SURF Data Persistent Identifier

The volume of stored data is growing rapidly across all fields of science, as is the number of data connections. Publications, which are themselves data objects, are supported by analysed data which is in turn based on raw data. Data corresponding to a specific publication may be housed in various data centers and recorded on various types of media. Storage locations are subject to change as well. This makes it increasingly difficult to guarantee the findability of, and access to, the data. At the same time, access is becoming ever more vital due to the reproducibility requirements of research and the reuse of scientific information.

Persistent identifiers: ISBN numbers for data

To resolve this issue, a coding system for data has been developed: persistent identifiers (PIDs). PIDs are comparable to the ISBN numbers applied to books. Just as an ISBN number provides a permanent, citable reference to a certain book, PIDs do the same for data. PIDs allow us to find data and refer back to it as well. One of the most important functions of a PID is its role as a fixed reference to underlying data, no matter where the latter is located. Any researcher consulting a PID must be able to trust that he or she will find the underlying data. This applies even if the storage location or physical form has been altered.

PIDs for researchers

SURF offers researchers the opportunity to register their collected data and to make it accessible through the use of PIDs. This is done as follows:

• SURF uses the handle software provided by the Corporation for National Research Initiatives (CNRI) as a structural foundation. This handle software uses a software model resembling DNS. A reference in the top determines where each PID is located.
• PIDs consist of a prefix and a suffix. The prefix is the first piece of the persistent identifier and can be requested directly from SURF by contacting us via helpdesk@surfsara.nl. The prefix belongs to the applicant. An applicant or institution may only submit PIDs that begin with their own individual prefix. As many unique suffixes as desired may be listed under a single prefix.
• SURF acts as a host for the PIDs. The PIDs are then replicated internally at SURF, as well as externally in the context of the EPIC consortium. 
• The prefix can be used to create, modify, search for and delete PIDs. This is done through a HTTPS RESTful API.
• The so-called PID resolver is an application that allows the user to determine the location of data, or to request the data object itself, based on a PID. The PID resolver is accessible via an HTTP interface. This makes it possible to use a browser or URL to resolve PIDs at http://hdl.handle.net. The PID resolver always works with one of the three identical PIDs.