Do you blow your load in 1 minute?

in #steem5 years ago

So the vote auction time is going to move from 15 minutes to 1 minute with HF21 and I wonder if it is going to be much better for authors - it is likely better for the bots of course.

After HF20, it moved from 30 minutes to 15 minutes and people have continually been adjusting their vote times to maximize their curation, downward. Some people don't seem to know what the auction is but essentially (at current times):

if voting at 0 seconds, 100% curation return goes back into the pool
if voting at at 5 minutes, 66% curation return goes back into the pool
if voting at at 10 minutes, 33% curation return goes back into the pool
if voting at at 15 minutes, 0% curation return goes back into the pool

So why not vote at 15 minutes? because of course the early voters get additional benefits when votes stack on top of their. I have seen front running large votes getting a 400% curation return, but this is only on small vote values. There are plenty of posts about this if you want more technical details.

What is important to remember is this bit though...

goes back into the pool

 
What does this mean? Well, if for example there was a vote of 10 dollars and was the only voter, under 50/50 the author would get 5, the curator 5 worth. But if the curator voted at 0 seconds they would get 0 curation and the 5 that they would have got goes back into the pool. That 5 then gets distributed on other votes instead. This means that other vote values are increased based on what is in the pool.

When people are trying to maximize their curation, they are actually going to have to give up curation returns in many circumstances while the author is always guaranteed the 50%. Most large curators on this platform get currently between 15-20% curation as generally they vote later and, there are no votes that can stack on theirs but small voters can quite easily get 40%+.

When it goes to 1 minute auction time, it means that pretty much all the bot votes will come in within the first minute and many at 10 seconds, 30 seconds, etc losing some curation that goes into the pool. However, unless they don't care about curation and just want to give the author 50% they are trying to frontrun larger voters to come. The curators to come can however better see whether they still want to vote on that post at all and, some will instead move their votes on to other posts, they will look wider.

On top of this, the curation maximizers will continually drop their curation time to frontrun those who are trying to frontrun them. A race to zero to to speak. This will continually put more into the pool for distribution on other votes. Then comes the downvotes that are also going to land on bidbot users who abuse them. This puts more into the pool again.

While I don't know what will actually happen, the more that goes back in the pool, the more authors will get an as it shifts from 25/75 to 50/50 curation return and there is likely to be much more risk in purchasing votes meanings, disruption to demand and supply might be affected as more people pull their delegation and play the voting on content game again. While it is hard to predict just what will happen in general, moving the curation to 1 minute brings in a few more variables that make that prediction harder again.

Bots are of course written according to prediction of what will happen and they will act accordingly. The bots on Steem aren't overly sophisticated so they aren't going to factor in all aspects, making them more human-ish. This might mean more returns to the pool and that means that it will go into distribution again on all the other votes. That means that with the authors guaranteed their 50%, it is always 50% of more than before the return to the pool. Authors that care about earning should also encourage downvoting of whatever a person considers unworthy of the reward. A blurry photo with a 300 SBD payout perhaps?

With one minute reverse auction rounds,

if voting at 0 seconds, 100% curation return goes back into the pool
if voting at at 20 seconds, 66% curation return goes back into the pool
if voting at at 40 seconds, 33% curation return goes back into the pool
if voting at at 1 minute, 0% curation return goes back into the pool

Where are you going to set your bot, which bot do you use and how accurate are the voting times, and how many votes are going to packed onto those 20 blocks?

I do not think this curation change does that much all up but it does bring in new games that some people enjoy playing by frontrunning large votes from bidbots, and then there is the downvoting aspect that is going to potentially affect their curation return anyway. It brings in more complexity to the point that no one is going to likely be able to maximize efficiently, there are always going to be losses and, that means more randomization of outcome as everyone takes slightly different paths in an attempt to satisfy their wants.

Randomization drives evolution, inefficiency is opportunity.

Taraz
[ a Steem original ]

For me? I am just going to vote when I vote - same as now.

Sort:  

Have an upvote for the title.

I wish I got upvoted on titles only. I have lots of good weird ones :)

Lol, you're headlines have gone quite sexual lately.

