HardFork20 and worrying about 1.2 SP votes
I read a funny milestone post by @v4vapid about being unfollowed by 100 accounts in the last week. Congratulations once again mate, you deserve it?... or something to that effect. Then there was the post about user retention bu @paulag that I read last night and quoted today about the churn of accounts decreasing and finally, one by @tcpolymath yesterday about Hardfork 20 and the effect it could have on curation trails and their voting. What do these have in common?
Maybe nothing, but since I like to connect dots even if I have to draw some in myself to begin with, I will.
Firstly, as I wrote in my own post earlier, I think the churn rate is potentially a good thing at this stage of the game as it leaves the dedicated and committed core users to go about the business of making this place something worthwhile.
But, this is more about @tcpolymath's post. There is nothing wrong with it per se, it is just that, is it really an issue? I know that all of these small numbers add up to large numbers over time and with many instances but, at the same time I question accounts that are casting votes that are using under 1.2 SP of vests. What I mean by this is that after HF20, votes under that amount won't count. I say, good.
People will of course claim that it is unfair on the small accounts who are coming in trying their best but in my opinion, if they are actually looking to earn here, casting 1% votes with 100 SP (or worse, 15 SP they got from Steemit) isn't the best way to go about earning and growing anyway. New and very small accounts should learn about curation but, forget about earning a curation return if they want to make it here in the long term. People are trying to maximize their earnings in the wrong areas.
Although I don't autovote nor do I trail, I have trialled both methods for a short period of time. I personally don't like it as I would see my vote on content that is not what I would naturally vote but, that is not the main risk. The problem is that passively voting is not going to be a significant earner unless a) one has a significant amount of SP or b) there are many accounts voting in a curation botnet. Obviously, the accounts with significant SP aren't going to run into problems with HF20 rules. It is the botnets and tiny passive users who aren't engaging that are going to be largely cut out of the game.
This is the issue with retention as it seems many of the accounts feel that they can come in and with their 15 SP or whatever it is, earn enough by setting and forgetting it. How many just, forget to ever return? I am not sure how this can be calculated but I would say that if we could analyse all autovoters, a large number of the smallest accounts there will actually be dead accounts, accounts that no longer interact in any other ways. This is what is going to affect the churn and burn rates too right? It is also going to affect the statistics of active users that will be used for marketing purposes one day.
As I see it, it would be great to be able to purge the system of accounts that don't actually offer anything to the blockchain and HF20 might do just that in some ways, at least statistically speaking as it will be easy to identify the accounts that vote but, don't grow as autovoters and, not active in other ways.
The problem with autovoting for the small accounts who believe it will get them somewhere is that they aren't engaging, they aren't building their account in other was, they aren't reading or doing anything to be a part of the community. Their tiny vote might add up to something in combination, but in the scheme of one @curie vote for example, they are useless.
Ok, so how does @v4vapid's post tie into this?
Using three of the examples he gave: @hashex, @kimval, @leeze...
All three joined in March 2018 and have respective account values of 14, 25 and 75 cents. All of them are resteem accounts and the highest value one has 120 "original" stolen image/quote posts that were spammed. These are the kinds of accounts that are making up a large majority of the resteeming accounts people have noticed lately targeting the people with high reps or in trending.
I don't know for sure but I would say that their follow and unfollow technique is a way to attract follows in the same way they do on Instagram. I might also make a large leap and suggest that someone is buying these accounts from users who aren't using them anymore to add them to their network. It is likely much cheaper than creating a new one and, it has the added benefit of already having at least some followers (retarded ones) meaning that half the work is done. They can advertise their resteem service to be seen by 100k accounts! Considering the estimate is about 60k active accounts out of the million and no active person n their right mind would follow a resteem service, what are those 100k accounts?
I have several accounts that autovote me often but, they haven't been seen on platform for a very long time. Unfortunately, they all only have about 5 SP and their vote is worth literally nothing. Some also vote at the time of posting which means not only is their vote worth nothing, they get nothing anyway. It doesn't seem like they really have taken the time to understand the platform does it?
People complain about the large account autovoters not being sensitive to the platform but, how much is being drained out of the platform by the small accounts or, being locked up by inactive accounts who will never come back. The massive botnet by DART/TARD was taking something like 22,000 dollars out in a 6 month period. Now, that is significant. Not only is it a significant amount of value, it is a lot of accounts that are doing nothing good for the blockchain.
As I see it, small accounts shouldn't be autovoting, they should be engaging and using whatever tiny amount they can muster to encourage other users. A 15 SP account should never be voting at 1% which is the reason that the slider originally had to be earned. Yes, 500 SP is too high but originally it wasn't, not that it matters because other interfaces enable the slider anyway or it is applied through the autovoter.
What this means is that accounts are being created hoping to earn but not actually be a significant part of the community. they want to draw from the pool without offering anything in return and as they realise it isn't going to work and what will is too much work for them, they leave. Churned.
@hashex, @kimval, @leeze... Now, tell me, what happens to these dead accounts in the future? What happens when there are 100 million users and 30 million names are dead but occupying the blockchain. No one can ever be called these names again and their crap accounts of resteems and plagiarized content is forever and immutable.
