As you may remember, in these two posts in I criticized the way how 'curation' currently works on HIVE (and before on STEEM):
-
I think one essential problem is that most of the time curators don't read what they upvote (so there isn't a really high correlation between the number of upvotes and the quality of the content), and ...
-
... one of the main reasons for that is the small 'curation window'.
Summary of what bothers me
As I wrote in these articles I think we should remove the five minutes curation window (directly after a post was written), because it clearly disadvantages manual curators who try to read and evaluate posts before they decide to upvote them or not. Before they could do that, most of the time lots of automatic upvotes have already come in.
It also makes 'curators' lazy (especially since there is 50 % curation reward): they simply upvote the same people again and again (there even aren't 'diminishing returns' implemented), while not enough users are seeking for new authors (respectively honor posts from old authors who aren't on their 'lists').
I see no reason to be rewarded less for a late manual upvote than for an early automated upvote!
And considering the high percentage of automated upvotes, I think it's simply a myth that manual curators find 'quality content' first, and then more upvotes follow and reward them (yes, that may happen ... as an execption).
New idea to improve curation
Recently I was thinking about how to make it less attractive to just upvote early without reading anything, and my conclusion is that curation reward should solely depend on the weight of an upvote, neither influenced by the date of the upvote nor by the (expected) number and strength of upvotes of other users (which currently can be a reason to decide either in favour or against upvoting a post). Curation rewards would be only determined by the available HP of the curators and the percentages of their upvotes.
"But how does that fit together with the convergent linear rewards curve, @jaki01?"
Well, I like the current convergent linear rewards curve (actually I suggested something similar long time before it finally had been introduced), so I think as curve for the author rewards we could stick with convergent linear, but for curation rewards the simple formula that curation rewards would only depend on the vote weight of the curator could apply!
Then there was no reason anymore to upvote certain users early just because of expecting other upvotes to follow (the same applies for the or the opposite case, not to upvote a post when it has already collected too many upvotes), while great authors still would benefit from getting many (strong) upvotes because of the convergent linear rewards curve which still applied for author rewards.
Disadvantage of my idea
It's true that self-voting would be somwhat more beneficial again as self-voters would receive more curation rewards than before. At least author rewards would still be limited by the convergent linear curve (and don't forget the 'free' flags since the introduction of EIP ...).
In case this bothers you too much, I offer you another suggestion which, however, is quite an attack on EIP:
when going back to 75 % author rewards (and 25 % curation rewards) the increased curation rewards for self-voting described above would decrease again ... But well, I better stop here, not to trigger another flag war against or in favour of EIP. :-)
Short summary
-
Convergent linear curve for author rewards (no change needed).
-
Curation rewards depend only on the own vote weight (HP, percentage of the vote).
-
Possible addition (if self-voting is considered as serious problem because of 2.): going back to 75 % author and 25 % curation rewards. :-)
I would like to contribute another example for the absurdity of the current curation system:
In our "Fascinating Insects" community as mods @faltermann and I are the experts when it comes to these little creatures. However, instead to follow our upvotes, some 'curator bots' with big pockets but no clue about the matter prefer to upvote any blurred ladybug pictures.
And why? Because they earn more when they upvote first ...
Now that I have read many comments and hopefully understand most of them to some extent, I have to realize that everything is not that easy and a fairer solution must be wanted by most people. When I look at the article, I am surprised that especially the big accounts seem not to want to take part in this enormously important debate. This post received the ridiculous 19.34 Hive Rewards after 15 hours. It shows quite clearly that those who swim against the current do not get a tailwind.
After publishing only 2.5 years on Steem/Hive within my possibilities, I realized that justice looks different. My friend Ritschi accompanies my experiment Hive as an outsider. He still doesn't understand how all this works, even though he reads all my posts. He asked me, why do I get a lower reward for an elaborate contribution than for a simple picture whose effort is limited? He asked me, why I don't post my pictures one by one, that would be much more financially rewarding, since one picture of me would still be better than 100 other blurry snapshots without description? My answer is clear, in the short term this would definitely be more lucrative. However, such a trivial contribution does not meet my personal requirements. I believe that in the long run good contributions that provide real added value will pay off. The question is whether we will be able to show people on Hive that it is worth the effort. The reward for this effort must be felt by everyone who comes to Hive!
