- Data sets NOT records for bibs
- RDF NOT MARC
- data input screens NOT encoding templates
Thanks, @alvet and @kartus. Lots of interesting ideas to unpack here. I think the notion of data sets (e.g. ‘not records for bibs’) closely aligns with the FOLIO Codex plans of many knowledge base sources. For instance, there could be many sources of records, such as from an electronic journals subscription agent, an inventory of the physical books in a library space, and from an institutional repository. Each of these sources coexists in the FOLIO Codex. We don’t have a lot of written information about the Codex yet, but we do have a working definition: A normalization and virtualization layer that allows for FOLIO to integrate metadata about various resources regardless of format, encoding, or storage location. Yes – it is a very geeky definition, but the core of what we’re trying to do is there.
Two other ideas fall naturally from the Codex. First, RDF-or-MARC would be a library’s choice; each would work equally well. Second, yes input screens and not encoding templates! For both of these, what matters is that there are apps that would handle the editing of each type of record: a MARC editor that edits MARC records, a BIBFRAME editor that edits triples in a triplestore, a PBCore editor that edits records about audio/visual materials, and so on. The Codex contains enough information to be able to do things like purchasing, circulation, and statistics gathering without needing to understand the intricate details of the bibliographic data.
The raw details about the work defining and creating the Codex are on Wiki.folio.org: https://wiki.folio.org/pages/viewpage.action?pageId=1414343