On a high level, I think the main purpose of creating this schema is:
formalize the description of terms used to describe crc datasets
generate the documentation
centralize the key-to-IRI mapping: documentation, discussion, formal specification (cuurently sfb docs provide user-facing description and tabby-utils repo contains slightly obscure IRI mapping, e.g. the tabby convention and some issues)
But on a practical level, what exactly should the schema describe, and for what it would be used in the catalog-generation workflow?
datalad-tabby
goal: describe the output of load_tabby, i.e. what comes out of a tabby-spreadsheet after adding a context with datalad-tabby?
usage: validate spreadsheets
tabby-utils
goal describe the (desired) output of further processing that is intended to go into the sfb catalog?
usage: validate catalog input
I think the two are not necessarily the same (there is some structure editing and enrichment from external lookups going on before passing the data to the catalog for a nice display).
On a high level, I think the main purpose of creating this schema is:
But on a practical level, what exactly should the schema describe, and for what it would be used in the catalog-generation workflow?
load_tabby, i.e. what comes out of a tabby-spreadsheet after adding a context with datalad-tabby?I think the two are not necessarily the same (there is some structure editing and enrichment from external lookups going on before passing the data to the catalog for a nice display).