Of course, this 5-minute window of curation also annoys me. As @jaki01 already writes, this must be changed. I am confident that Hive will experience a positive change quickly and fairly.
Wherever there is money involved, there will be people who do not think long term enough and only chase the quick profit. Most users do not want to deal with the mathematical calculations of comments, but this automatically puts them at a disadvantage. We have so many creative people here who, no matter what area they share their creativity with us, should have more attention and visibility. This is the only way to promote real quality and new members recognize the added value behind Hive.
Performance and quality must be worthwhile!
I think it's detrimental to favour automatic voting when we really need real curation. I benefit from a load of votes minutes after I post, but I would rather have more actual readers.
Well said!
I know, concerning the number of upvotes I am also privileged compared to many others, but concerning real engagement in the comment sections of my posts I also have experienced better times! Nowadays people leave their upvote (take curation reward), and then consider their job as done (only if HIVE itself is the topic, engagement is a little bit higher). :)
In my last post I am trying to promote a HIVE soccer tipster competition, and before publishing it I wondered if the maximal allowed number of 300 participants would be enough ... but so far only six HIVE users have joined! :) Of course not everyboday is interested in predicting soccer results, but I guess that the other reason for the (so far) very low number of participants is that the majority of upvoters just didn't read my post ...
Part of the problem is the small community. For a lot of interests there is just not much of an audience. BTW I'm not a soccer fan. We need to build on what communities we have and not focus quite so much on rewards. Of course some people are desperate for any rewards, but a lot of us can play the long game to try to make Hive more attractive. We need to reach some critical mass where it's a go-to destination.
Sure, but to reach that aim we must discuss our problems to improve and become more attractive for new users ... and I think we are just doing that (for example) here.
I don't think the masses care that much about details of the rewards system. They are happy to make anything, but real engagement has value too.
They don't care about the mathematics/the mechanics behind it ... but they do care about the results!
And of course they wonder about the rewards generated with posts like this, this or this one.
The frequency of this type of overvalued posts depends for example on exactly that stuff which we are discussing here.
Let me be clear that I'm not holier than the pope; got quite a few people on autovote also.
But nevertheless from day one I found this 5 minute window a ridiculous thing.
No-one is able to read, evaluate and upvote "properly" within that short time-frame.
I totally agree with you and your idea.
I wouldn't claim that, too, concerning me. :)
However, if certain kinds of behaviour are discouraged, maybe even unholy people like us will start to act in a better way. :-)
Good to know that others agree with my thoughts, so at least it seems I am not completely off the track.
'maybe even unholy people like us will start to act in a better way.'
Amen to that, hahaha.
Sometimes, the auto is required for some curators to attain the 5 minutes window claim. But for me, I tried to read more and I curate too as well with the little HP I have.
Some good points. But for my point of view on this chain, the biggest Problem is the less upvoting on comments....
Most Stakeholders votes Articles alone. Mostly open the comment section is not being a option. "Nobody" read and wrote UNDER a article on this chain.
Me personaly make it sad, becouse i love reading and commenting much more, than wroting own articles.
I did wrote thousands of comments, with way more content as my articles. Sometimes i get the answer:
"This comment can be a own article."
Yeah, nice, but it is just a comment and i was just inspired from the topic words and thought in the article.
We need more interaction first, so much more people do write less articles and more comments on this chain. With that, for me the worth of a "Social Media" is growing really.
I like your idea but I would add a 60% penalty on self-votes -> you better use your votes on other people if you want to maximize your rewards.
I recently published another idea though. Quoting myself now:
I could imagine something like this as alternative system:
"Content that is actually being consumed is the most valuable":
-automatic rewards allocation based on reading time of individual mac-addresses (not IP-adresses or vitual machines)
-all posts get two new counters, number of individual mac-adresses reading the post and combined duration of reading of the post.
-the reward pool allocates resources automatically to all posts that have been written 7 days ago based on those metrics (maxed out at 5min of individual reading per mac-adress to avoid "reading bots").
-add in a small percentage on top, that increases the payout of written posts based on staked tokens of the author.
