Initial work on a MAJW build process
to : Martin Holmes
Minutes: 460
I’ve taken on the role of generating the TEI output from the database XML dump, and spent most of the 26th getting familiar with the db structure. There’s still a lot that remains opaque, but I have a couple of XSLT files in place, and I’m already generating an expanded placeography (whereby the multi-place pseudo-locations in the db are linked to their component places, as well as taxonomies for businesses (not sure what those are yet, but they might be types of legal action), relationships, and languages. I’ll have to merge in some existing TEI which includes the modern personography and the prefixDefs, but that should be easy to achieve. This build will be a precursor to proper diagnostics which should be able to check internal consistency.
In subsequent days, I wrote transformations for the rest of the trivial tables, and began addressing the complexities arising from the fact that primaries and relatives may actually be the same person in many cases (in other words, the same person has entries in both tables because they perform both roles). I’ve written some code to help identify those cases and have been discussing remedies with AWB, and we may make some changes to the db itself to address this.