Posted using Partiko Android

I wear a lot of purple and listen to Prince.

i'll vote all at zero seconds. Give the reward pool 100% of the rewards, it's one of a small few not complaining about something lol

Will be interesting to see how many will do that. frontrun and take curation for no personal gain :D

It really doesn’t make sense to me...Most of the posts are way longer than 1 minute reads...

So I guess this is going to be a race of who’s gonna vote first in order to maximize their curation rewards only. As for the post itself?

Like their are pushing people to vote first read later

Oh...screw it, we can always read it later or tomorrow or the day after tomorrow...right? Sigh...

So I guess this is going to be a race of who’s gonna vote first in order to maximize their curation rewards only. As for the post itself?

Automation is going to happen anyway. As will people who want to read anyway.

Most of the posts are way longer than 1 minute reads...

The idea isn't that you read in less than one minute. It is to take all the bot/autovotes which happen anyway and stuff them all into one minute and get them over with. After that you can take as long as you like to read and vote and not be racing to compete with bots (but you will still be competing with other readers).

Most people not involved with vote bots and autovoting will be able to just ignore the timer now.

Just a thought, but if voting becomes a race to the bottom in terms of earning curation rewards then it really doesn't help the best content get discovered.
"Blowing your load in a minute", as you elegantly put it, means that no one is actually reading the content of the post before voting which to me kind of defeats the purpose.

Yep, but it depends on who is doing this too as the content that should be rewarded would still get voted on top by people like yourself, the content that is getting frontrun for curation alone won't. Plus, you might choose that it is overrewarded and remove some of those curation returns with your 2.5 a day and instead place it on something that you think does deserve it.

I don't think it is going to pan out so accurately for curation as the voting bots will not be to the second. Not sure how it will all play out yet though, just speculation at this point.

Yeah, I was thinking about this a little more and it's true. I don't think the changes will influence the way I already vote that much. I guess the point is to try to encourage larger accounts to be more active with their stake/votes that aren't presently manually curating.
We'll see how it goes, right? I doubt the changes will solve the culture of greed but it's step in the right direction I feel.

The greed culture seems to be an endless cesspool, but perhaps for those who don't extract into it - a decent approach can be found that brings benefits without harming others.
Baby steps.

The problem with this now is steem auto voting. It is not accurate enough and too erratic on voting. I am hoping more use it as I will leave it alone after the hardfork. I will do better manually than relying on something that cannot guarantee you a set time. Luckily I have given this some thought already and know what to do and literally can't wait for this to kick off.

The problem with this now is steem auto voting. It is not accurate enough and too erratic on voting

problem? :D

Luckily I have given this some thought already and know what to do and literally can't wait for this to kick off.

It is going to be interesting because the potential is that some of what you vote on might have to stop buying votes after a few downvoted losses. Who knows how it will all play out though.

I agree and I think the first week you leave the bot posts alone. I don't understand who is going to use a bot as you are going to be handing money away. You are looking at 25 % less returns than you are now and are just hoping that people will cover your losses by voting because of the bot. Certain users who put fortunes on are in for a rude awakening and it will be good to see how true their following is or if they have any followers at all.

Steemauto at least only allows you set votes in 1 minute intervals, so I guess everyone will just have to set at 1 minute.

I don't know if there are other auto curation tools which allow second intervals?

Will you be in attendance for the Witness show on Sunday - 6PM GMT btw?

So much to discuss, or listen to for me TBH and resist getting snarky in the chat.

I don't know if there are other auto curation tools which allow second intervals?

Good point @revisesociology. Second intervals would be good even now and I believe it will be more important after HF21. Let's see if they are going to implement it.

you think they can. It depends on many things and I don't think most have the hardware to do it.

Will you be in attendance for the Witness show on Sunday - 6PM GMT btw?

Home alone with the daughter today so unlikely. I post when she has a nap :D

Not quite 7 hours from now, yeah?
Did you see Steempeak has given authors a countdown timer option, so we don't need to worry about timezones :)

It's going to be a very complex system. I will set my autovotes on some authors at 1 minute but going to vote manually on other authors like before probably.

yep, it is likely that many won't notice too much change in curation (other than the 50/50).

