He doesn't need to personally nit pick every case, or really any, but by setting a comparatively very objective standard for leaving a negative rating the VAST majority of conflict is simply avoided.
There will be no hindrance of stopping scammers, but in fact it will be easier to spot them as they don't blend in in a sea of red. Also people will feel more comfortable contributing to a community where they don't have to worry about everything being taken away over some petty squabble at any instant.
I agree with this, the more people adopt this system the more it will balance itself. Initially it might appear a Cats vs Dogs drama, the more people adding their trusts/distrusts the more our personal "reputation" would form based on objective observations from the community. I might be able to plot with a bunch of people against Cats for 1 month, maybe 2/3/6 months, but the real value of cats and dogs will come out overtime. Personally I woudnt even give power to any DT level indeed:
the sum of someone Trust would then come out based on the entire community feedback. Scary? Behave then...
I don't think that would work. Let's say I piss off a scammer by sending negative feedback, what would stop him from creating/using a bunch of alt-accounts to wreck my trust level? I like the idea of a non-DT system since the latest DT changes have caused a lot of childish behaviour.
I don't understand how someone can wreck your trust level using alts.
Feedback left by nobodies is disregarded in counting your points (you can look at my own trust page for examples).
If someone manages to create enough sockpuppets, gives them voting rights, and unlegitimately include people in DT1 just so he can tag you in red, without getting blacklisted in the process, not only we could consider the trust system flawed, but also ask what in the hell did you do to piss off this person.
That is because of the current trust system. I believe Gunthar was trying to address a system without any form of DT, just community feedback (similar to eBay).