three-letter codes for identifying languages: Migrating from in-house standard to community standard

A foundational aspect of documenting an endangered language and preserving that documentation for long-term access is identifying the language itself. The web version of the Ethnologue has become the de facto standard for identifying the more than 6,800 languages spoken in the world today. The system of three-letter codes that uniquely identify each language has been used within SIL for nearly three decades as an in-house standard, but now there is increasing demand for these codes to be used by other organizations and projects. This paper describes four changes that SIL International is implementing in order to make its set of language identification codes better meet the needs of the wider community. The changes seek to strike a balance between becoming more open while at the same time becoming more disciplined. The need for language identifiers A foundational aspect of documenting an endangered language and preserving that documentation for long-term access is identifying the language itself. Effective retrieval of resources depends on the uniform identification of the languages to which they pertain. Simply using language names in metadata is not adequate since the same language is typically known by many names and those names change over time. Furthermore, different language