Don't care I vote for content I like by people I like, when I see their posts and have time to, you know actually freaking read them!

I'm a Minnow, as of today actually, I accomplished this organically and it took over 8 months to do so. In that entire 8 months I've made maybe 10 SP so another potential 25% boost isn't worth me gaming a process that dehumanizes steem and does nothing in regards to engagement. The human touch is vastly more attractive to me than jostling for scraps with joe blow auto vote or botty the brainless wonder.

In fact I didn't even no this time limit was a thing until eip exposed this. Because, steem has such a steep learning curve and there's no emphasis on assaulting new victims with the complicated economy(rightly so) you get 8 month daily users that are clueless they're clueless about curating..

you know actually freaking read them!

No one is reading my posts in one minute.

The human touch is vastly more attractive to me than jostling for scraps with joe blow auto vote or botty the brainless wonder.

Other than a few over a 2 week period in 2017 for testing , all my votes are manual. so far I have 54,618 votes.

In fact I didn't even no this time limit was a thing until eip exposed this. Because, steem has such a steep learning curve and there's no emphasis on assaulting new victims with the complicated economy(rightly so) you get 8 month daily users that are clueless they're clueless about curating..

Lots of people are clueless about many aspects and I am hoping that this will all make more users clueless so that it is harder to game, or easier for everyone to game - making it harder to game.

Haha, well said indeed!

And yes I'm a quick reader but 700 words a minute or so is out of my league.

If I had a large stake i would definitely be trying to learn how to best maximize my votes so I'm not knocking the hustle.

That's an amazing amount of votes given, I commend you, and I follow for reasons of realness as demonstrated by your daily efforts over years of steeming..

Posted using Partiko Android

Optimizing one's vote within a one minute window will involve much more randomness than within a 15 minute window. Why? Because there are things like network delay that could easily be in the seconds or occasionally even tens of seconds. However, it's clear that all the bot votes will come before the first minute is up. This will definitely favour autovoting. I wonder if I should really come around to writing a voting bot of my own. It could be an interesting challenge.

I've been saying for quite some time now that some Steem-Engine token or future SMT could try not taking the voting order and reverse auction into account at all. Only the voting power used would matter. What people do is read what they read (their own feed) anyway. Guaranteeing everyone the same ROI from their SP regardless of any other factors might free them to reward what they enjoy without any interference from bots. That could work better than what we have now.

Optimizing one's vote within a one minute window will involve much more randomness than within a 15 minute window. Why?

Yep.

This will definitely favour autovoting.

I think so but, what are they autovoting on? put flags on top of the usual suspects and things change again.

That could work better than what we have now.

It could but in general, the small accounts grow at a faster rate than the large as the large can't frontrun really. While people scoff at the small amounts, when there are so many small accounts, it adds up over time.

I think so but, what are they autovoting on? put flags on top of the usual suspects and things change again.

They could autovote established and popular authors who do not use bid bots.

"That could work better than what we have now."

It could but in general, the small accounts grow at a faster rate than the large as the large can't frontrun really. While people scoff at the small amounts, when there are so many small accounts, it adds up over time.

Front running is but a small factor in the growth of small accounts, although it would become more significant with the 50/50 split. I think @ocdb is by far the best on Steem so far to help smaller accounts along. It has turned many Redfish into Minnows and Minnows into Dolphins. Note that @ocdb a centralized curation project in the sense that getting whitelisted is a decision made by a handful of people as well as getting delisted. Musing was another island of centralization that worked tremendously well at spreading stake.

My vision of successful Steem is a collection of communities where community standards are enforced by the members whose social standing depends on being a good community member or apps with a lot of centralized stake that are capable of effectively maximizing quality for the benefit of the company running the app as well as rewarding quality accurately.

Lemme try to time this right as I've seen you get your votes right at the 15 minute mark..

Posted using Partiko Android

Coin Marketplace

STEEM 0.18
TRX 0.14
JST 0.030
BTC 57983.59
ETH 3132.93
USDT 1.00
SBD 2.44