
- 13. Keywords and subjects aid discovery
- 14. Embargoes and restrictions affect access
- 15. Checksums help verify files
- 16. Citation needs version context
- 17. Repository literacy supports many careers
A repository record is a durable meeting point for files, people, dates, rights and identifiers. A Zenodo record combines descriptive metadata with one or more uploaded files and a landing page. Resource type labels help distinguish datasets, software, publications and other outputs. This gives science learners a concrete way to separate a landing page, a file, a DOI and a claim.
Try this
Evidence question: What research object is this record meant to represent? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: List the resource type, title, creators, publication date and files before reading the abstract. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for a record packages files and metadata, its identifier or version context, and one limit that must remain visible.
Versioning matters because a citation should resolve to the same research object a reader actually used. A draft can be edited before publication. Once published, the record becomes citable and archived according to repository rules. It also shows how preservation, attribution and reuse rights support reproducibility.
Try this
Evidence question: Is the object a private draft, restricted record or public published record? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Write the lifecycle stage beside every screenshot or note. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for draft and published states differ, its identifier or version context, and one limit that must remain visible.
Metadata can improve after publication even when archived files must remain fixed. Zenodo registers a DOI for a published record when it does not already have one. The DOI supports durable citation and global metadata discovery. The record is useful precisely because its lifecycle and identifiers can be checked.
Try this
Evidence question: Which DOI belongs to this record, and was an external DOI supplied instead? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Resolve the DOI independently and confirm it reaches the intended landing page. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for dois provide persistent identity, its identifier or version context, and one limit that must remain visible.
A community collection helps discovery, but the individual record remains the object to inspect and cite. Zenodo documentation states that published files and the persistent identifier cannot simply be modified, protecting the object a citation refers to. Metadata can be updated. A careful citation therefore names both the object and the version context.
Try this
Evidence question: Has the change affected files or only descriptive metadata? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Compare the record history with the file list and separate the two change types. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for files become fixed after publication, its identifier or version context, and one limit that must remain visible.
A repository record is a durable meeting point for files, people, dates, rights and identifiers. When published files change, Zenodo versioning creates a new linked record with separate metadata, files and persistent identifier. This gives science learners a concrete way to separate a landing page, a file, a DOI and a claim.
Try this
Evidence question: Which version contains the file used in the analysis? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Record the version-specific DOI and filename in an analysis receipt. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for new files call for a new version, its identifier or version context, and one limit that must remain visible.
CHAPTER 6 OF 17 · Inspect files and rights
6. Concept and version identifiers answer different questions
Versioning matters because a citation should resolve to the same research object a reader actually used. Versioned records can be linked across an evolving work. A general identifier may resolve to the latest release while a specific identifier anchors one release. It also shows how preservation, attribution and reuse rights support reproducibility.
Try this
Evidence question: Are you citing the work in general or the exact files analysed? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Choose the appropriate identifier for two example sentences and explain the choice. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for concept and version identifiers answer different questions, its identifier or version context, and one limit that must remain visible.
Metadata can improve after publication even when archived files must remain fixed. Published metadata such as title, creators or description can be edited without changing the DOI. A metadata update does not silently replace archived files. The record is useful precisely because its lifecycle and identifiers can be checked.
Try this
Evidence question: Which field changed, and does the correction alter your interpretation? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Record a retrieval date and note any visible metadata revision. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for metadata can be corrected, its identifier or version context, and one limit that must remain visible.
A community collection helps discovery, but the individual record remains the object to inspect and cite. A software release, dataset, poster and report require different checks. The resource type guides discovery but does not guarantee documentation quality. A careful citation therefore names both the object and the version context.
Try this
Evidence question: What files and documentation should exist for this resource type? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Create a type-specific checklist before download. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for resource type shapes expectations, its identifier or version context, and one limit that must remain visible.
A repository record is a durable meeting point for files, people, dates, rights and identifiers. Zenodo requires a licence field and lets depositors choose rights appropriate to the object. The default shown in current guidance is CC BY 4.0, but users must read the actual record. This gives science learners a concrete way to separate a landing page, a file, a DOI and a claim.
Try this
Evidence question: What licence is attached, and what attribution or share conditions follow? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Translate the record licence into a plain-language reuse note. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for licences define permitted reuse, its identifier or version context, and one limit that must remain visible.
CHAPTER 10 OF 17 · Choose the version
10. Creators and contributors support attribution
Versioning matters because a citation should resolve to the same research object a reader actually used. Creator metadata names people or organisations responsible for the object, while contributor roles can describe other participation. Identifiers such as ORCID may improve disambiguation. It also shows how preservation, attribution and reuse rights support reproducibility.
Try this
Evidence question: Who should be cited, and are roles explicit? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Build a citation from the record rather than guessing authorship from a file cover. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for creators and contributors support attribution, its identifier or version context, and one limit that must remain visible.
