duolingo

[ben-domingue]

see code here: https://github.com/ben-domingue/irw/blob/main/data/duolingo2018.R
at present this leads to a ragged table. we need to straighten this out. you can get some sense of the problem if you just look at the attached in a spreadsheet [first 100 lines of duoling__listen.csv]

dl.csv

[ben-domingue]

i think it might even work to just cut what is past part.speech