ConceptPermanent identity, not permanent location

The fundamental insight behind DOI is a separation of identity from location. A traditional URL points to a place — a specific server directory path. When the server structure changes, the URL breaks. A DOI, by contrast, identifies the object itself independently of where it happens to be hosted at any given moment.

When a publisher moves content to a new server, changes its URL structure, or even changes its name after a merger, they update the DOI resolution record. The DOI itself remains unchanged. Readers clicking the same DOI link ten years later will still reach the paper.

AnatomyHow a DOI is structured

DOIs follow a specific structure governed by ISO 26324:

10.XXXX/SUFFIX

The “10.” prefix identifies the identifier as a DOI. The “XXXX” is the registrant code assigned to the publisher or organization by the registration agency (usually Crossref for academic journals). The suffix is chosen by the registrant and can encode journal, volume, issue, and article identifiers.

ResolutionHow DOI resolution works

When you click a DOI, your request goes to a resolver (typically doi.org), which queries the central registry for the current URL. The registry returns the current location, and you are redirected there automatically. This happens in milliseconds.

The central registry is maintained by the International DOI Foundation (IDF), with Crossref and DataCite handling the vast majority of academic registrations. The resolver is free to use and requires no account.

LimitsWhat a DOI does not guarantee

A DOI guarantees persistent linking, not permanent access. If a publisher goes out of business and no archive takes over the resolution record, the DOI may eventually break. This is rare but not impossible. DOIs also do not guarantee quality: predatory journals can register DOIs through some agencies.

However, for the vast majority of academic publishing, DOIs work exactly as intended. They have dramatically reduced the rate of broken reference links in scholarly communication.

Key Takeaways

• A DOI identifies an object permanently, independent of its current web location.
• The structure is 10.XXXX/SUFFIX, governed by ISO 26324.
• Clicking a DOI resolves through doi.org to the current publisher URL.
• Always use DOIs in citations when available — they are more reliable than URLs.
• A DOI guarantees persistent linking, not permanent access or quality.

FAQPeople also ask

Is a DOI the same as a URL?

No. A URL points to a location on a specific server. A DOI identifies the object itself and resolves to its current location through a central registry. URLs break; DOIs persist.

How do I find a paper’s DOI?

Look at the top of the article PDF or the article’s landing page. Most journals display the DOI prominently. You can also search Crossref at search.crossref.org or Google Scholar.

Can a DOI change?

Generally no. The DOI itself is permanent. The URL it resolves to can change, but the DOI record is updated to reflect the new location. This is the entire point of the DOI system.

Do all journals use DOIs?

Most established journals do, but not all. Some smaller or newer journals may not have registered DOIs yet. The absence of a DOI is not necessarily a red flag, but its presence is a positive signal.

Can predatory journals have DOIs?

Yes, through some less selective registration agencies. However, DOIs from major publishers and Crossref-registered journals are reliable. Verify the journal in DOAJ or a major index, not just by checking for a DOI.

SourcesReferences & further reading

• DOI.org — Official Resolver
• Crossref
• DataCite
• ISO 26324: Digital Object Identifier
• DOI Key Facts

Practical GuideHow to Find and Use DOIs

Finding a DOI:

1. Check the top of the article PDF — most publishers display the DOI near the title or header.

2. Look on the article’s webpage for a “Cite this article” or “DOI” link.

3. Use doi.org — enter what you know (title, author) and the resolver will find the DOI.

4. Use Crossref’s metadata search at search.crossref.org.

Using a DOI in citations:

Always use the DOI format: https://doi.org/10.xxxx/xxxxx. This is the resolver URL that will redirect to the current article location even if the publisher changes platforms.

Common MistakesDOI Misconceptions

Mistake 1: Using URLs instead of DOIs in references. URLs break when publishers migrate platforms. DOIs persist. Always prefer DOIs when available.

Mistake 2: Thinking all articles have DOIs. Older articles (pre-2000) may not have DOIs. Some journals, especially in humanities, still do not assign DOIs. In these cases, use a stable URL or ISBN/ISSN.

Mistake 3: Confusing DOI with PMID. A PMID is a PubMed identifier specific to the PubMed database. A DOI is a universal persistent identifier. They serve similar purposes but are managed by different organizations.

Mistake 4: Using the wrong DOI. Some articles have multiple versions (preprint, accepted, published) with different DOIs. Always cite the version of record (the final published version).

FAQFrequently Asked Questions

Is a DOI the same as a URL?

No. A DOI is a persistent identifier for a digital object. A DOI resolver URL (https://doi.org/…) functions like a URL, but the underlying DOI remains constant even when the resolution target changes.

Who manages DOIs?

DOIs are managed by the International DOI Foundation and registered through agencies like Crossref (for scholarly publications) and DataCite (for datasets).

Do DOIs ever change?

No. A DOI is permanent and immutable. If an article is retracted, the DOI remains but may redirect to a page noting the retraction status.

How do I create a DOI for my dataset?

Use repositories like Zenodo, Figshare, or Dryad, which automatically assign DOIs to deposited datasets. Your institution may also have a DataCite membership.

Can I access an article for free using its DOI?

The DOI provides persistent access but does not guarantee free access. However, many articles are now open access, and tools like Unpaywall and Open Access Button can find free versions using the DOI.