CSL Discourse

`keyword` and `language` field formats


This seems to be the case, but for confirmation:

  • Is keyword normally a semicolon-delimited list?
  • Is language normally a >-delimited list/vector of ISO 639 (+ ISO 3166) codes?


The vector syntax in “language” is a CSL-M artifact, but if systems can recognize that an entry in that format is setting the primary language of the item to the “target,” that would be a good thing.


Thanks! BTW, I realised I already asked about the language code itself: