Systemic Failures: Metadata Systems and Missteps
In answering a metadata question recently, I was reminded again of how the first system in which we encounter a concept can having lasting effects on how we think about that concept. It's certainly true for metadata and is one of the many reasons why I think it would be a bad idea for metadata teachers to teach "practical" stuff like DSpace, even though I see the appeal.
So. Here are brusque, dissmissive, and prejudicial assessments of how three major systems for digital metadata work will mis-influence your understandings. Take them with as many or few grains of salt as you need.
CONTENTdm
Everything is Dublin Core and nothing is happy. Collection by collection you create new profiles where you can call a field anything, so long as it's Dublin Core.
DSpace
DC Terms becomes Dublin Core + dots. The idea "dc.format.extent" makes sense to you. You can qualify Dublin Core with anything. Anything.
Fedora/Hydra/Samvera
Oh for joy you work in linked data. But ... it's flat like a pancake. No blank nodes or nuance and the triples are mostly a way of encoding text once you admit it to yourself.
That's it. That's the brusque, dismissive take I'm feeling right now. Disappointed in a world where shortcuts and simplications and systems let down not just what we can do but how we grow and think about metadata and how we break beyond any of that.