24 ms·

Not quite. The social signal that a large number of people disapprove of content is what they are looking to hide. It isn't just important for us to know if t

Not quite. The social signal that a large number of people disapprove of content is what they are looking to hide. It isn't just important for us to know if the majority of viewers think a fortnite walkthrough is shitty. We also need to know when a politician is lying to us. Do most people think this is bullshit? That is what the like/dislike ratio communicates so succinctly.

pcwalton 5y ago
An enormous number of public opinion polls broadly show acceptance of the COVID vaccine, including mandates [1]. Yet every video demonstrating the efficacy of the vaccine that makes it to the front page of YouTube is spammed with dislikes (80% or more). It's statistically impossible for the polls to be off by so much that YouTube dislikes are representative of actual public opinion. This is why it is necessary to take antivaxxers' toys away. [1]: https://news.gallup.com/poll/354506/update-american-public-opinion-vaccination-requirements.aspx https://news.gallup.com/poll/354506/update-american-public-o...
DSMan195276 5y ago
The idea that the dislike ratio indicates any of this is kinda silly. At best, a "large number" or "majority of viewers" isn't even close to a 10th of those who have watched it, the rest either liked the video or didn't care enough to leave a like/dislike. And as is pretty commonly known, people are much more likely to leave feedback if they're unhappy than happy, so clearly the numbers are going to be at best unreliable and likely very tilted in the dislike direction.
Quarrelsome 5y ago
I don't think I've ever used like/dislike as any sort of indicator beyond seeing that there's some drama going on. Its a very unreliable metric. If you want discourse about a video and its propagandiness then IMHO reddit would be a better place to get that because it has conversational features, unlike other social media spaces.

pcwalton 5y ago

An enormous number of public opinion polls broadly show acceptance of the COVID vaccine, including mandates [1]. Yet every video demonstrating the efficacy of the vaccine that makes it to the front page of YouTube is spammed with dislikes (80% or more). It's statistically impossible for the polls to be off by so much that YouTube dislikes are representative of actual public opinion. This is why it is necessary to take antivaxxers' toys away. [1]: https://news.gallup.com/poll/354506/update-american-public-opinion-vaccination-requirements.aspx https://news.gallup.com/poll/354506/update-american-public-o...

DSMan195276 5y ago

The idea that the dislike ratio indicates any of this is kinda silly. At best, a "large number" or "majority of viewers" isn't even close to a 10th of those who have watched it, the rest either liked the video or didn't care enough to leave a like/dislike. And as is pretty commonly known, people are much more likely to leave feedback if they're unhappy than happy, so clearly the numbers are going to be at best unreliable and likely very tilted in the dislike direction.

Quarrelsome 5y ago

I don't think I've ever used like/dislike as any sort of indicator beyond seeing that there's some drama going on. Its a very unreliable metric. If you want discourse about a video and its propagandiness then IMHO reddit would be a better place to get that because it has conversational features, unlike other social media spaces.