RE: Voting Abuse and Ineffective Curation: A proposal for blockchain-level change
I mentioned this elsewhere, but figured I'd take a moment to write a reply about it as well.
I'd definitely support changes to the 30m window and am all for exploring ways to maybe create a level playing field in the first 5 minutes.
What I don't think I'd support is just hardcoding the system back to 50/50. I'd much rather see those values turn into a configurable amount that can be decided by the platform it's being posted on. On a site like Steemit - being about curation and content discovery (both needing work), 50/50 might be a good number, but it would be a hinderance to platforms like chainBB where curation is effectively meaningless and votes are only an expression of wanting to reward someone.
A configurable amount (set inside the comment_options operation for a post at the time of creation) would also let the community experiment with every imaginable range of values, from 0% to 100% for both authors and curators, and the market will decide what to vote for. Some platforms may flounder if they choose the wrong values, while others may succeed by setting numbers we would have never expected.
I hope we can get some action around this (and a number of other issues that have nagged us for ages) early in 2018 - there's a lot to do still to make this platform rock solid.
I think this is a really important concern that @jesta mentioned. For DSound, in my experience as the developer of it, and by analysis of the last 4 months of it's live alpha, it is also not interesting to have 50/50 back. Maybe in DSound it would be even better to have it entirely as 100/0 in favor of the author, as the listener is already having the benefit of listening to the track... And that would mean more author rewards for the musician which is who technically needs them in the first place, as the curators are already having their fair share as free listeners. So, having this property customisable would benefit all and it should be only settable in the initial post and not changeable, to avoid different apps to mess with it...
Really happy you chimed in as someone else building a platform on Steem.
These types of considerations have to be accounted for, if Steem is for all platforms, and not just for the steemit.com website. What works for one site might be detrimental to another.
Changing reward budgets per post could be problematic.
First, there is the issue of user education. Currently, users know that they get paid for posting and curating. The math is publicly available in the source code, but doesn't need to be advertised to everyone for them to understand how to use Steem. If we added such an option then it would need to be reflected in the UI so users knew how much of the rewards they could earn.
Second, while this might start as a way for different interfaces to incentivize different content, it would not stay that way for long. Savvy users would see that a configurable percentage creates a market for potential voters. If there are identical posts, one with 25% curation rewards and the other with 75%, which one will most people vote on? The answer is both, but at some ratio defined by the curation rewards curve.
This would turn curation into a market. How much of your rewards are you willing to give to curators for votes? This a slight change from the intended mentality today that curation is budgeted for rather than authors paying curators for their votes. The difference is subtle, but the perception of ownership over tokens is a powerful one. This is why many users get upset when they are down voted. The rewards are not yours until payout. But that doesn't prevent users from feeling like the rewards are theirs as soon as they receive an upvote.
It's powerful but is it beneficial or detrimental? I think it's the latter, the general entitlement mentality is something we want to get rid of.
It's already a market. How much SBD are you willing to pay to minnowbooster for votes?
Configurable rewards will encourage real curation instead of vote buying/selling.
To provide a response to each point:
Ackshually... It is.
A society is more than an economy. Given the existence of folks that literally care about nothing else but money, creating a market for free speech inevitably causes freedom to speak to become treated as a commodity - and this demeans freedom itself. Monetizing freedom results to some degree in slavery for profit.
OTOH, it's preferable to being demonetized on ideological grounds, or being forced to listen to the speeches of Marvin Bush at gunpoint, until you pray for that sweet, sweet release only death can provide.
The market is unavoidable, so making it as free as possible is the best possible option for human freedom.
tl;dr Nope, it's the best answer.
Dear @vandeberg ;
I need any help to stop @grumpycat hurting innocent people.
We have to show that Steemit is bigger than any bully who is trying to impose his own rules by using his high SP on innocent people.
The post below is the summary of the situation :
https://steemit.com/life/@firedream/stop-the-grumpycat
Thank you for any help to stop the actions of @grumpycat.
Best Regards.
FD.
A configurable amount may end up with a race to the bottom. Big players may decide not to vote on certain posts based on this parameter and the content itself will not drive curation (not that it does right now but I feel that this will create the same problem but in the opposite direction).
They might decide not to vote on something - but not everyone cares about curation. I know I'd vote on things regardless of the percentage.
Would it make sense to bring votes from others (by sending them notifications of recommended topics) by referring them to mentions?
Example...
I like a specific content I found on STEEM. Then I know certain users will likely like to read and curate that content, but some other that do not give a shit.