This system is hard to game, no votefarming, not easy for circle jerkers...
With this system you could also reduce the rewards-(pool) to authors by 20% and increase the staking rewards by 10% to encourage hodling, leading also to reduced inflation.
Also maybe think about "the longer the HODL (meaning no powerdown of the tokens, first in first out), the bigger the HODL rewards" (increase by 1-10% on top of your HODL rewards)
Sure, but often the biggest abusers also have the biggest number of different accounts ... Maybe something like 'diminishing returns' when upvoting the same accounts (also own one's) again and again would be more effective?
Interesting of course ... but wouldn't the maximizers start creating 'read bots'. That shouldn't be too difficult. You may oppose that long reading times only might benefit the author, not the reader ... but still: there could be 'read buying' or 'read trading': "I 'read' your posts all the day if you 'read' mine." :-)
The nightmare of a system like this is aggregating data from multiple sources. How would you pull and combine viewer & reader data from hive.blog, PeakD.com, eCency, Dapplr, 3Speak, Vimm, WordPress blogs using the SteemPress plugin, etc. etc. I just don’t think it’s feasible to automatically measure attention at the front end levels in a way that distributes blockchain level inflation.
Yeah, that's unfortunately true.
But we should come up with some clever way to distribute rewards differently, incentivising what we want and disincentivising what we don't want. I don't really see discussions about this anymore.
setting penalties for self-voting will never work due to sockpuppets
back in the day we had diminishing returns where your posts would earn less and less rewards from the votes on the 4th, 5th, 6th post etc and that was also obsolete as people would just create more alt accounts to vote on instead.
Read and approved 😉
OK, I start with 3rd, because I have nothing to add to the first two points:
Sure, but often the biggest abusers also have the biggest number of different accounts ... Maybe something like 'diminishing returns' when upvoting the same accounts (also own one's) again and again would be more effective?
How do you want to 'forbid' that in a decentralized network?
I would say: flag posts upvoted by bid bots, and it's not worth anymore to pay for them! That's one thing that works rather well already on HIVE in my opinion.
Well, maybe only count these visitors who stayed at least for a certain amount of time ... But right, I agree.
One question should be "Where does abuse start, where does it stop?" That alone is difficult to define. I have observed - even on myself - that the (moral?) boundary shifts the richer you get. In the beginning I never used a selfvote - I'm not my biggest fan myself. I even found it disgusting when I saw this - even with small users - in others. At some point the penny dropped that the quality of the articles had long since become a minor matter and that's when I lost any "shame" as well.
Maybe 'diminishing returns' are a solution, but also a bit unfair to those who are very diligent, each article is a very good one and for this reason alone I want to support them regularly.
Since the rigorous solution - each user only one account, selfvotes forbidden - is not possible in this system, there is no meaningful control in my opinion. And on the insight of the people can unfortunately not bet.
Flags are okay. But who uses them? Most people are afraid of a flagwar and without a "higher instance", it is anyway hardly possible to find out every selfvoter, every votebuyer.
If I see that a user has bought excessive votes or does so regularly, I won't vote. Others only vote because they know exactly that the user pays bots and so they get a higher CR if they are earlier than the bot.
How you do it, you do it wrong - we cannot change people and their attitudes. But it's certainly always good to call a spade a spade and provoke new thinking, as you do here.
I won't let myself be annoyed by the circumstances any more. As long as I enjoy it, I blog. Those who like my texts, vote for them, those who don't like them, leave them alone. I don't question it anymore - sometimes I still grin... ;-)
I try not to consider the topic in particular as moral question (even if it partly is one). If someone upvotes themselves or 'curates' a bad post because that brings them more curation rewards, they do it because it's possible (we are - some more, some less - greedy humans by nature)! :)
I try to find solutions which discourage certain kinds of behaviour which in the long term damage the platform.
As long as "regularly" wouldn't mean several times per day (the definition @haejin prefers), the impact of 'diminishing returns' would be rather small. The good thing about 'diminishing returns' is that nothing was forbidden, but with every further upvote, the rewards were getting smaller (however, your voting power when upvoting your 'best friends' would of course recover after a while).
Yes, of course flags are (too often) misused, but at the same time served to control @haejin here and bid bots.
