I disagree. There are a lot of videos that I find just “meh”. I might not regret watching them, but wouldn’t recommend them nor watch again.
Then there is content which I find pretty good/bad but not extremely good or bad. For such cases a more nuanced scale is better.
For other users this might be less informative, since they will be seeing just the average anyway and can therefore only determine general perception; except if the distribution is also made available.
But for a personalized recommendation system I think a nuanced scale can work better.
From a content creators perspective one can also evaluate better whether there is room for improvement and by “how much”, in case one is interested in such.
I disagree. There are a lot of videos that I find just “meh”. I might not regret watching them, but wouldn’t recommend them nor watch again.
Then there is content which I find pretty good/bad but not extremely good or bad. For such cases a more nuanced scale is better.
For other users this might be less informative, since they will be seeing just the average anyway and can therefore only determine general perception; except if the distribution is also made available.
But for a personalized recommendation system I think a nuanced scale can work better.
From a content creators perspective one can also evaluate better whether there is room for improvement and by “how much”, in case one is interested in such.
What about a trinary system? Like, dislike, and meh?
Quaternary Like, dislike, meh, “what in tarnation?”
I like the quinternary system like a strong dislike, a dislike, a meh, a like, a love.