How programme metadata is stored? Each programme have a unique id (see ....). The minimal data for a programme which is enough for various listings and search results should be easy to access. This includes: title, creator, dates, abstract, icon, rating, language, length, type, genre, rights (?) Otherwise there must be a possiblity to store arbitrary new metadata elements. Basically we prepare for metadata elements from XBMF and SOMA. There are the following repeatable elements in SOMA: title.alternative creator creator.role (??) subject contributor contributor.role type (??) format (??) format.extent language relation.* coverage.* extendedInfo Persons and organizations are described with contact records (sotf_contact) which are reusable in several metadata records.