next up previous
Next: Resolution of URNs and Up: Resource names Previous: URNs and LIFNs

Format of URNs and LIFNs

NOTE: The format of RCDS URNs and LIFNs described here is based on recent discussions in the Internet Enginering Task Force and elsewhere, and is subject to change until the standards are finalized. RCDS can thus serve as an early URN testbed, and will migrate to whatever URN format and resolution schemes are adopted as Internet standards.

A URN consists of three parts:

  1. The fixed prefix string URN:.
  2. A namespace identifier (NSI), which identifies the format of the remaining portion of the URN.
  3. A namespace specific suffix (NSS), which is an identifier assigned according to the rules for that particular name space.

The NSS will usually be subdivided into a naming authority, and a string which is assigned by that naming authority, which may itself be subdivided if further delegation is needed. So URN:inet:foo.bar:mumblefrotz would be a URN that was assigned by the naming authority foo.bar.

The location of the naming authority within the NSS is not fixed; rather, it is a characteristic of the name space. This allows URNs to serve as an ``umbrella'' for other naming schemes (e.g. ISBNs, SGML Formal Public Identifiers, Usenet Message-IDs) that have a variety of structures.

For RCDS URNs, the name space identifier will normally be inet (though any URN can be used) and the naming authority will be an Internet domain name. A LIFN is a URN from the name space lifn; the naming authority portion of a LIFN will also be an Internet domain name. While any Internet domain name could potentially be used, domain names used for URNs and LIFNs should be chosen to allow them to be persistent for the useful life of the resource.



Keith Moore
Fri Feb 7 11:53:58 EST 1997