So, what should the platform be able to do? if you mention people in a post and those people actually value the post where you commented on.. then you had some value in doing it... and you should receive better curation for it (aka more height). But on the other side if the mentions do not vote, then you should be removed from winning any rewards for any votes on your comment, and likely loose reputation, based on the amount of reference you are doing.
To prevent bots or bad users here is quite simple... let's make the weight of your rewards based on averages... make too many references to several others using bots and creating a wave effect, would be stupid, because, the higher the number of references, the same average it will create... and therefore not so optimal the advantage. If on the other hand, a very powerful wants to put some value on the post and for some reason, someone had mentioned lots of users to try to win some bucks, that user will actually lower the average per user... by being mentioning everyone.
Likewise, if the same user tries to mention only big users... and the content get's only validated by a single user. then the average will be catastrophically low!
The concept...
These mentions I am referring should be different from current mentions. They should be some kind of "recommendation" and should be seen as the user receiving the recommendation, filtered (highest to low rank) so, the useful recommendations are taken into account first. Also, the longer it passes without your mentions taking action, the less your curation should be.
This will likely solve, BOT auto-voting problems, thefts trying to impersonate people, SPAM, and too greedy readers that wish to strategically promote content using big references.
Does it need more detail? the idea? Shame I am not a good coder like you guys. otherwise, I would be creating my own commit already.
While I confess I am confident I don't fully understand the ramifications of your idea, I will say that this is exactly the kind of out of the box, original thought I most admire, and that makes Steemit to me like Heroin to a junkie.
Thanks!
Yah @jesta Your information is right.Thank you for your feedback.
Your argument, whilst certainly credible, is similar to arguing in favour of the minimum wage because not having one causes a race to the bottom among desperate workers, and thus output quality will suffer. This is possible, but I think in practice some natural equilibrium might be found if we allowed it to be, where the overall dynamic of the platform improved.
To manage the transition, perhaps we could initially limit the curation to a 75% maximum for example.
Unlike some proposals, this doesn't look to be too difficult to implement from a blockchain perspective either.
I agree that supply and demand will eventually reach an equilibrium. However the asymmetry in the distribution of Steem Power may push that point in favor of the large accounts at the expense of content producers.
There may be other business cases that are not centered around content production that could greatly benefit from a totally flexible allocation of rewards. In the end this may outweigh the possible negative effects. After all the Steem blockchain can be used for much more than social media or blogging.
That's true enough. The distribution could have a negative influence on where the equilibrium settles. The recent trend in greater whale delegation may reduce the impact of that too though by somewhat equalising the effective vote power distribution.
Yeah, a race to the bottom is what I worry about as well. I think allowing it to be configurable is a good idea, but between reasonably confined limits, not 0-100%.
I agree. 25-75% seems like a fair range to me, 20-80% at most.
This is also a fantastic point; I think everyone who provides their .02c on the solution need to consider the fact that steemit.com isn't the whole enchalada. These proposals are to the steem blockchain. Meaning it will have an effect on every platform out there; dtube, chainbb, dsound, steemiz, utopiaio, busy, zappl and tons more. Those are just the most popular of the many platforms that have sprouted during the beta phase of STEEM (a beautiful notion by the way. Think of all the ideas that have been born and dev teams STEEM has helped bring to together.)
That, and once SMT's are a thing there will be countless more interfaces using STEEM tech. So if we are to put forth our ideas for UAHF/SF's they need to consider the direction of steem's future as a blockchain, not as an interface.
Don't forget the underdog handicap we would be losing by replacing the 30 minute window with a 5 minute one. https://steemit.com/curation/@beanz/the-problem-with-the-current-curation-system#@sigmajin/re-beanz-the-problem-with-the-current-curation-system-20170120t025022414z
Why do you upvote your own comments all the time? It's kind of distasteful.
This is my self voting percentage for the last 3 weeks.
I vote for myself only for visibility. If steemit inc includes a UI for declining payout on comments I would use it.
Thank you for explaining yourself. It seems you care a lot about Steemit so I thought I would ask why you were pumping your own comments.
I get it now. Whoever has the most money that's who's comment should be first. Right?
I don't have the most money. I pay for the delegation so in my mind at least I'm paying to have the comment seen, yes.
I personally feel that it takes away from perhaps a really good comment from a user who can't afford large personal upvotes that's all. Not a lot of people will scroll the entire comment section and read everything where the gem comment of this entire article may live.
It is self defeating to expect people to invest in steem and even powerup to steem power (a min 13 week commitment to the platform) and then criticize writers who choose to use their steem power to promote their writings.
