RE: The Main Reason I (and my witness @l0k1 thereby) are Anti-Self-Voting
Thanks for the mention. I appreciate that. But it should be noted that I'm not for any blanket policing of self-voting and taking it a step further to police voting on possible alt accounts.
If there's a specific case of abuse, then by all means, look into it to determine if there's an actual problem. But blanket countering, shaming, or digging for "links" between accounts is counterproductive and can do more harm than good.
Nobody can determine whether a vote on another account is for a friend or family member. If someone votes for the same person every day, does that mean they are a friend, relative, or a sock puppet? Are we just to assume the worst?
And should all self-voting be opposed? Is it an abuse of the system to vote on your one post per week? On one or two comments per day? What if you're upvoting 100 other posts every day?
I prefer handling this at the protocol level as much as possible. Take human subjectivity out of it and create the abuse-mitigation that is "fair" and can scale to millions of accounts instead of leaving it in the hands of a few people and their bots.
It is not possible to advocate a position, and still have the respect of the community, while condoning the violation of that position.
It is precisely the elimination of the 'low hanging fruit' that is addressed and targeted by this current project. I think it says a rather large amount about it being the correct position that the most egregious self voter, is like a mosquito with zero reputation kick to their giant voting power, and so quick to act without considering more than 5 minutes into the future about what happens next.
We discussed the option of such as a cumulative or windowed averaging that allows say, 10%, 20% or some arbitrary line between. If the protocol needs self voting, a rational, 100% applicable defence must exist, or you create a slippery slope where the very same rationale gives you 50,000 different possibilities and no way to agree.
But if, instead, we narrow our criteria so that if it cannot be implemented by a bot, but rather, a simple, universal rule, such as 'may not directly vote upon own content', you see that 'bypassing it with alts' does not violate this simple, automatic rule!
A much simpler ruleset than the cumulative average arbitrary, arbitration-dependent human-dependent and impossible to standardise way of mitigating bad behaviour.
I don't even have to search far to show you an example of how self upvoting to 'alter the sequence of comments' cannot be a justification here: https://steemit.com/steemit/@holoz0r/steemit-licence-plates-are-finally-here
'Lol awesome', voted up to nearly $12?
When I read the part about detecting "linked accounts" I was also concerned. I have a follower who I don't even follow who consistently upvotes every single one of my posts within the first 10 minutes. It's not at all a collusion, I don't even know who he is beyond a username.