The XianYu use case looks very practical. Randomly identifying few members of the network with certain credibility to resolve disputes makes sense.
The selection procedure for arbitrators should be handled by the network without any interference by any third party. A creadibility parameter should be implemented and followed by the network to maintain transparency.
It's a very broad question and hard to cover within the span of this thread. But one thing is certain, that it can be achieved only through interference free technology which can ensure utmost transparency and integrity. A lot more brainstorming is needed to reach a finalized concept execution method.
Great answers!
1) It is important to add, however, that on XianYu, only active users of the platform are eligible to become Jurors. This limits the potential of totally unbiased and decentralized decision making, as it is much fairer to include people from outside and give them the chance to judge on certain affairs, if they are interested to do so. With Oath, we're sharing communities, so not only EOS users could rule cases on the EOS chain but a NEM- or Cardano user could do so as well. This makes more sense in our opinion and also brings along advantages for all partnering platforms in our ecosystem.
2) Decentralized arbitration is always kind of the "Third Party". Whether it's the platforms users who are not affected by the cases' results, or a much broader pool of jurors. What Oath provides is the Jury and the verdict. Execution, user-behavior, user experience and technological standpoints will not be affected nor interfered.
3) I wanted to imply here, that code is always limited by itself. Human rationality, or whatever the majority thinks is true, truly covers each scenario possible and thus extends the usability + security of smart contracts / technologies drastically.