Finance of Curation
Recently I came across a post by @costanza, and it is one of the numerous KE ratio related articles that I tagged to lately :) Here is the post. The reason it sparked my interest because it is related to Curangel. It is a curation program I am involved in for a very long time. I want to give a full disclosure: I delegate to Curangel and I am very happy to do it, and nothing written on that post and this current post will change my decision how I am involved with that project. I respect all the curators there and many of them are my personal friends.
For those who need some background on KE, here is the post. I am not going to repeat what KE is and how to use it. I am not going to give anyone any prescription on how to use KE. You do you, Universe will do Universe!
First, I thought the title image was a bit sensational :) Curangel account having a high KE ratio doesn't bother me at all. In fact all curation programs which returns part of the curation to delegators WILL have high KE. This is a FEATURE and not a BUG!
Okay, with that out of the way, I did like some of the data presented in costanza's post on individual curators. I have decided to plot them, because, plots are always better than looking at a table. Disclaimer: I didn't look at the validity of the data, I am simply plotting the data he has listed and making an analysis.
There are 13 curators represented in this small sample, and this plot above is their individual KE ratio. High ratio is generally considered 'bad'. I used parenthesis, because good or bad is not simple, and more importantly, I shouldn't be the one deciding good or bad. It is not my job, I can only make that determination for my actions.
So in summary, if KE ratio for an individual account is 10, that individual currently holds 1/10th of the HP of their total rewards. However, for a some of those people, they don't even post, so whatever their KE is doesn't matter. Others with high KE post very infrequently, so for them, again I don't really care that their KE is high.
Who are they voting
I feel this is the more important part. As a delegator if this KE number is very high in the median, then I should maybe a little bit concerned.
Here is the data. This is the plot of the medians for the accounts our curators are curating in this sample. I generally consider not voting accounts greater than KE = 3. I only see 5 curators during such a thing. Again, this is my personal cut-off. I am not even aware if all the curators are even aware of KE, not to mention the cut-off. Also they certainly do not have to follow my personal cut-off. They can curate a post based on their personal quality cut-off.
There's that. I just wanted to get this out there, so that we can take a look at it. I don't really see anything concerning at all from my personal point of view.
I'm currently curating the Splinterlands art contest and together with isaria we added some new rules, some users didn't agree or found these rules asinine because I had comments like: "we just want to share our art".
I was interested in participating in the contest because, there are many users in power down or some shared the art created without at least explaining what it was about.
I know that maybe it needs better changes but, if we move everyone who has a high ke we would be left without participation 😅.
I wouldn’t support any person with a constant power down.
I don’t care what art they share.
Use the KE cutoff will be my request.
Nice follow-up!
The main point that I was trying to make is that there is a clear pattern (this isn't aimed at Curangel specifically, I just took them as an example). Go to any of the Curation Compilation posts, click on a random post that got curated, check the activity on the account and the wallet, and you will see in most cases that:
It just feels like many curators are taking a rather easy approach of having a list of accounts where they know the content at least looks nice to curate (or where they know more upvotes will come in on). While many of the content creators think that Hive is a platform where you post content and earn money which simply isn't a sustainable model if everyone is just cashing out and nobody is investing.
There really needs to be some awareness that if you want to get something, it also requires to give something, and Curation groups and their curators being aware, while at the same time raising some awareness around KE Ratio would be a good start.
In the end, it's fully the responsibility of those who make the upvotes or delegate their Hive Power to ones who make the upvotes. It needs to be near impossible for pure value extractors to do so. Also, I think Leased Hive Power should count toward KE Ratio.
I agree with your three bullets, and I will add three of mine :)
Medians might not be representative of the extremes, for example median might be just over six, but the highest KE of the upvoted author might be much higher...
So you don't think there should be an extra evaluation of those authors with KE 6+? I saw a bunch of KE 16+ all the way into KE 70+...
well, you made me look :), and I shouldn't have
https://peakd.com/@evildeathcore/wallet
Yep. This one is an extractor.
I don't care what he posts, I would have NEVER voted him :)
had to unfollow that one recently, wasn't really about the extraction but the constant shoving out posts but barely any comments left anywhere was shocking.
Yes. Typically there can be arguments when the numbers are 5-10, especially if the posting habit is sparse and with quality. However, when it gets to north of 10-20-30.... pick a number, usually there is no arguments.
I checked that author. Zero engagement!
It's now 83.6
The extraction is crazy though. And the fact he's fairly consistent with his posting.
Maybe, it's a tough time for him or something and he's using his HP to survive.
If the distribution is log-normal, and in this case it is. I like Median as a better distribution than mean.
Look at this data for example
Mean is way too high because of the two high data points. Median is more representative of the sample. It is always the case for a log normal distribution.
If I plot the data as a probability density function and do a log-normal fit, you can see why median is better
perhaps KE 70 was a superlative post! I don't know, but I can't judge, as I haven't seen it :)
I am just looking at the data from a statistical point of view, and the median looks good to me.
Here is the one with over 70:
But you are right, two high scores in this case also are messing with the Average KE...
I looked up the 78 :)
It is a disaster!
Don't make me look up anyone else!! :) :)
I think a lot of people get upset about KE, but it's just another metric people can use for their personal preferences. If someone wants to factor in KE to their voting, so be it. If not, great for them.
