DOI Frequently Asked Questions

• What is a DOI?
  DOI stands for Digital Object Identifier. DOIs are used to uniquely identify files or other resources on the internet. Address details need only be maintained in one place in order to guarantee that links to a file or resource, wherever the links are, will never be out of date, so bringing to an end the all too familiar "file not found" message.
  
  
• What can DOIs be assigned to?
  DOIs can be assigned to digital files which are the actual "product", but it is worth emphasising that a DOI need not resolve directly to the object it identifies. For example, an e-book may be a chargeable item, so its DOI may lead the user to a "shopping basket" page with the item already added to the basket; another item may only be available to subscribers, leading the user to a log-in page before the actual item is revealed.
  
  
• What is metadata?
  Metadata in this context is the information about both a DOI and the object the DOI identifies. It allows an end-user or a computer program to discover the required DOI and/or interoperate with it.
  
  
• What is a Registrant?
  The term "registrant" is applied to the person or organisation that registers a DOI and maintains its associated metadata. In theory, someone who has no connection with a web-page or other internet-based file could still register a DOI relating to it; however, no registration agency would be likely to permit this knowingly. The "registrant" will therefore normally be the publisher of the item which the DOI relates to.
  
  
• What does a DOI look like?
  DOIs consist of a unique alpha-numeric character string made of 2 parts:
  
  • A prefix, which is issued by a registration agency to a single publisher, begins with "10.", followed by a number, then a "/";
  • A suffix, which must be unique within the prefix. (The registration agency will ensure uniqueness before the DOI is registered.) Some agencies generate a suffix automatically, whereas other agencies, including Nielsen BookData, require the suffix to be determined by the registrant. This allows publishers to integrate their DOIs with other systems and identifier schemes.
  Although the suffix can be a string of any printable characters, for practical reasons (to do with invoking and transmitting DOIs within XML and URLs), it is recommended that the following characters are not used:
  
  

  ;

  /

  ?

  :

  @

  &

  =

  +

  $

  ,

  *

  "

  %

  #

  <

  >

  '

  }

  {

  |

  \

  ^

  [

  ]

  `

  SPACE

  
  DOIs are case-insensitive (e.g. 10.1000/abc is the same as 10.1000/ABC). The suffix can integrate other standard identifiers such as an ISBN, although no meaning should be automatically inferred from the suffix of DOIs in general.
  
  
• Mandatory guidelines on DOI creation
  
  • UTF-8 is the required coding.
  • If you use characters that have special meaning in URI Syntax, when you embed DOIs within URLs (for example to link the resolution system: http://dx.doi.org/10.1000/abc) you have to hexadecimal encode them. Find here the URI Syntax (http://www.ietf.org/rfc/rfc2396.txt).
  
  
  
• DOI character set
  DOIs string may incorporate any printable characters from the Universal Character Set (UCS-2), of ISO/IEC 10646, which is the character set defined by Unicode v2.0. The UCS-2 character set encompasses most characters used in every major language written today.
  
  However the Handle System (the underlying resolution technology for DOI system) at its core uses UTF-8, so UTF-8 is the required coding. UTF-8 is a Unicode implementation and so in its pure form has no character set constraints at all: any character can be sent to, stored in, and retrieved from a handle server.
  
  The IDF imposes no additional character set constraints.
  
  
• Special characters encoding
  UTF-8 is the required coding.
  Special characters can be encoded as XML entities. For example the following:
  
  
  
  
  could be encoded as:
  
  <Title language="eng">
  <TitleType>01</TitleType>
  <TitleText>Prova caratteri speciali β con beta
  </TitleText>
  </Title>
  
  Otherwise special characters can be encoded directly in UTF-8.
  Both these cases are correctly handled by the NBD registration system.
  
  
• Why are DOIs so useful?
  Put simply, a DOI provides a name for referencing a digital file, web-page or program, which can be used without needing to know its physical location or "real" name. This means that its physical location or file name can be changed without any links to it, eg. from other websites outside of the publisher's control, needing to be updated. This is called "persistency" and protects publications from the usual knock-on effect of "broken links" which prevent a publication from being obtainable.
  
  
• How should a DOI be referred to on a web page or in print?
  The correct way to cite a DOI on a webpage is as follows:
     doi: 10.1000/186
  The "hidden" html code would be:
  <a href="http://dx.nbdrs.com/10.1000/186">doi:10.1000/186</a>
  where the first part of the URL is the address of the specific DOI proxy server you wish to use. Note that http://dx.nbdrs.com is Nielsen BookData's own proxy server; DOIs can also be resolved using proxy servers run by other organisations.
  Consider also that DOIs can be "hidden" on a web-page just like URLs, such that the visible link to be clicked on can be something completely different.
  
  
• What happens if I type a DOI into my browser's address bar?
  Current versions of browsers do not recognise a DOI unless it has a proxy server address preceding it. There are free plug-ins available which "enable" the major browsers to recognise a DOI without having to type in the proxy server address as well, but unless you are sure the desired audience for a DOI have all got this plug-in, it is advisable to specify the proxy server.
  
  
• Who is using DOIs?
  Globally, around 14 million DOIs have already been registered. Many well-known international publishers, including Elsevier, Wolters Kluwer, Blackwell, Wiley, McGraw Hill, Springer, Klett, etc already use DOI technology to ensure persistent links to their publications.
  
  
• What are the alternatives?
  It is possible to use the Handle System, which is the technical foundation of DOIs, in isolation from the DOI system, and at first this may appear to be a cheaper option. However, the value of DOIs is as much in the social infrastructure around them as it is in the underlying technology. DOIs are guaranteed to remain active and resolve to something, so that even if the object it refers to ceases to exist, there will at least be a useful message confirming that the object did exist once, saving the user from wondering if they simply have an incorrect link.
  
  
• How does the service work?
  Registrants are provided with their own online account via which they can manage information ("metadata") about DOIs, using either an online entry-form or file-upload facility. The first method is a simple, convenient way to add or update just a few DOIs; for larger publishers who use other computer systems as well, the file-upload method allows multiple records to be input to the system at the same time. The file-upload can be triggered manually or can be invoked directly from automated programs via a Java program interface.
  
  
• What are the associated costs?
  The costs associated with publishing a DOI are as follow:
  
  • an annual account fee, entitling the registrant to use a number of different DOI-prefixes;
  • a small registration fee for each new DOI;
  • a small annual maintenance fee for each.