I understand that the devil is in the details and all of these small amounts will add up somewhere but, for the most part the difference in curation return costs that will be lost by small accounts voting could be made up for multiple times with even a modicum of interaction with the community. Yet, they won't or they can't because they are bots. Should we spend our time worrying about a mass of accounts that are never likely going to try to be a part of the community or, should we work out ways to take the small amount of people who are trying and helping them grow more?
Perhaps the autovoters could adjust settings available based on SP so that no one can set a vote (easily) if it doesn't reach the threshold. If there are autovoter accounts who don't understand the new rules and adjust themselves, it probably means they are dead. Perhaps the small accounts that do trail could up their percentage but lower how many posts they vote on, perhaps every 3rd or 4th instead of every post.
Again, although these things are important, I don't think it is going to cause much harm to anyone who is actually active here as they will be engaging and learning and out earning their curation by magnitudes until they are at a point where their vote is significant enough to split or use to trail. As said and in my opinion, small accounts shouldn't be trailing and autovoting at all, they should be engaging as much as they can and, that includes with their vote as it will get them much further than any curation earnings.
So many of the churned accounts I would say are either alts that got tiring or people who thought they could earn here without doing very much. They got their Steemit delegation, set their autovoters and left. When they come back they should have no more than then because, they haven't chosen to put anything into the platform in time, energy, content or, investment. For the trails, the numbers from these accounts aren't going to be significantly missed.
It is hard out there for small accounts to be sure, but being drowned in shitsoup of inactive accounts that are resteeming BS isn't helping matters. Disincentivize poor behaviour and find ways to incentivize good.
Taraz
[ a Steem original ]
That part itself makes some sense, the weird part is that they accomplish it by taking (a 100% vote at 1.2 SP) worth of rshares away from every vote. Not just the tiny accounts, every vote in the whole system. Yours, mine, @freedom's, all will lose the same amount per vote. Basically every account loses somewhere between 1.2 and 3 effective SP depending on where they keep their VP, which doesn't matter too much to you and me and George but can matter a lot to someone with 30 SP, who is already at least a little committed here.
I'm not terribly upset by it, but it's kind of a bizarre choice, and I wanted to see how much it mattered.
Yes, I understand that it trims every one a little with the greatest effect on the smallest but, from a curation perspective, an engaged user will outstrip that very quickly. I don't know how to do the math on it but, What is the weekly curation expectation on 30 SP considering that currently, it takes 90 SP to get a 1 cent vote.
Lets say that 90 SP it is all used to vote each day at 10x 100% = 10 cents, 70 cents a week, 20% curation equals 14c. Double it for an awesome curator = 28 cents. That is with 90 SP. with 30, that is ~10 cents a week. Yes, significant if an awesome curator who can pick great content but, at current prices, that is 3 half decent comments on any one of my 30 odd posts a week. Which is easier for them to earn? Now, add them in the trail and they are voting behind other larger voters so their curation is zero.
Wouldn't it be easier to comment to not only cover the difference but, have the potential for much more? Wouldn't it be easier in the long run to be part of the community and far outpace the curation earning? I don't know but for me, engagement for a smaller account will likely make a massive difference to their value while curation return is negligible. If it comes to self voting, are they going to post 10 times a day and only self vote? That would cost them about 0.5 of a cent but, the community 9.5 cents. It might attract some more votes too though but is likely spam.
I think at the end of the day, it will bring more benefit than loss to the engaged accounts.
Very near zero. Unless you're frontrunning bidbots or voting in front of Utopian you need about 100 SP to earn curation at all. Below that curation rewards don't often get to the 0.001 Steem level they need to avoid being dusted.
But I don't care about curation earnings (look how I vote). I care about how much influence a user has over the rewards, and removing 10% of their influence is still something, even if that influence is small to begin with. Possibly especially if it is. For one thing it makes it even harder for something like the Minnow Uprising to work.
It didn't end up being a very high number, but even so removing $750/year collectively from Curie's rewarded posts doesn't help us.
Yes, I agree losing influence doesn't help but I also think at that stage, the influence isn't in the SP, it is in the ability to engage with others in the community. I would assume that there will be other factors coming that will cover this in different ways too as overall, I think the HF and Hivemind will improve the environment for small/new accounts and offset any potential harms.
I think this would take a lot of effort better spent reading a little and commenting well.
I think I made almost as much SP from one comment making fun of Jerry as I've made from curation in the lifetime of my account. Thinking about curation is almost always counterproductive.
Lol. Earning off Jerry gets double points.
"we work out ways to take the small amount of people who are trying and helping them grow more"
^^^^^^^^^^^^
I use an auto vote to do some of this. There are accounts where the content is just not my thing all the time, but their contribution to the community is outstanding.
People need to engage, not just newcommers, we are short readers (and and middle class). Using auto votes with very little SP is a waste of time and resources
Some of those accounts have earned less in the last 3 months autovoting than I would give them for a half decent comment.