Yeah, it’s unfortunate that they do. This is just a conversation
Look at the curators KE's too, at least one of them is ridiculously high. Are these people doing good for HIVE, just taking.. rinse and repeat?
I saw that. I don't know the background. I am worried about the curation more than curator's own KE.
I am glad that the content still matters :)
In fact, I don't think it does that much. I am not saying it shouldn't but just stating the fact. Statistically.
Perhaps the content of the person's character, to quote a historical figure? :P
Yes now it's better.
It is important to know people. If it is just an account then I am only going to apply statistics, but if it is person I likely know the background.
I had doubts about the last graph, but I read the comments and understood perfectly.
Valuable information.
*As a delegator, currently, there is no way to control that high KE votes are not cast, right? *
And Do you know if there is any curation programs with a KE < 3? Or least, below 10
I still believe that KE is valuable information that everyone should interpret as they wish; that is the essence of Hive.
However, almost a year later, I still don’t understand the drama surrounding this issue
You can certainly voice your opinions.
For example, I can certainly tag @ewkaw and politely request to reconsider voting or not voting evildeathcore, however decision is up to the curator.
Regarding your drama comment: traditionally hive people like drama. :)
But jokes aside, people typically don't like to be told what to do. They just want to do whatever they want. In vacuum that is perfectly fine. However, we don't live inside a vacuum. If someone's actions hurts or affects someone else negatively and it is detrimental to the society, then we have a situation and someone got to speak up.
You are from Cuba. You are very familiar with this, not that it is strikingly different in the US now :)
I think having curator gives example that they need to increase their HP is good
I mean in Hive for some new people curator are like influencer on Web2 social media and in my opinion when curator just extracting it will gives example to newcomer.
But that only just what i thought.
Nothing offense to all curator since they are having their own financial program
Yes that would be preferable
KE scores makes me sad sometimes. I saw some cool art posts. I checked KE of the authors and it was huge. Like 20 or 30? So I just moved on.
Curators should have some kind of automated reply to raise awareness that says, 'Your post was going to be curated but your KE Ratio is too high with some explanation'. I assume that many on Hive don't do it with bad intentions and just believe Hive is a great platform where you can earn money for making content.
Unfortunately, that isn't the case as there is no real revenue, and the earnings that are getting dumped are just creating sell pressure while there is little to no reason to buy Hive as it's a race to the bottom this way.
interesting idea. Perhaps that would be possible.
Yes. I don't typically vote KE = 20 unless I specifically know the person and in that case I have requested them to lower it.
Hi Azircon, can you explain the last graph a little further? Is it a graph of the KE of the accounts receiving the upvotes? What is the difference in voting activity between the highest and lowest here? Is the lowest voting accounts with low KEs?
Yes, I can. Every curator votes multiple people during one cycle. This is taking a single cycle and list the median KE of all the account a particular curator votes.
Thank you! So the curators on the left side of the graph are more likely to vote authors who stake their rewards and the curators on the right side are more likely to vote authors who extract a large portion of their rewards.
That makes sense. It's a good metric. Thank you.
Yes. On that sample that is what they did. I am not sure if that is a statistically significant behavior or not. Currently the sample size is one.
I am simply plotting this data
Till now I don't still understand the KW ratio very well and since it's no longer in our profile, I don't think a lot of people are even aware of it😂😂
The curators should still try and pay attention to what and who they're voting on though.
What's the need of our transactions receipts being public if those giving us money aren't checking where its going to?
You can make it visible on the profile for everyone. It is a setting in PeakD.
Go to PeakD -> settings and turn this on. You can see all KE
I always check everyone’s KE
I see...
Thanks
Very interesting, thank you as always! Brought me some extra insight, I hadn't thought about curation projects having a high KE as they pay out the gains... Makes sense, though.
I must say I'm surprised that I see so many curators with a high KE themselves. About who they curate, I'm not sure. I'd like to see that they're responsible with their votes, weighing all the different angles. holoz0r did a great comment on that in the post you tagged, and the string of comments following were mostly the same quality. He basically points out all the thoughts that I had, so I'll just leave the link - if you haven't read it yet.
One thing I might add - KE is relatively young. It still has to filter through and become an accepted metric, and also be understood enough to use it as a tool to complete a picture. There should be a way for everyone to have a little explanation field within the KE-Display, to state why their KE is so high. That would make curation easier, too, though of course everyone can lie.
People can explain. I see holoz0r explaining away a lot :)
Yes, but they have to do so every time again. Though it doesn't really matter, anyway. If they are a constructive part of hive, they will have a network of people who read and curate manually and probably even make a whole post on why they decide to divest or to re-structure their finances on hive, and they can pin that on top of their blog or something, if they really think it's important to explain it. That way, KE-oriented curators could have a quick insight while checking the KE.
I rarely vote someone that I do not know, personally. Now I do follow some curation trail that votes whatever that votes. I need that to spread my votes also.
As soon as I'm as big as you, I will do that, too. I just recently reduced my curation-trail vote, as I'm voting a lot more manually and my votes are quickly depleted. And it's not even that much... But I feel like it's better to get used to a curating strategy now, as I'm a small fish, so I do a better job when I grow.