Limitations of proprietary identifiers

Companies in the music industry supply chain make wide use of standard identifiers when communicating data to their business partners. Within company databases, proprietary identifiers may also be used for such purposes as distinguishing different encodings or product configurations of recordings. Most companies use proprietary identifiers as database keys within their local systems.

However, it is possible that an entity may not have an appropriate standard identifier. It may, for instance, be the case that musical work has not (yet) been allocated an “official” standard identifier. Or it may be t that a company needs to differentiate between two different encodings of a sound recording. It is therefore not an unusual practice for companies to transmit their proprietary identifiers as part of their communications with business partners.

Such a practice has, however, a few problems that have to do with the nature of proprietary identifiers:

Contrary to standardised identifiers such as ISRC, ISWC, GRID, ISAN etc., there are no published rules for disambiguation. Event if such rules are documented, such documentation is typically not public. And, critically, the granularity of the proprietary identifier has been chosen to meet only the allocator’s requirements. In addition, a company that allocates a proprietary identifier is free to change these rules of allocation and granularity whenever it wishes (e.g. when their supply chain software is being updated). But if another company is using such a proprietary identifier for matching, it must be aware, that these identifiers could be changed at any time and without notice.

Therefore, recipients of proprietary identifiers simply cannot rely on their persistence or accuracy for their own purposes. They can be used for matching purposes, though: if, for instance, title, main artist and proprietary ID of two entities are matches, then it might be legitimate to state that the two entities are the same.

 Note: the term “standardised” identifier here means formally standardised identifiers, but may also include proprietary identifiers where the allocating organisation has openly documented its criteria and is expected to not change them without notice.