Every card says who found it
Attribution comes from the Room membership, not from anything the sender typed, so it stays right the moment more than one person is helping.
A helper firing findings at you is only useful if you can read them without stopping talking. Everything about the queue is built around that one constraint: you are mid-sentence, looking away from the screen, and you have about half a second per card.
Not the title, not the thumbnail, not the source. The whole reason you asked was to find out whether this is worth two hundred dollars or two, so that is the element set largest and boldest, with the source named underneath it.
Attribution comes from the Room membership, not from anything the sender typed, so it stays right the moment more than one person is helping.
Junk comes in visually muted with a marker on it, so a worthless result costs you a glance instead of a decision.
The thing your audience is looking at right now is the first card in the list, which is also how you avoid airing the same item twice.
Air it or dismiss it. A queue you have to think about is a queue you stop reading, so there is nothing else on the card to consider.
Once something is up, its actions are replaced by an on-display marker. You cannot fumble a second tap onto the thing that is already showing.
This is written on the queue itself, and it holds where it matters rather than only in the interface: putting something in front of your audience is your action to take.
It is a seller surface and it is never sent to the audience display. Your viewers see what you chose to show them, not the pile you chose from.
Suggestions live for the length of the broadcast. Tonight’s pile does not greet you at the start of tomorrow’s show.