Thanks! Here's another fun idea (sorry I'm not implementing this myself and submitting as a pull request; do feel free to ignore me): For the ones that have lipsum because they don't have text, you could use the top comment. Something like:
"COMMENT CONTENT," USERNAME said when prompted
USERNAME had this to comment: "COMMENT CONTENT"
We interviewed USERNAME, who said: "COMMENT CONTENT"
"COMMENT CONTENT," USERNAME said when prompted
USERNAME had this to comment: "COMMENT CONTENT"
We interviewed USERNAME, who said: "COMMENT CONTENT"
etc.