While it may seem that under DPOS or similar protocols that the crowd is wise and well informed, instead what we actually see in practice is very low participation rates.
You could change a couple words and you'd be talking about democracy itself. This is an old problem. Democracy needs ideally informed citizens, highly educated, in order to function properly. Most people would rather choose one thing and do it well, rather than learn about every issue that's up for a vote. I don't learn about medicine before going to the doctor: that's why I pay him, so I won't have to learn what he learned. Being a politician is like any other practice, it's amenable to argument and science and rational thinking. So I think it should be handled like other practices: egregious violations of the public interest in favor of personal interest should be punished, for instance, and accidental mistakes that can't be helped must be tolerated as a necessary evil bug in an imperfect system.
This is why I think EOS will fail. We cannot have informed citizens with the current technology so in the end there is technocracy and all the problems we see today in Steemit will only become progressively worse under EOS.
Tauchain seeks to resolve this problem by scaling discussion so that collaborative decisions can be made. In other words, I see EOS as having the same bottleneck that we see in traditional institutions because DPOS which to my understanding is the basis behind EOS and it has those flaws. I think these problems can be resolved but it requires using algorithms, machines, AI, and to scale decision making in ways which go far beyond what DPOS currently offers.
I will never be educated enough. I can only read so many books in the time I have due to the scarcity of my attention. The only way to get beyond the scarcity of my attention is to somehow encode my preferences and values in such a way that machine intelligence can increasingly focus my attention on what is in my interest.
DPOS in my opinion will lead to politicians and politics. Also the transparency agenda behind EOS will promote maximum conformity possibly to the point where people cannot stray from orthodoxy to even express their true opinions. This could have very negative long term consequences but it's for EOS to experiment and discover for the cryptospace what those consequences will be.
My take on it is we need to scale our thinking and relieve our bottlenecks. In my opinion the best way to create a better life and a better world is to focus on improving ourselves individually. There is no institution which can help us if we don't scale ourselves, as in our thinking, our ability to be moral, our ability to make big decisions. Global sentiment is very important but without an automated means of factoring it in, we cannot know how our decisions could impact other people.
Accidental mistakes are not well tolerated in society though even if you believe it should be. The data does not show that just because something is an accident that global sentiment is not going to turn negative toward the entity which did the accident. Accidents also scale up as the stakes scale up, and as EOS becomes more scalable, it becomes a global thing with global implications and world level stakes.
Interesting thoughts. I'll keep in mind.
And I agree with the moral sentiments of the penultimate paragraph.