At the recent annual meeting of the Linguistic Society of America in Pittsburgh Jeff Good of University at Buffalo and I organised a tutorial session (Friday 7th January, 1.5 hours) and poster session (Sunday 9th January, 3 hours) on the topic of metadata in language documentation and description.
The tutorial talks covered general topics such as how to design a metadata system and what it can be used for, what kinds of metadata researchers are collecting, how linguists’ metadata relates to that developed by anthropologists and archaeologists, and what information archives need for the best description and preservation of language materials. The poster session presented specific case studies from on-going archiving projects.
Jeff and I are able to bring together field linguists, computational linguists, language archivists, anthropologists, and archaeologists to discuss the issue of metadata from an interdisciplinary perspective. The poster session included presentations of a number of archives of endangered languages materials and displayed their approaches to metadata.
One thing that became clear from the presentations and posters was that early work in language documentation starting around ten years ago was heavily influenced by library concepts (eg. Dublin Core), and that key metadata notions were interoperability, standardisation, discovery, and access (see, eg. OLAC,