someone with 10 BP has the same vote as 10,000 BP?
This sounds pretty silly when reduced to two users, but I'm sure this approach is put into perspective by a larger set of users. But, to stay with the answer to your question: Yes.
Yes, because it reflects opinions, not "poor" or "rich".
I'll try to bring an example - also highly simplified:
Article A
This is a very good article, 99 users like it, it has 99 votes.
The users are all smart people, but they are poor. They all have only 10 VP.
--> 990
Article B
Only one person likes this article (maybe it's even a selfvote). This user is rich, VP 10,000.
--> 10.000
Today, a single rich user determines which article is the better one: Article B, although only one person (persons, not money, should have an opinion) likes it.
Now we first collect all VP in a pool. --> 10.990
At the end article A gets 99% (10.880) of it, article B 1% (110). Already article A is the better one, which represents the opinion of many.
The result would be the same, if the rich user votes for article A, a 10 VP user votes for article B. A is considered the better one by the majority of thinking users.
Of course, the CR of the users should be taken into account, which is especially important for the rich user. The CR are also calculated from the total pool. That is, the rich user gets 25% of the CR payout, even if he did not vote for the good article A. He has paid into the pool, so he should get his reward for it. The fact that article B simply doesn't deserve more because it's bad in the eyes of most users, even if the rich user has voted, is something the rich user no longer has to worry about.
This is my idea in a nutshell. Maybe a discussion can actually be built on this. I must confess, however, due to time constraints I can not participate very much in it. Constructiveness is important to me, we should not peck each other's eyes out in this debate. That's not worth the investor-blogger dilemma, we are all human. One is smarter, one is dumber, one is greedier, one is too altruistic, one is black, one is white... we need to find a common consensus, don't all have to love each other. But if quality is desired in this community, it must not be defined by poor or rich.
Thank you for being a part of this discussion. It speaks to your character the proposal you submit given your amount of stake.
Unfortunately, I see not one but three issues with this.
The first is the one rycharde mentions. I'll use the whale from Steem/Hive as an example of this to an extreme. He had 100's of known accounts, some speculate thousands. He deployed them often with bots. It would be nothing for the likes of those like him to create thousands of accounts and just upvote their own posts and win the largest slice of the pie.
The second problem I see with this is it would discourage people from holding stake. If the return isn't based on stake, then most would only stake 1 Blurt and get the same return as anyone who staked a lot. This would be even more certain if the savings were rewarded as rycharde mentions with two pools, as everyone would stake 1 Blurt and then throw the rest in savings, if they kept anything here at all.
The third flaw I see with this is it would see the folks here to extract as much as possible all flocking to the same posts that have a lot of votes to maximize the return on the vote. I don't think this would encourage quality but would create for lack of a better term return popularity. Hive i believe did something similar, only it was for posts that achieved more than 20.00 in votes got a higher slice of the pool if I remember right.
It seems to me that the one vote same pull approach would make this no longer a proof of stake system.
This already exists (sort of). Its Reddit, Twitter, and many other social sites. It seems by default that once one introduces money into the equation it changes the way many interact with one another. This is clear by the lack of comments for example, even clearer by many of the crap posts made that share nothing of spirit or the person posting it as well. It's not uncommon on Twitter for example to see threads with thousands of comments, and no one got paid a thing. Here (and the other graphene chains), many posts are barren of comments.
That was part of my venting in the post. If more people would focus on being real, and seeing one another I do believe most of this would work itself out. So the question for a fix if it can be done is
How can we get people to let their hair down. Share their real passions, the ache in their hearts, the joys that made it all worthwhile despite that pain. Share from their soul and let us know one another. To laugh with one another whether it be great news or sharing ones folly and the rest of us shake our heads laughing as we remember falling victim to the same mistakes and tragedy.
I can say this. I've interacted with many fine folks in my three years on these sites. Due to the low participation from my part of the world here, I've had to become more proactive at looking at communities such as the German and Spanish ones. I have been both touched and amazed at the overall cohesion and critical presentations I witness. Amidst all the crap posts, the selfish maneuvers I see those like yourself, Afrog, rycharde and so many more. I content myself that these are gifts brought to me through this. That there is also more than simply monetary rewards here. And unlike the other sites, we have good intentions from the top promoting what it is many of us yearn for. Quality, abundance and friendship. I see this in many powerful accounts here. While mine isn't quite so powerful, I feel good with my position and that given time it will one day be powerful as well.
Together folks like ourselves can help foster a place of quality, by subsidizing that which we want to see more of. This principle always works, and we have structures being put in place that aids us in this.
Thank you again for weighing in. I appreciate the quality you bring to this community, combining both wit and humor. A winning combination. Now, here in this comment section you display compassion and a sense of sharing for others. Amidst all of the greed and lack of quality those like yourself shine bright. Thank you.
Wow, once again a very deep, meaty comment from you - thank you.
First of all, I want to apologize to you for not responding directly to your text, but "jumping" on Rycharde. Your text is very good, it contains so many points worthy of comment. But - as you already know me - that would also be very labor intensive. So Rychard's cue came just in time for me, I was able to latch onto it to say what I have to say.
Further, I want to thank you very much for your compliments. I really appreciate your perception of me. Yes, that is indeed a small gift in this virtual world where it is becoming more and more obvious how many are just chasing the mamon, losing sight of human interaction.
But now to the matter at hand.
"Problem" 1:
I didn't even have this huge abuse with multiple (thousands?) accounts right in front of my eyes anymore. This is - like any other kind of abuse - possible in both systems.
So it is necessary to set up rules or definitions of abuse. Constructively, sanctions must then be considered (downvotes are not a way, because here again everyone, especially "the big ones", can play out their might). This was probably already discussed among the Witnesses, Jacob wrote it here:
https://blurt.world/blurt/@jacobgadikian/hf4
"Problem" 2:
I think this could regulate itself through the principle of "give and take" and the experience of users (at some point you know everyone).
