This might be a bit convoluted. I saw a statement: posts that are not in consensus get very few likes, while posts that are in consensus can have millions of views. So, high-liked posts aren't worth reading. That sounds quite right. Then I took a look at my own data, and I couldn't help but feel a bit uneasy. The tweet with the highest views in the past few days had 260,000 views and 449 likes. By this logic, that tweet is also "in consensus" and not worth a look. But that tweet talked about how the Atour pillow made my nose bleed and my experience with OpenClaw... what kind of consensus is that? So I think this framework might be missing a layer. The real situation is probably like this: non-consensus content indeed gets ignored when it is first posted. But if it is correct, after a while it will become consensus, and then the likes will come in. When BTC was at $3,000, calling for a buy was non-consensus. Once it rises to $126,000, everyone says they knew it would go up, and it becomes consensus. Non-consensus is not an identity; it is a time window. All non-consensus, if correct, ultimately becomes consensus. So using the number of likes to judge whether a piece of content is in consensus or not might be asking the wrong question from the start. What you see is not consensus vs. non-consensus; what you see is how far along this viewpoint has come. What you should really be wary of is not the high likes, but when you like something, not being able to tell if you saw it at the first step or at the last step.