You are viewing a single comment's thread from:
RE: Proposal: Change Up and Downvotes to bids
basically the idea is that your vote is still worth a fixed amount, but it has a tendency into a certain direction. (Around the value you set for the post). To make sure its not over or under.
Of course you could setup a custom frontend with a bot which makes sure your vote is like you want it. But undoing unvote to upvote etc costs a) RC and b) voting power
Your vote does not have a fixed value. The value of your vote varies by three different metrics: the first is the SP you possess, the second is the weight of the available VP you assign it, and the third is how depleted your VP is from casting votes,, or becomes during the vote window as you cast votes thereafter. Another issue is that some folks have even more variables they consider prior to voting, such as curation rewards. Some people only vote because they get paid. This is already a nebulous target, because it depends on who besides you votes how hard and when they do so.
I'm not even going into circle jerks, purchased votes, or promoted posts, and how this might complicate them.
Most people have but little idea of how to calculate the third variable, and even for those that do know how, the work to do so if not often undertaken. I don't undertake it at all, and just always vote at 25%, because I have discovered that if I vote at that weight, I generally remain within VP that can recharge in ~1 day, which maximizes the VP I have to vote with.
Finally, I have to agree with @lextenebris on one thing: voting is already too complex (witness my rule of thumb regarding vote weights) and this would make it more complex. I do not agree that people don't vote at all due to lack of robust comprehension. I observe that even idiots express opinions, or vote, despite demonstrable complete lack of comprehension of relevant issues and consequences, so reckon very few people actually even vote less when incompetent to do so rationally and well.
I was simplifying it for the sake of simplification in the comment, I'm well aware that the value varies based on many things (VP, reward pool, curation, SP, etc).
But, for sake of simplifying the implementation, all this "could" stay the same, the only thing changing would be anchoring the value around a certain value the person assigned to the post.
The algorithms that compute the value of your vote on a post do not generalize that value. They set it precisely. Since the target value you assign a post is attained, or not attained, depending on all the votes cast on a post, the calculations become impossible as the VP of voters is depleted or recharged by voting.
Multiple variables that are calculated exponentially increase the complexity of the calculations to account for each other, and this level of complexity is unattainable with our present tools.
But the same problem exists at the moment too.
No. Your proposal adds a layer of exponential increase in calculation difficulty. The decay of VP is calculable since there's only one factor to calculate. By introducing the target payout of the post, you multiply that calculation by all the VP applied to the post. This is incalculable.
The user still applies a certain VP to the post. If less is used he gets it back after 7 days, end of complexity.
You appear not to math well. There is no way to ascertain the result you seek.
First of all, I think you don't know the fine rules of discourse too well, because that's not how you discuss.
Second, I am quite decent with math and code.
If someone can with a specific quantity of VP "bid" on the value of a post to guarantee it. And in the end, let's say the post is highly over or undervalued all this VP will be used, no adjustments needed. If the value of the post is exactly what he voted for, at the end of the 7th day, the chain calculates this, and refunds him the unused VP. End of story.