RE: Exploring the Practice of Self Voting & Upvote Rings.
I dunno. We decide what's okay and make the math follow. My instinct is the penalty curve should start gentle, but get more aggressive.
The problem is the unintended consequence of that might be an increase overall self-upvoting since the reward algorithm basically would be saying "one or two self upvotes are fine". Right now we have a lot of social pressure to not self upvote too often: that social pressure has less room to justify being more restrictive if the algorithm is more finetuned.
On leasing delegation: I lease out SP to others via minnowbooster. The benefits are not just the math, but often the increased social benefits/connections people can make with that SP. So, I'll agree that the raw numbers are hard to justify the returns, but I think more SP brings more chances to make wider connections too. That brings long-term gains. Mind you, I do notice a large number of upvote bots and trail-accounts leasing SP. I think most of them are anticipating their own future growth.
There is certainly no doubt this is a hot topic for conversation! You've got a good point with putting in the algorithm being a statement that self upvoting is acceptable to a point. But is that a bad thing?
I'm thinking out loud here, but are we approaching this from the wrong angle?
What if instead of trying to penalize people for voting for themselves, we make it more rewarding to curate new content? Some sort of additional reward system where you earn an extra percentage for upvoting someone who you have never upvoted before. Of course then some people would just create a bot that randomly upvotes new users without any actual viewing of the content.
So many things to consider!