In my eyes it's not the aim to punish every single self-vote etc. but there should always be the threat that it could happen if one does it too often (similar like not every person who drives too fast is catched, but the probability is rather high ...).
We aren't living in a perfect world and won't reach perfection on HIVE (even if I think it's worth trying to improve the system). I like your attitude to try to have some fun anyway.
I enjoyed reading the comments section and truly learn a lot from your conversation.
I am also trying to figure out each and every points you presented here.
Overall I agree with you and I can see how knowledgeable and devoted you are to the improvement of this platform.
Honestly, I'm doing my best not to upvote my own post, I also believe that its best to be used to upvote others.
I only upvote my own work if I find it unrated even if I did my best in coming up with that post and spent hours just to make it a bit attractive to curators and informative for those who really read.
Although my votes doesnt make any significant difference at all but I pity my post and myself for getting only cents, very discouraging.
Lastly I prefer the 75/25% sharing of rewards hope it will be applied back soon.
Thank you and more power!
Everything written in this post is clear that's happening in the chain. I agree totally with the new idea but the fact that as of the moment, everyone's concern is all about getting the best out of curation, only worsen the current situation if will be implemented. And I'm not surprised this is not everyone's main concern or maybe they don't really care, they just let the normal process sink in and interact within the chain.
Maybe that's a good sign that they're not into the potential money of this network or they don't care who's getting more rewards and who's barely earning it. But that would be a selfish act against those who invested real money as the price will keep on tanking no matter what because Hive is an active token. It's not mainly about protocol-driven distribution but emotions bound to affect and disorient the network.
But I don't lose hope for we still have great curators and leaders inside. I owe every token I have earned in this blockchain to them except for those tokens I actively moving on exchanges.
I use autovotes too but I would gladly adopt the change you offer. I must admit that sometimes, when curating manually, I have a hard time upvoting a post that I like because it already got a lot of big votes and the potential curation reward would be next to zero...
What do you think about this experiment? It seems that it enables an author to decide whether they want to stick by current blockchain code (order of curation votes matters) or to use @reward.app which will distribute curation rewards based solely on vote weight (order of curation votes doesn't matter). In addition, it even enables the author to specify a percentage higher than 50% as the post's curation reward.
One thing which I really love about it is the granularity of control - different authors can choose which rules they want to use (they're not stuck with the blockchain's rules), and can even use different rules for different posts - e.g. for post A I might go for 50/50 split, but for post B I might go for 20/80 split. So customizable!
I fear that after authors could choose the percentage themselves, in the end that would lead to more and more reduced author rewards.
I see that risk, yes. What about downvoting low-quality content? To my mind, this can serve as a counterbalance to low-quality high-rewarded posts. So if curators vote for a post because it will give them more rewards, but the post is not seen as high quality by others, then they will lose their curation rewards when those others downvote.
At least in theory. I know in practice we've had a lot of challenges with having downvoting work well.
I fear (again) that the biggest 'curators' (even if many of them don't deserve that name as they just don't work but let programs work) are big stakeholders, so I doubt that there would be enough users with courage to downvote them.
I think many stakeholders would choose the biggest curation rewards (no matter the content) instead of considering that in the long term it would be beneficial for HIVE to see great authors succeed.
If you (and too many others) are against '75/25' maybe the self-vote problem could be also solved with flags, because I think apart from this mentioned, and hopefully not too serious disadvantage, my idea to make curation rewards independent from the date of the upvotes and the number and weight of other upvotes could help to solve the curation problem.
I am not against '75/25', I am for distributing the reward pool to high-quality content and if '75/25' can help that when combined with your idea of decoupling the order of the votes then I'm all for. I wonder how such experiments can be made in an easier way. A thing I liked about @reward.app is that it allows making an experiment without having a hardfork. It's much faster and more granular (not everybody has to abide by the same rules). So I wonder if they are able to provide the ability for different communities to use @reward.app to experiment with different models: '75/25', '90/10', with order of votes mattering, without order of votes mattering, etc., etc. I think that can open the door to essentially having behavior like SMTs without having to code in all of that in the blockchain. @acidyo @roomservice what do you think? Is this possible?
