Frequently asked questions

A Persistent Identifier is a permanent and unique identification code attached to a digital object (a scan, audiovisual file, metadata record, website etc.). The identifier is separate from the object’s location. It is a unique code registered at an agreed location, ensuring that the object can always be located on the Internet, even if the object’s name or location changes. In this way, the object can always be clearly referenced and located.

We currently use URLs (Uniform Resource Locators) to identify a digital object on the Internet, but we have all encountered broken or dead links referring to web pages that are no longer available. This is because URLs can change over time due to technical or organisational changes, such as a name change. This, in turn, has unwanted consequences for links and references to the digital objects in portals like Europeana, for instance. That is why a Persistent Identifier is needed. It serves as a permanent identification code attached to a digital object and always refers to that object, independent of the underlying technology used to access it.

The Persistent Identifier (PID) can be compared to the International Standard Book Number (ISBN), the unique identifier attached to every published edition of a book. International agreements were made about how ISBNs are assigned, and by who. A central administration tracks which ISBN corresponds to which work. The same goes for Persistent Identifiers. When a PID is assigned to an object, the PID and information regarding the object are entered in a central register, along with the object’s location. Whenever a change occurs in the name or location of the object, this change is registered and a reference to the new location is added. The PID, then, always refers to the object’s most recent location. Technical solutions alone are not enough, however. Measures must be taken at an organisational level as well. In principle, an object can only be sustainable as long as the organisation in question guarantees and maintains its durability.

The image of The Jewish Bride, made accessible online by the Amsterdam Rijksmuseum, has the following Persistent Identifier: http://hdl.handle.net/10934/RM0001.COLLECT.5223. When you click this link, it reveals the current web address: https://www.rijksmuseum.nl/nl/collectie/SK-C-216. The Persistent Identifier ensures that The Jewish Bride remains accessible, even if this web address should change in the future.

Technically, it is possible for an organisation to make sure that their links remain functional, but this solution is far less effective than using PIDs, for five crucial reasons:

  1. The large PID systems are aimed at longevity and affiliated with large, international data centres belonging to several governments. This international network is surely more robust and durable than your own organisation.
  2. If you are ever forced to divide your collection over multiple organisations, it will be virtually impossible to maintain the same URLs.
  3. In the long term, an individual solution is more costly and laborious than a shared solution.
  4. An individual system is less user friendly, as users cannot tell from the URLs for your objects that they are intended as Persistent Identifiers. As a consequence, they cannot gauge the reliability of these links if they intend to refer to these objects
  5. By yourself, you lack the ability to keep track of how your Persistent Identifiers are used.

Furthermore, Persistent Identifiers are increasingly seen as an indicator of reliability. Their absence will ensure that end users and fellow institutions will no longer wish, dare, or be allowed to refer to your material.

The costs consist of the one-off costs of implementation and the structural costs of maintenance and membership of a PID system.

The structural costs depend on the PID solution chosen and range from a few dozen to a few hundred Euros annually.

The one-off implementation costs cover the adaptation of your collection management system and website. The Digital Heritage Network’s Persistent Identifiers project helped providers of commonly-used collection management systems implement PIDs in their systems and make PIDs as simple and affordable as possible. 

In many cases, the impact of implementation can be softened by reusing existing numbering systems and leaving the registration of the PIDs to the collection management system.

It is possible to take a step-by-step approach to implementation, starting small, provided possibilities for expansion are factored in from the start. Please use the 'Stappenplan' (stepwise guide) for the implementation of PIDs elsewhere on this website.

No. A number of standard solutions are in place. A different, individual implementation is also an option, but solutions for that exist at other institutions as well. 

The Persistent Identifiers project has collected good practices of organisations who implemented PIDs. They are available elsewhere on this website.

Various methods of implementing Persistent Identifiers exist. We recommend using one of the international standards applied in the Netherlands: ARK, DataCite DOI, Handle or URN:NBN.

The Persistent Identifier systems ARK, DataCite DOI, Handle and URN:NBN are based on techniques that have been used for over 15 years. In the rapidly evolving world of the Internet, they have more than proven their durability, and these systems are increasingly well-known and widely used. Since 2001 some 8.2 billion ARKs have been created by over 1000 organizations — libraries, data centers, archives, museums, publishers, government agencies, and vendors. The Handle system has existed since 1990 and is used in over 75 countries by over 10,000 universities, research centres, libraries, archives etc. DOI is an implementation of Handle which has over 100 million objects registered. URN:NBN has been in use at national libraries and archives for over 15 years.

The project has developed a guide to help you choose the most suitable PID for your organisation. Please see our PID Guide

Your organisation is ready to implement Persistent Identifiers as soon as you can uniquely identify and describe those objects that you wish to preserve for the long term, and are planning to make these descriptions and/or objects digitally accessible online.

PIDs represent a way to persistently identify and reference objects. As such, they are suited to every object that you wish to make accessible and referable for the long term.

It is recommended to use PIDs for collections and objects that you wish to publish online (as a record or image) and to which you intend to ensure sustained access. The PID Guide developed by our project can help you make this choice.

For ARKs contact the ARK Alliance.

For URN:NBNs, in the Netherlands you can only turn to the KB, National Library’s registration agency.

For Handles, you can go to SURF. They offer PID services in collaboration with the European Persistent Identifier Consortium (EPIC). These services are free of charge only for scientific institutions, but also available to heritage institutions.

For DOIs, the most prominent provider in the Netherlands is DataCite Netherlands.

For the Handle and DOI systems, however, you may also register PIDs with a different party.

The PID resolvers of the PID suppliers mentioned elsewhere on this website or their partners have proven to be robust and durable in recent years. In addition, PID suppliers are often not-for-profit organizations that have a commitment to long-term availability.

However, as an institution you can choose to have the PID resolver hosted by a party of your choice or to do it yourself. In principle, these solutions are less durable and reliable.

In cases like this, the most common scenario is that (the management of) the prefix (and all accompanying PIDs) is handed over from organisation A to organisation B. If required, the locations to where the PIDs point will have to be updated (e.g. to their new URLs at organisation B). With URN:NBN, the organisations submit the changed locations of the objects to the national resolver.

A PID almost always consists of a prefix and a suffix. The prefix is related to an organization, the suffix to the object. If one or more PIDs change organisations, both organizations will have to make agreements about the PIDs. Typically, organization B will create new PIDs for the objects. Organization A can have its PIDs point to it. The organizations can also agree that organization A creates a 'tombstone' and refers to organization B on it. Organization B can include the former PID in the metadata for the objects, so that it can still be found. The various PID suppliers can tell you more about the (im)possibilities of their systems.

Yes, you can! On our About page, you can download a PID Guide toolkit that you can use to build your own PID Guide.