DOI (Digital Object Identifier) is a unique permanent string of symbols (letters and digits).

Any object can be tagged by DOI – physical, digital or abstract – which needs identification, primarily within specific community or for intellectual property protection. Objects for DOI allocation are books and their chapters, collections, contributions to collections, articles of electronically published magazine, dissertation, norms, technical notices etc. DOI is an analogy of bar codes, which is used for physical commodities.

DOI are the only broadly accepted permanent identifiers for scientific theses. They appear in printed materials and on the web (relevant objects, as references…).

International DOI Foundation (IDF), which maintains the basic infrastructure for DOI registration and application, takes care about the DOI system and its development. Furthermore it provides licenses for registration agencies, which register DOI and offer services for application in various fields.

Registration agency CrossRef specializes in scientific research field and in services for professional literature publishers.

DOI itself consists of two parts – prefix and suffix – separated by a slash. Prefix is assigned to publishers by CrossRef after they become members of CrossRef, while suffix is assigned by the publisher. Each and every combination of prefix and suffix must be unique – one object can have only one DOI and no two objects can have the same DOI.


CrossRef is an official international registration agency providing organizational ant technical infrastructure for DOI registration and application in the field of scientific research.

The most visible and most significant outcome is the possibility to obtain stable, permanent references to electronically published scientific theses. These references appear primarily in the list of references in the publisher’s document (most frequently scientific article) and they direct to the material, which is cited by the author.

In the CrossRef system each DOI has its set of allocated metadata and a URL, so-called “response page” – web page with a full bibliographic reference of a document, DOI and information about access to full text of a document. It is recommended to import abstract here as well. If the user clicks a link containing DOI (for example in reference to scientific article), the system redirects the DOI to that URL. (The Actavia system automatically creates this page.)

This way the entry’s contents are clearly specified and a permanent hypertext reference is saved on the internet (resistant to changes in location after web reconstructions, in case of object’s owner change etc.) The publisher has access to full text under control (he can enable access to authorized users only, impose a charge etc.).

The publisher is obliged to add outbound links to every document’s reference which refers to a material (e.g. scientific article or book) registered in CrossRef system. This is also done by the Actavia system.

CrossRef is not a database of articles (does not store full texts or abstracts) –the DOI uses scientific article references for reliable linking to their full digital texts (or their abstracts), it works as a switchboard basically.

By March 2014 more than 4800 cooperating institutions (big and small institutions, commercial and noncommercial publishers, scientific companies etc.) have joined CrossRef system and they have registered more than 89 million DOI.

Major advantages of DOI in CrossRef system

  • Principal advantage are permanent DOI references (URL may change, but DOI stays the same) – publishers and other CrossRef users create reliable permanent links in references and database records.
  • DOI enables reliable identification of content – If a user accesses an article via DOI, he can be sure that it is a publisher managed version.
  • A single agreement with CrossRef serves as an agreement with all cooperating parties, whose materials it is referred to. It is not necessary to sign a number of bilateral agreements with article publishers.
  • Evaluation of electronic publications – readers increasingly expect online materials to contain hypertexts to cited sources. Thanks to references from other users, the accessibility and turnout of publisher’s content is getting better at the same time.
  • Popularity boost for authors – In case of documents it is possible to display present information about how often they are cited, including material links in which these citations appear (Cited-by Linking service).

List of publisher’s obligations in CrossRef

  • Allocation of DOI and XML format metadata sending to CrossRef system for all documents, which full text is available online.
  • Addition of DOI outbound links (if available) to the list of references.
  • All CrossRef DOI must link to valid response page.
  • Check their own transactions with CrossRef on regular basis, check notifications and correct issues.
  • Update DOI metadata, especially URL sources.
  • Resolve duplications of allocated DOI. (CrossRef system checks if one DOI is not allocated multiple times and it eventually alerts of the error.) In the same way the publisher is responsible for objects to have only one DOI.

DOI & CrossRef on

  • Allocation of unique DOI.
  • Creation and export of metadata to CrossRef.
  • Analysis of references including search for reference DOI and their automatic linking to original source.
  • Automatic creation of response page – a web page with the document’s information.
  • Download of information about article citation rate (CrossRef cited-by).

The doivup application automatically analyses references, finds their DOI and adds a link to their original source (if it is saved in the CrossRef system). An advanced reference analysis is a key condition for sending a full article description to CrossRef. Referential parser success rate is among the top algorithms of this kind worldwide. The system also tries to search for non-linked references.

University’s cited-by mode is active, which enables information gathering about citation rates and links to the cited materials for all allocated DOI. Citation rates data are obtained daily by CrossRef agency’s Cited-by Linking service. One condition is sending a list of references in metadata for CrossRef – their processing and sending is managed by doivup application.

Many other search servers are able to extract the bibliographic description from so called Dublin Core headings, which are added to the response page of every document by doivup server.