E.g. I see some people on Steem (I'm still active there) who have downpowered their entire stake (even "bragged" about having put everything into Hive or Blurt) but still post. Sorry, they don't get a vote from me, because it's clear that they 1. can't give anything back and 2. have no interest in interaction.
Besides, everyone has an interest in receiving CR, after all. These equal zero if you always have only 1 VP active.
"Problem" 3:
No, I don't think so, if we make sure that the CR for the voter depends on his effort and not on how many others vote. Then the CR will stay the same no matter where you vote.
That of course also contributions are voted by the "best friend", although they are not so good, is clear, for that we are humans who act here. But this behavior will be put into perspective as long as there are people among us who are looking for good content out of genuine interest and are willing to move out of their own circles to do so.
Dear @practicalthought, I am exhausted and close the "novel" here. Nevertheless, it is always a pleasure for me to exchange ideas with you. Only in the dialogue thoughts come into the roll, which should result in a common result to the satisfaction of most.
Warm regards,
Chriddi
All of this is predicated upon one thing: 1 user = 1 account.
So, either design a chain where that is enforced, or design rules where that is not an assumption. Our experiment is based on the latter.
In real life, we do not (yet) demand everybody's full credentials before every interaction; we assume they do not possess a body-double. Here, we assume everybody does :-)
We had a lot of chats about this theme before on Blurt. I can remember even Jacob and Megadrive has been involved and there where plans already to solve the multiaccount problem.
But as it was said on conutless other places: We can't envolve Graphene very well. It seems, we have to wait for the development of a chain on Cosmos hub. It is fact that Graphene has no future because nobody knows the codebase very well.
But now is the time to talk about the layout of a better game on Cosmos. We should not stop to think about how it will work better.
I'm glad they are continuing with Blurt as they move forward with Splash. Due to intrusive government regulations, it appears that for those of us in the U.S. Splash will be (more of) a tax nightmare. Under the current system, our pull from the pool is 50% given to those we vote for. So we only get half that tax liability. Under the Splash model which will be modeled after Whaleshares, we will be responsible for 100% of the tax liability.
Thank you for not only being a catalyst for this post, but for weighing in. You are one of the accounts I look forward to seeing a post by. The combination of intellect and passion you possess and share leaves me grateful.
This right here. The whale that took joy in destroying accounts at the other chains had perhaps thousands of accounts.
Strange, isn't it?
1 user = 1 account. Yes. For voting, yes. Of course some people sometimes need more accounts (for example for different themes).
I think, it's all a matter of definition. People need rules, everything else is unsettling. We have known this at least since the Old Testament.
Besides I'm lazy now... ;-)
https://blurtter.com/blurtlife/@chriddi/qqx8ug
check out the Code of Hammurabi
https://en.wikipedia.org/wiki/Code_of_Hammurabi
"People need rules". Like children. What better trap is there than to encode those rules? ;-)
Every set of rules will have some optimal behaviours, eg, you can be a crook if you're big enough to pay for the law to be friendly.
Once you add money, those optimal behaviours may change to maximise income. Then the game changes. Most people seem to expect the old game with the same rules - not happening. This is still a new science - and far harder than being a mere economist.
Thank you. I know the code - I am a teacher for mentally disabled children.
These children partly can't understand why it is shabby to cut off the very biggest piece of cake, because they are hungry after all and the needs of others don't interest them at that moment. So you have to support them directly with the action regulating.
Is it shabby to farm with 1000 accounts or the biggest stake? It is such a smart idea to make gold from straw very quickly - what do I care about the needs of others?... ;-)
... So you have to support them directly with the action regulating. About coding, great trap.
In the discussion about determination of codes, we must basically assume that not everyone has the same moral sense and the same cognitive prerequisites, which is why nothing at all will regulate itself all by itself.
Your cake example reminds me of a basic, yet interesting, optimisation problem.
How do you get 2 children to share fairly a cake?
As we know, people will look at the tiniest difference to get the larger share.
The solution is that you let one child cut the cake into two parts, then let the other child have first pick of their slice.
The solution is sooo simple... ;-)
However, it is (still?) a mystery to me how we can break this metaphor down to our core problem with many, many children. And the coding of it even more so! But in the end this is fortunately the task of competent mathematicians and computer scientists.
Exactly! The multi-person cake-cutting problem is, however, not so simple.
https://en.wikipedia.org/wiki/Fair_cake-cutting
It is one of many algos I am looking at - thinking takes time tho ;-)
What the h...! Chapeau if you manage to tailor this insane challenge as an algorithm to our concerns!
Maybe step by step, trial and error? Rome was not created in one day... ;-)
Happy Easter,
warm regards,
Chriddi
Very good spoken @chriddi! That's the fact I always tried to say but didn't found the right expression.
Thank you very much.
I hope it's become clear that this is just a discussion starter, because of course in any system there are many ways to abuse it. And if the possibility exists, there are always exploiting users. But then it is a question of defining what exactly is meant by abuse. Only then can this be sanctioned (e.g. no rewards after warning - downvotes are not the right means in my eyes either).
Be that as it may, I am convinced that the 1-vote=1-vote approach is the right one, especially since everyone can see that it is not working the way it is now. Together, we have to discuss the pros and cons. And then we have to try out individual elements (not all at once, that can't go well). Then constantly evaluate them and, if necessary, rethink or develop them further. To reach a goal, you also have to take a step back to make the right turn at the fork in the road.
I am excited and happy to contribute whenever I can. What I don't want is to go through age-old discussions over and over again, only to see that nothing happens and old habits are retained without reflection. Because then we won't need a third, fourth, twentieth chain.