Metadata can improve after publication even when archived files must remain fixed. Records can link associated publications, datasets, software or earlier versions using relation types. The direction of the relationship matters. The record is useful precisely because its lifecycle and identifiers can be checked.
Try this
Evidence question: Does the link mean cites, isSupplementTo, isVersionOf or another relation? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Draw arrows between two records and label each relation exactly. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for related identifiers connect outputs, its identifier or version context, and one limit that must remain visible.
A community collection helps discovery, but the individual record remains the object to inspect and cite. Zenodo communities provide shared areas where projects, institutions, domains or conferences can review and organise records. Community inclusion aids discovery but does not replace record-level evaluation. A careful citation therefore names both the object and the version context.
Try this
Evidence question: Who curates the community, and what does membership actually certify? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Open one community and inspect an individual record separately. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for communities curate collections, its identifier or version context, and one limit that must remain visible.
CHAPTER 13 OF 17 · Preserve the citation route
13. Keywords and subjects aid discovery
A repository record is a durable meeting point for files, people, dates, rights and identifiers. Descriptive terms help searches and indexing. Broad or inconsistent keywords can still produce noisy results. This gives science learners a concrete way to separate a landing page, a file, a DOI and a claim.
Try this
Evidence question: Which terms reflect method, topic, organism or output type? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Rewrite a search using one identifier and two controlled or specific concepts. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for keywords and subjects aid discovery, its identifier or version context, and one limit that must remain visible.
CHAPTER 14 OF 17 · Preserve the citation route
14. Embargoes and restrictions affect access
Versioning matters because a citation should resolve to the same research object a reader actually used. Zenodo supports visibility and access settings, including embargo or restricted files where configured. Metadata may remain visible even when files are unavailable. It also shows how preservation, attribution and reuse rights support reproducibility.
Try this
Evidence question: What can be inspected now, and what requires permission or waiting? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Write an access-status note without claiming that unavailable files do not exist. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for embargoes and restrictions affect access, its identifier or version context, and one limit that must remain visible.
Metadata can improve after publication even when archived files must remain fixed. Repository file information may include checksums that allow a downloaded file to be checked for accidental change or corruption. A checksum verifies bytes, not scientific validity. The record is useful precisely because its lifecycle and identifiers can be checked.
Try this
Evidence question: Does the local file match the repository checksum? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Calculate or compare a checksum and state exactly what that result proves. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for checksums help verify files, its identifier or version context, and one limit that must remain visible.
A community collection helps discovery, but the individual record remains the object to inspect and cite. A good citation uses the record’s creators, title, date, version where relevant, publisher or repository and DOI. Analysis notes also preserve filename and retrieval date. A careful citation therefore names both the object and the version context.
Try this
Evidence question: Could another reader obtain the exact same object? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Create a citation plus a technical receipt for one file. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for citation needs version context, its identifier or version context, and one limit that must remain visible.
CHAPTER 17 OF 17 · Preserve the citation route
17. Repository literacy supports many careers
A repository record is a durable meeting point for files, people, dates, rights and identifiers. Researchers, software developers, librarians, archivists and data stewards all publish and maintain research objects. Students can practise the same habits with class projects. This gives science learners a concrete way to separate a landing page, a file, a DOI and a claim.
Try this
Evidence question: Which role ensures identity, rights, documentation or long-term access? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Design a mini Zenodo-style record for a school investigation. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for repository literacy supports many careers, its identifier or version context, and one limit that must remain visible.
Contents · Previous chapter · Continue at the Science Learning Hub
A persistent link becomes reproducible evidence when the version is clear
Zenodo records help research objects remain findable, citable and connected. The decisive reading move is to identify exactly which record and files a claim uses.
Keep DOI, version, filename, licence and retrieval date together. Then follow related identifiers to the paper, software or dataset needed for the next evidence question.
| Reading object | What it can support | Question to keep visible |
|---|---|---|
| Record DOI | Persistent identifier for the published object | Is it concept-level or version-specific? |
| File | Archived research object | Which version and checksum apply? |
| Licence | Permission and attribution conditions | What reuse is allowed? |
| Related identifier | Connection to another output | What relation direction is declared? |
Sources and next reading
Substantive resource claims in this article were checked against the primary pages below. A database record is a structured route into evidence; it does not by itself prove a new biological conclusion. Recheck the live record, evidence labels and release context before using it in assessed work or research.
Primary sources
- Zenodo about records — accessed 24 October 2026.
- Zenodo DOI guidance — accessed 24 October 2026.
- Zenodo version management — accessed 24 October 2026.
- Zenodo licence guidance — accessed 24 October 2026.
- Zenodo communities documentation — accessed 24 October 2026.