I mentioned this to @acidyo ages ago, I think the 5-minute window is a dumb idea too. Maybe it was there for a reason once, (30 mins, then 15 mins, now 5). I don't see the point and your stake should be the weight.
I don't self-vote so the other issue is a non-issue for me. Flags will sort things out like that it people start taking the piss.
The issue with linear curation and going back to 75/25 is that the bid bots will come back.
I think all we need to change is the 5 minute window and perhaps make rewards fixed for the first 24 hours before they taper off; then people can take the time to curate manually and outcompete the bots.
Lets 'welcome' them with flags. :)
Maybe, but I still see no reason to reward early voters higher at all than late curators (at least not as long as the majority of upvotes are auto-votes) ... Why should I upvote a post, when any of the large upvote bots was faster than me? To increase their automatically generated curation rewards? :)
(In some cases I ask the author to write a comment for me to upvote, so I have not to support the curation maximizers by following them.)
I agree with all your points here @jaki01. But changing the system might take years to happen, if it does happen. The resistance will come from people with higher stakes than those in favor. It is still a worthy cause but meanwhile I think the best path to take is the one you mentioned above:
This could be the best course of action because you won't be adding to the rewards of the early voters and you can take your time in curating posts that you really like. And although you will be going over the 5 minute rule, at least the bigger share of the rewards go to the author and not the bots.
This is as far as my understanding of the matter goes. I'm no expert at these matters but upvoting the comment instead of the main post is a good palliative for the moment that things cannot yet change.
Yes, I will do that even more often in future.
Agree on a surface level but we need to provide some incentive to up vote something that is better than for example just this comment. Linear curation rewards encourage just up voting of anything as the rewards are all the same. People will self vote and claim the rewards; there is no disincentive other than vigilante mobs of down voters which makes the platform feel more like a failed police state.
I like to upvote comments because they are essential for real interaction and communication in this community.
In the eyes of many newbies HIVE must look like a real desert, an unpeopled ghost town, inhabited only by bots!
However, what I meant was: when I see that some well known bots have 'curated' a great post which I would like to upvote, I often abstain from that and ask the author to write a comment which I upvote then instead of the post because I don't want to increase the curation rewards of these voting bots.
I disagree! :)
If rewards were the same anyway, then you would be free to upvote quality content! You didn't need to think about your curation rewards anymore but could focus on what you like (and I think the big majority of users would do exactly that)!
The few undiscerning ones who wouldn't care at all could be flagged.
For author rewards the convergent linear rewards curve would still apply ... and if we had '75/25' again, self-votes wouldn't be that beneficial.
In addition one could think at further measures like 'diminishing returns' when upvoting the same (also own) accounts again and again.
Furthermore, I think something similar like "Voting CSI" in SteemWorld could also serve well to detect abuse.
I agree too. I think it's ridiculous that autovoting is allowed at all (but apparently it cannot be stopped) so why encourage it with this five minute rule.
How does the 5 minute rule encourage auto voting?
The 5 minute window is a "reverse auction." It's meant to somewhat discourage auto voting. Any votes made on a post in the first five minutes lose part of their curation rewards. I'm not sure if it's exactly linear... but as an example, let's assume you were the only person to vote on a post and your vote is worth .20
If you vote at 5 minutes, you get a .10 curation reward. If you vote at 1 minute you get a .02 cent curation reward and .08 is returned back to the broader rewards pool.
It doesn’t help auto voting... in fact it keeps auto voters from all voting immediately as a post is published. Instead voters “bid” (thus the reverse auction) by voting earlier than the five minutes, gambling that the curation reward percent they are foregoing, will be less than the increase they’ll see from voters coming in after them.
The 5 minute rule is not the culprit in encouraging auto voting. It’s a mitigating factor. The mechanic where early voters earn a percentage of subsequent voters curation rewards is the root of auto votes.
The problem is that as a manual curator, who really tries to evaluate a post before upvoting it, you will be always behind the huge army of automated profit maximizers (even in the very rare cases that you found potential interesting posts shortly after they had been published). That's simply a fact, and if you don't believe it, ask users like @acidyo who have tried both methods and then compared the generated curation rewards (and by the way no auto-upvoter upvotes a minute after post release).
