baskaufs
commented
2 years ago Because the labels we are using are by default from the ACT output, they will generally all be in English. There are a few exceptions, but I did not have any trouble scanning down the column of labels and identifying them.
The way I handled this in the test was to copy the label column to the title column, then manually delete the non-English ones. This was not that hard to do, but it was a pain to have to copy the reference columns to go with it, so it would be better to just fix the script to populate all of the columns associated with the title and delete all of the cells in those columns for the non-English ones.
Non-English titles would have to be done manually unless we work out a system to populate them using data collected as described in issue #14
I ended up kicking the can down the road on titles. Since they have to be language-tagged with only a single title per work, we need to decide whether the "official" title is in English, or some other language. This issue probably needs to be addressed along with #14 since it would be the likely source of any title text that wasn't in English.
If we decide on a strategy to decide which English labels should be official titles, we could write those even if we save the non-English ones until later.