Funny, as I was reading this post I had the following pop up on Gina:
That's a solid Reputation right there! :D
As in other comments elsewhere, i'm not overly concerned, even taking into consideration comments here by tcpolymath. The % loss is bigger for smaller accounts, who may indeed be committed on working their way up. I think this is worth the hit to cut some of the BS out :)
I think that the actually committed will hardly feel it if at all. I am not sure about my numbers above but, i think they are close enough. The committed are going to engage, I know, I spend a lot of time upvoting them n my comments section :)
Yes, this is extremely trivial to the vast majority of users. I'm not sure what BS it cuts out, though. Votes under 50m rshares already don't count (you'll even get an error if you try to cast one) so the only new things are taking them away from everyone else and smuggling a little superlinearity into the curve, and taking away the error message so we're not telling people their vote didn't work anymore, just letting it fail invisibly. Neither of those seem like benefits to me.
The good point about these dust votes not meaning anything in the future is all that dust is going to add up to a lot when it gets redistributed to the reward pool for the users who have a vote that is worth literally something.
I don't get why take it out on everybody else joining who are legit users coming aboard. If it wasn't for a certain person who talked me into coming over here I more than likely would have gave up myself but I stayed to support him in his endeavor though that support wasn't financially profitable that wasn't the point. Seems there would be a better way to deal with inactive accounts. Personally I think they should eliminate auto voting altogether but evidently it must be advantageous to the higher ups otherwise they would.
Autovoting is advantageous to everyone in some way. It really isn't taking it out on newcomers as I see it and those who are legit will more than make up for it. For example, I will upvote your comment 4 cents. That is more than a new account can most likely earn in curation returns in a week. If they engage well, they will do better than if they don't. Automation however isn't going anywhere in this world, it is increasing.
Without any incentive plus being kicked out of the system because of bandwidth how do think that will encourage new users unless they are going to get rid of the bandwidth problem.
HF20 brings in a new system for bandwidth. It should fix the issues that new users experienced last time.
https://steemit.com/steem/@steemitblog/blockchain-update-4-resource-credit-implementation-details
i agree with the prinicipal with what you're saying - but i'm not sure that the way it's being implemented is very logical (although admittedly I haven't read about HF 20 apart from what you've written here) - a limit of 1.2SP for voting might seem logical now - but is it future proof? what if the price of steem goes to $1000 - and that 1.2SP means you all of a sudden need $1200 invested before you can make the most miniscule of votes. It makes the whole thing seem not very scalable.
There was talk of making the minimum a witness-set parameter, and if it sticks around long enough for Steem to go to the moon I imagine that will happen.
Is it the percentage of the vote You cast or the amount of the vote that is the important factor.
Example: The majority of my votes are at 10-14 % votes. which is just over 0.011
The others votes range from 25% to 100% depending on my voting power %. and who I vote for
Unless you're @hr1 it really doesn't matter very much. At current prices the per-vote loss is 1/150 of a cent, so the cost of voting 70 times a day instead of ten is about a penny every two and a half days.
Ok, but that takes care of # of votes.
Maybe I am trying to solve the big picture instead of all the little pieces first.
As long as it goes over 1.2 Steem it is recorded to the chain.
So a 1% vote with 120 Steem Power @ 100% voting power. You won't be affected very much, nor will anyone that is here longterm really in my opinion. On a very large scale, it will show large numbers like the 750 TCpolymath mentioned for curie in a year but, that also is a tiny percentage of the massive amount the distribute yearly. It also goes back into the pool so it is still distributed as far as i know.
OK got it and it makes sense. Thanks
That will incentive new users to powe up and engage more on the platform.
To tell you from experience of when I was trying to identify and eradicate the bots when they first arrived, I'll tell you that the dedicated bunch will consolidate (and they're all whales anyway), so identifying the bad actors is extremely difficult. Not to mention the amount of legitimate people who are using voting bots.. as well as the bad actors who are using voting bots.. and they're all augmenting their earnings with garbage posts. There isn't enough active curation to find all of the junk.
If someone were to build an AI to monitor real posts and real users who provide real responses to their commenters, I bet that we could eradicate most of the bots and spammers and scammers. Many will still remain though.
Until then, we're going to be stuck in a perpetual cycle of the bots and bad actors winning because the whales who are trying so hard to continue augmenting their income with their 100 different bot/alt/promotion/bot-using accounts look so similar to those who are real people. It's effectively talking about trying to fight the people who own the majority share and who run the voting bots and who have the most to lose if Steemit became the platform that it was meant to be. It's a catch 22. Lose all of the whales because you've made it so hard for them to earn "easy money", or let the vicious cycle of corruption continue and lose the opportunity to grow. Because let's be honest.. Steemit is falling apart and it isn't pretty.
Call me cynical, but with the amount of effort that I personally invested in trying to identify scammers and spammers (at a low success rate) while leveraging my extensive professional experience in performing exploratory data analysis for fraud and abuse (and being quite successful in those endeavors), I don't see anything good coming out of HF20 for the little guys...