Sorry, the analogy was indeed too extreme.
But the point remains: when you think content is good, you upvote. If you think content is not good, you don't upvote. But you don't flag, unless it's a clear abuse (scam, phishing, etc.)
Flagging posts that you think have higher reward than look ok to you means that you are questioning the judgement and defying either the people who upvoted ("you think this is good content but I think you are wrong, this is bad content") or the very system of steemit (with bots and all). If the Steemit system leads to absurd rewards because of bid bots and other abuse then the correct approach is to tackle the issues with the system, not each and every post that exploits a broken system
I personally question the judgement made explicit in page 15 of the whitepaper (see above) as I believe that even if completely eliminating abuse is never possible, it definitely MUST be an explicit goal of any sane and self-respecting society. But I'm not starting my own militia war by downvoting abusive posts. It's profoundly destructive to the very system we want to see thriving
Yes that's exactly right and people should be (and must be) skeptical of each other. In a decentralized system the answer to "Quis custodiet ipsos custodes?" (Who Guards the Guards?) is "each other". Our voting decisions are all open to scrutiny and disagreement by each other. That is the only way this system can work.
I agree with that, but downvotes are also part of the system. If the system can work (at least better) with a new and better emphasis on downvotes, and new and better ways of downvoting, then everything is just fine. The goal is a system that works well, not a system that works without effective and intelligent use of downvotes to express and mediate disagreement between voters, whenever needed (not just in specific cases decided by you).
I believe this discussion is important and I like a lot that we can have it. As answered on another comment, I am preparing a standalone dedicated post on this topic.