I disagree. Having a larger 'curation window' the probability would be much higher that a manual curator noticed a post before the time of maximal curation reward had been reached. This alone would be an advantage compared to a very small curation window. Furthermore, as already mentioned, it's nearly not possible to find, read and evaluate a post within the first five minutes after its appearance.
Here I agree. And that's why I suggest that curation rewards should be independent of the date of the upvotes and upvotes of other users.
I really wonder how easy it would be to implement a curation window depending on word size of post, this would of course not work well for videos and those 5 page footers with donation addresses some people like to make but yeah. Or give authors the option to set the window before posting.
But why a window at all?
I often notice great posts a day or even later after they have been published, but it's me who takes the time to read, evaluate, comment them and then decide if/how strong I upvoted them (that's real curation work), whereas many other upvoters don't even know that they have upvoted the post, but nevertheless get more curation rewards only because their auto-upvote programs noticed the post faster than my slow human eye. :)
You may also read what I wrote concerning my insect community.
For now I adopted the following method: when I see a great post with many auto-votes, especially by bots like for example @upmewhale or @appreciator (yes, sometimes they accidentally also upvote great stuff), I place a comment in which I ask the author to write a comment. Then I upvote this comment to support the author without increaasing the curation rewards of the curation maximizers (/vote sellers etc.).
without a window the maximizers would have it even easier by voting on popular authors at the same block/ 0-3 seconds. and yes, they some times vote up posts if they have very little rewards (close to 0) as that's what gets them a nice ROI, so best would be if you could downvote their main posts a bit and then upvote a comment, but of course you're going to need considerable stake for that to be worth it as it's not easy to get past the curve tax
With my idea, described in the post, it wouldn't be an advantage anymore to upvote popular authors!
One could freely upvote whatever one likes because with a linear 'curation curve' there was no need anymore to care about the date of an upvote or the upvotes of other users!
As the curve for author rewards would still be convergent linear, it would still be worth it to create great 'quality content'.
PS. Still manually curating here for 4+ years, only tested the automated version on shteem
Exactly!
Good ideas. Can we find enough witnesses to support these proposals?
Am I the Oracle of Delphi?? ;-)))
(I guess not but I think it's always worth a try.)
Automatic upvote sometimes is recommended when the main Curator is not physically present to read the articles but nevertheless, we need more of consumers which I will say readers than the producers ( the Curators) . Morealso, the needs to be place of the same balance sheet for more progress on Hive.
Sure, but that's not the main reason for all the automated upvoting ...
Thank you, for caring about this platform & trying time come up with ways to make it better.
I hope Hive thrives & does not go the way of Steem...
Hi @jaki01 😁,
I agree with you on that the timeframe of 5 min should be extended to 10 min at least.
I don't agree on it. In my point of view, that means less efforts for voters, they will be upvoting only the posts of their friends since no extra rewards to upvote others or to discover content. I also like the idea to penalize the curators if they upvote more than 1 post per day per author or more than 5 posts per week per author (give less curation rewards). I know some spammers that are being upvoted more than 10 times per week by the same whale. Remove the selfvoting is a nice temporary try, because eventually people will create an alternative account and delegate their stake and continue selfvoting.
@Mahdiyari (the hive.vote provider) should stop his service, we will have less jerky circles like @backscractchers or @honeybee. At least it will be more difficult to set up jerk circles. As well, developers should obfuscate the code to avoid people follow trails and set up fanbases. Consequently, more people will be aware of their investment and upvote manually. The fact is that, high stakeholders will need to pay for manual curators like me or delegate their stake to the bidbots 😃.
It's ok to avoid spam and promote investments.
Nope. A lot of content creators are using likwid or reward.app to liquefy rewards. The powerdown journey of 13 weeks is nice for the platform and for our security.
Yes you're right sir.i noticed that some of upvoter upvote a post without reading it. Its result to that one of quality content will not get vote and post with no quality in it get upto 14$ vote.
And as you say the curation reward must be depend on vote weight.
I wish some developer implement this idea and reflect the changes soon. Thanks
Thanks for supporting my idea! :)
You're more than welcome. This is the need of every Hive user. I personally know some person who write quality content but he didn't get vote and also knows some who just take advantage of their high level