Downvoting and Reputation - My thoughts on what works and what doesn't

There is a lot of discussion about this subject, and after reading some comment threads I decided it was time for me to weigh in with my thoughts developed over quite some time thinking about it. One of the elements of a previous post I made that I forgot to mention, was a reputation system for the decentralised universal system I devised, which I called The Agora Network.

I think that there should not be a monetary penalty for downvotes, but rather that these downvotes should contribute to a personal reputation network scoring system. Instead of a universal system-wide reputation, the system calculates the reputation and therefore the height to which others posts get to on your feed to be modulated by this network proximity algorithm. Thus, if you upvote someone, this raises their score only in yours, but then also is factored into whoever up or downvotes you - if they downvote you, you invert the values as to how their score is calculated as to who you upvote, because obviously if they don't like you, to some extent they will not like who you like either. If you follow someone, this should also have an impact on how other peoples' scores are rated. Maybe something like an automatic upranking of them and everything they vote up.

By using a network proximity algorithm to modulate reputation scores, you can avoid some very serious gaming attacks that can be done, especially by sock puppets, also known as Sybil attacks. When you receive votes, these votes should be locked in after a cooling off period, and downvotes do not alter rewards, but only how high you can appear on someone else's feeds. If downvotes take money away from you, then they amount to theft. Rather, they should only limit your exposure to this other person.

I base these ideas on what I found was most effective on Facebook when I encountered unpleasant people I didn't like - when I simply blocked them, my mood improved and I wasted less time on unproductive arguments. Downvotes are no different to flames on other people. They just cause antagonism and unnecessary stress. If you simply don't see their posts, then your mood is improved. So it's not just about your own personal state of mind, it also affects the network because fights are ugly and unpleasant and effectively downgrade the whole network. This is why places like 4chan, Facebook, Twitter and Reddit are increasingly becoming seen as cyber-ghettos, because this kind of behaviour is lowest common denominator, pulling the reputation of the whole network down.

If Steemit adopted this kind of policy, people would be incentivised more to participate because negative interactions would result in increasing the network distance between you and the others who you don't like, or don't like you. I know that there is some people who think that this creates 'bubbles', but please explain to me exactly what productive outcome there is from open antagonism?