We are just starting up a dspace archive for oral histories and I am making my way through the dspace wiki to understand best practices.

One question, in particular, needs to be resolved before we can proceed. Each oral history has at least two pieces--an audio file (sometimes more than one) and a transcript relating to that file. In addition, there may be subsets of that master file (clips, podcasts) with related bits of the transcript.

What is best practice for putting these items in the archive? If we upload each item as its own element, that's great and enables us to capture all of the file-specific metadata appropriately, but would mean re-entering general metadata (e.g., noting the interviewer and interviewee, where the interview took place, the interviewee's bio, key words related to the interview, etc.)

If we upload files together, then how do we capture file-specific info (e.g., "extent" or filetype, checksum, etc.)

Am I missing this is an obvious place in the online docs? If not, what practices are people using?

Many thanks,
Ari Davidow