DOI, ISSN, ISBN, URN and ORCID: Identifiers for Proceedings

Each identifier points at a different thing: a paper, a series, a book, a web resource or a person. Mixing them up is a frequent source of wrong references.

Last updated 6 November 2026

What each identifier identifies

A persistent identifier is a label meant to keep working even when a web address changes. Persistence comes from the organizations behind the label, not from the label itself. The DOI Foundation says as much on its own site.

The sections below explain each one. Compare how services use them in Proceedings Repositories Compared.

DOI: the identifier for a paper

A Digital Object Identifier (DOI) has a prefix and a suffix separated by a slash. The DOI Foundation's own illustration is 10.1000/182. The DOI System is standardized as ISO 26324, and the DOI Foundation, a not-for-profit, governs it.

To resolve a DOI, put it after https://doi.org/. The system then redirects you to the linked object, most often a web page. The DOI Foundation says resolution is free.

Use the DOI exactly as printed on the paper's record page. Do not guess a DOI or build one from a title. A fictional example in the preferred link form is:

https://doi.org/10.0000/example.2023.1

If the link fails, the DOI may have been mistyped, or it may not have been registered. Verification steps are in How to Verify That a Citation Is Real and Correct.

Crossref and DataCite: who registers DOIs

DOIs are obtained through a registration agency. Two agencies matter most to readers of proceedings.

• Crossref is a not-for-profit membership organization founded in January 2000. It shares metadata openly through free APIs, and its stated mission is to make research objects easy to find, cite, link, assess and reuse.
• DataCite is a nonprofit membership association. It supports DOIs for research outputs such as data, preprints, images and samples. Zenodo, for example, states that its DOIs are registered with DataCite.

The practical point: both agencies expose metadata that you can look up from a DOI. When a reference manager imports a record, it often asks one of them. If the metadata looks wrong, compare it with the paper itself.

ISSN: the identifier for a series

An International Standard Serial Number (ISSN) has eight digits, written as two groups of four with a hyphen, for example 1234-5678 in the form only. The eighth character is a check digit computed with a modulus 11 method, and it is written X when the result is 10, according to the ISSN International Centre.

An ISSN identifies a continuing resource regardless of medium. It carries no meaning about origin, content or quality. Each medium has its own ISSN, and an ISSN-L links them. For instance, CEUR-WS.org states ISSN 1613-0073 for its series, and PMLR states ISSN 2640-3498, with 1938-7228 for its earlier name.

Because an ISSN names the series, do not use it as a paper identifier. Put it in a reference only if your style or venue asks for it.

ISBN and URN: books and named resources

An International Standard Book Number (ISBN) identifies a book edition. Proceedings published as books often carry one. Different formats, such as print and electronic, normally carry different ISBNs. We did not check ISBN rules on the agency's own site for this page, so confirm the details with your national ISBN agency.

A Uniform Resource Name (URN) names a resource independently of its location. CEUR-WS.org states that each volume has a URN of the form urn:nbn:de:0074-NNNN-x. Treat a URN as a backup label for the volume. It is not a replacement for a DOI.

ORCID: identifying people, not papers

ORCID stands for Open Researcher and Contributor ID. It is a persistent identifier for researchers, and ORCID states that individuals can create and maintain a record free of charge. Many submission systems ask authors for an ORCID iD so that two people with the same name are not confused.

An ORCID iD does not appear in most reference styles. It helps you check an author's identity when you verify a citation. Use it with care: an iD shows who someone is, not that a paper exists.

When a proceedings paper has no DOI

No rule says every paper must have a DOI. CEUR-WS.org states that some volumes have a DOI and that DOI assignment is being tested. The ACL Anthology states that it does not assign DOIs to third-party materials from sister societies.

If no DOI exists, cite the proceedings title, editors, year, publisher or series, page range and a stable URL. Prefer a URL from the hosting service over a search-engine link. Check the page again before you submit. For archived copies, see Link Rot: Archiving the Web Pages and PDFs You Cite.

Common mistakes

• Using a series ISSN where a paper DOI belongs.
• Typing a DOI by hand and adding a trailing full stop to the link.
• Assuming a DOI proves the paper is peer reviewed. It proves registration only.
• Citing a DOI of a volume when you used one paper.
• Guessing a missing identifier.

More on this theme is in Ten Common Mistakes When Citing Proceedings.

Frequently asked questions

Sources and further reading

Related topics

• How to Verify That a Citation Is Real and CorrectStep-by-step check that a reference is real and correct: search the title, resolve the DOI, compare Crossref and DBLP, and spot fabricated or AI citations.
• Proceedings Repositories Compared: ACL Anthology, ACM DL, IEEE Xplore, PMLR, CEUR-WS and MoreA neutral overview of places that publish or host proceedings: ACL Anthology, ACM DL, IEEE Xplore, Springer, PMLR, OpenReview, CEUR-WS.org, arXiv, Zenodo.
• Glossary of Proceedings and Citation TermsPlain-language glossary of proceedings, publishing and citation terms: DOI, ISSN, camera-ready, preprint, version of record, CSL, BibTeX and open access.