Real rows out of dash_comment_ideas: the comments are what viewers actually wrote under Historically's video, the bends are what the engine actually generated. Nothing here is filler. Every draft uses YouTube's own comment skeleton - 40px avatar, bold handle, timestamp, body, thumb up with count, thumb down, REPLY - and puts the bends where YouTube puts replies.
"how did you become the Pope?" "I watched Historically's online guide"
Closest to the screenshot you sent. The bends are one click away, so the feed scrolls like a comment section: one ask per row, nothing else competing.
"how did you become the Pope?" "I watched Historically's online guide"
Everything visible, nothing to discover. Taller per card, so fewer asks fit on a screen.
Being a bishop is hard. Imagine only being able to move diagonally for years
The one thing a real comment section gives you for free and this feed does not: which video the person was watching. Costs one line.
Being a bishop is hard. Imagine only being able to move diagonally for years
One idea leads, the rest wait. Keeps the card short and still shows a real title rather than a count.
Nobody expects the Spanish inquisition!
Every pop culture reference I noticed Enrico Pucci (JJBA Part 6) Pomni (TADC) Susie (Deltarune) "nobody expects the Spanish Inquisition" (Monty Python)
Two people asked for the same video. This is the only draft where the card itself proves demand - the number of faces IS the number.
All five keep: the exact comment text, the real author and avatar, the real like count, and a link to the comment on YouTube. None of them show a view count, an outlier badge or a thumbnail - nobody has made these videos, so there is nothing to measure.