Twenty-seven real YouTube titles, English only, out of the same catalogue. Pick the five a stranger would most want to watch and the five they would least, and say which rows are not really video listings at all. No translation, no puzzles — just the judgement an editor makes a thousand times a day, with a survivor's memorial speech, a council meeting and David Lynch's weather report in the pile.
3 videos featuring this round
Rounds stay open: later videos added their new models to this one.
You are helping to build the front page of a site that surfaces YouTube videos by the city they were filmed in. The batch below comes out of the catalogue exactly as it was stored. The title is all a visitor sees before deciding whether to click — there is no thumbnail, no channel page, no view count, nothing else.
Below are 27 titles, all in English. Tell me which five a general audience would most want to watch, which five least, and which of them are not really video listings at all.
Who you are choosing for
Someone browsing with no prior interest in any of these places, no local connection, and nothing invested. They are not looking for anything in particular. They will click on what looks worth the next few minutes of their life.
Watchable, not important. The most significant video here is probably not the most watchable one, and you must not confuse the two. If your top five is a list of what deserves attention rather than what would get it, you have answered a different question.
Rules
Exactly five in each list. Most watchable best first, least watchable worst last. No id in both lists. Echo each id exactly as written, and copy each title from the batch as it stands.
Judge the title as it stands. Several contain ', " or & — that is encoding damage, not the uploader's writing. Ignore it. Do not comment on it, do not fix it, and do not mark a title down for it.
The reason is one sentence, and it is about the video, not the title. Say what a viewer would actually get. "Promises a specific abandoned place and names it" is a reason; "well-optimised title with strong keywords" is not. Never restate the title back to me — it is already sitting there at the front of the line.
— an engagement plea with no subject, keyword soup with no sentence, an advert, a phone number. A dull recording of a real event, correctly titled for the small audience it is for, is a legitimate listing and belongs nowhere near this list. If you think nothing qualifies, keep the heading and write under it.
Answers
Sorted by cost. Open a row to load its answer.
The third list is for rows that are not video listings at all
Boring is not spam.
None.
Say when it is close. If two rows are genuinely level, order them either way and say so in the reason. If a call is arguable — a listing that is an advert but also something people watch for pleasure — that judgement is worth more to me stated than hidden.
You may use what you know. If a title is dull on its face but you recognise the person, the place or the event in it, that is exactly the kind of thing I need told.
Output
Markdown, nothing else — no preamble, no commentary afterwards, no code fence around the answer. Exactly these three headings, in this order, one row per line:
## Most watchable
1. **title copied from the batch** — [id](https://www.youtube.com/watch?v=id) — one sentence.
2. …
## Least watchable
1. **title copied from the batch** — [id](https://www.youtube.com/watch?v=id) — one sentence.
2. …
## Not listings at all
- **title copied from the batch** — [id](https://www.youtube.com/watch?v=id)
Every link is https://www.youtube.com/watch?v= followed by the id exactly as written — no extra parameters, no shortened form, no tracking. The id is the link text as well, so I can check it against the batch and open the video from the same place. Nothing in these titles needs escaping: put the title between the ** as it stands and leave it alone. One line per row, and no line breaks inside a row.
The batch
Each entry is an id line, then the title verbatim on the next line.
[1] id=V_1UW2AS9Bs
David Lynch's Weather Report for 12/14/20