So you're saying it's good for the platform to upvote your own comments that you make on other peoples posts because you invested in Steem?
Sounds to me like you need to read a little bit more into what you've invested in then my friend. Your invest won't be worth very much if everyone upvote every comment they made.
Here you can be just like this guy in the future. Upvoting your own comments because you invested so it's cool right?
https://steemit.com/art/@paco/discover-kevin-winger#@paco/re-paco-re-paco-re-paco-re-paco-re-paco-re-paco-re-paco-re-paco-discover-kevin-winger-20171130t062652179z
Read the post attached to the comment then tell me it wasn't an important point to be made and seen by readers of this post.
Noted on the gem comments. I'll read from the bottom and try to bring up some constructive comments.
I have read your article and realize that you 10 month so ago made near the same proposal as blocktrades has now.
Thank you reading from the bottom up and considering upvoting some not so wealthy stake holders who have made some good comments!
I will read your post!
@beanz believes his contribution tl be valuable.
*her
I would say everyone would believe their own comment to be valuable. Wouldn't you agree? Why would one take time to write a comment which they thought was stupid or invaluable?
if there are a lot of comments but none have been voted up and
I think i have something of value to contribute i will vote myself to get to get it as high on the list as i can.
But if its not so important to me i don't bother. But will vote on my own posts : )
Steemit has now implemented eward by choice on your comments the same way as posting now. If we feel we have a comment of value like you are saying above we can choose the option decline payout which you force our comment to the top without the reward for the huge upvote. I don't think voting your post is a problem. Only people that make 20 posts a day
Ha, I'm new here but I am one of those who will read through the entire comments! I know, I know! What am I doing with my life? Growing the steemit blockchain with my time on the platform... (smile!)
Yes, so what are your thoughts on what you've read?
Ummm... can I ask where to get a pic like this for my own upvotes? (It seems website generated, that's why I asked)
https://steemreports.com/outgoing-votes-info/?account=ahmadmanga&days=14
Steemreports.com has a lot of useful stats like this.
I do like your idea @jesta. And for the sake of trying to solve a problem, opening to more options might actually reveal how to solve the problem. Just like open source decentralized blockchains. The rule for the 5 or 30 minute, could also be something interesting... and I would even put another TWO versions on the table, which are, 1st "random times for each curator" (within a min and max, implied by hardcoding), OR count how long you are reading a post and save that locally, that would be used when issuing your upvote. The locally saved time can be easily used like some other effect of "proof of work".
I wish that can be possible, but I don't think that can effectively be done on the backend... And thr blockchain has many different front ends (steemit.com, busy.org, chainbb.com)and that would mean all of them should willingly count the time the window is left open...
If this idea can be made practically then I'd be the first to support it!!
Calling it for a challenge! =) I would be happy to propose it on utopia, but I would need some additional thinking to make it worth others spending time on it.
But I am with you, "if it can be made practical, I am the second to support it"
Good luck, if you posted it on utopia.io give me a link.
I will copy -paste just a few sentences, telling about the roots of a problem:
Changing 30 minutes to 5 minutes....I don't know if that's going to change anything, as most bots are already set to vote after 20+ minutes after post is published.
You're right - the problem is much larger than just this, and this won't solve that larger issue.
But the 30 minute rule for curation rewards is just an unneeded complexity in the system that needs to be made less complex. Each part we make simpler, the easier it'll be to solve the entire problem.
So this!
I think this is a good way of thinking...
This would make it easy to find the problems in it after simplifing it... and make it easier to find problems in other parts of the system since making one part simpler will decrease its effect on other parts.
Just changing 30 minutes to 5 minutes would probably have little impact at all, I agree. The more important part of the change is to eliminate the transfer of rewards from curators to authors when the curator votes during the window.
And what is the possibility, whales will show any interest for this kind of change? If this is going to be processed at all...
Of the whales I've talked to, most seem in favor of the general idea, at least as a first step towards improvement.
Many good ideas to think about.
Here are a few more that might help simplify things:
Thanks for reading these suggestions to ponder.
Saya sangat senang membaca komentar anda, luar biasa, ada solusi yang disertai analisis yang kuat. Sebenarnya kita semua sedang menunggu solusi dengan pertimbangan-pertimbangan profesional sehingga semua pihak dapat memahami kesimpulan ini dengan nalar terbuka. Terimakasih atas konstribusi profesional yang telah membuat semua merasa lega.
sory
How about you stop spamming?