Open nh36 opened 2 years ago
Yes, I'd argue also: we can ignore the more finegrained annotation attempts made in this spreadsheet and first get them in line with our models. THEN, of course, one can discuss to which degree one can reflect parts of the models Chris had in mind.
Mattis has made a model using Baxter of a CLDF complaint dataset: https://github.com/hanproj/baxterocrhymes
Our next goal is to look at basic formats for unannotated rhyme collections.
The work for the 蒼頡篇 is largely done: https://calc.hypotheses.org/2525