BidBots, Abuse & Curation: Addressing the Elephant in the Chain, Gazing Into The Abyss.

in #utopian-io5 years ago (edited)

2017-05-10-18-54-47-900x646.jpg
Source

“Beware that, when fighting monsters, you yourself do not become a monster... for when you gaze long into the abyss. The abyss gazes also into you.”
― Friedrich W. Nietzsche

Delegating to Bidbots as a Refusal to Curate

Users that delegate to bidbots are intrinsically users that refuse the present curation system in Steem. The SP "BidBots" have is nothing more than the SP of accounts delegating to them, whether the owners, other accounts or a combination thereof. Said accounts, by delegating to the bidbots are refusing to participate in the Steem content discovery system, also known as Curation, presumably because they don't think the curation rewards are appropriate. Instead they choose to "maximize their stake", in other words: Making as much money as practically possible without self-voting.

Now, why is self-voting badly seen in the first place? Well, it is because self-voting is seen as a refusal to participate in the content discovery system of steem, aka curation, which is the foundational stone of the platform. Self-voting in essence is saying: "why would I reward anyone else if I can reward myself instead?"

Delegating your SP to a BidBot is, in practice, nearly the same, since in both cases your stake does not contribute to the content discovery system. Moreover, if you delegate to BidBots, your stake interferes with the content discovery system, not only does it not contribute to it, it works against it.

Self-Voting vs. Curating vs. Delegating to BidBots: The Numbers

pay937884_960_720.jpg
Source

Let's delve into the numbers. Let's say you have 40k SP, that's about a $1 vote, what would happen if you vote one of your posts and nobody else does? At payout you would get $0.25 worth of SP in curation rewards plus you would get $0.375 worth of SP and $0.375 worth of STEEM/SBD in Author Rewards. That's $0.625 is SP and $0.375 in liquid STEEM/SBD. In other words: you get the full $1.

But, self-voting requires posting content, either from your account or from an alt, and those posts are subject to flags by the community observing the self-voting. Even posting what is colloquially called "shxtposts" takes some time, and your reputation will suffer if you vote said "shxtpost" whether in your own account or others. In short, it doesn't have a guaranteed return, you will suffer reputational damage and is not sustainable in the long run.

What would happen if you voted somebody's post with your $1 vote and nobody else voted for it? You would get 25% under the present curation rewards system, $0.25 worth of SP.

Now, let's see what would happen if you are delegating your 40k SP to a BidBot. It would vary of course from bidbot to bidbot, from window to window, with price fluctuations, etc. but to give the reader a rough average number if you delegated the equivalent SP of a $1 vote you would get in liquid (STEEM or SBD) around $0.50 per each one of your $1 votes. Yes that's double what you would get if you curated and half what you would get if you self-voted.

Have you heard of the EIP? 50/50?

Under the proposed new content discovery/curation reward system curators would get instead of 25% in curation rewards, 50%. That means that under the proposed system people that curate will get at least as much as in average they are getting by delegating to bidbots now: $0.50 for each $1 vote.

The EIP is going to incentivize curation in the sense that since curation rewards are going to double, certain stakeholders that weren't curating due to not being satisfied with 25% curation rewards now will begin curating. This assumes those stakeholders are indifferent to the liquidity of the rewards. Remember, in average delegators to bidbots get $0.50 in liquid for every $1 vote they sell, but under the proposed 50/50 of the EIP they will get $0.50 in Steem Power, non-liquid rewards that take 13 weeks to be withdrawn.

Additionally, even assuming these accounts that were delegating to bidbots were to switch to curating, it is important to note that these accounts have demonstrated that they are interested first and foremostly in maximizing their rewards over contributing to the content discovery system and so the curation they will perform will naturally again be focused on maximizing curation rewards over contributing to the content discovery system.

Now, that 50% in curation rewards is overall, distributed to all the voting accounts, but there is also the new convergent curation rewards curve, what does that mean? It means that if you have a $1 dollar vote and you vote a post you will get $0.50 if the post is at payout at $6 or above, namely, the curation rewards go linear, equal rshares per steem in voting SP at around a post reward of $6, if the post rewards are below that, the voter with the highest SP gets more rshares, a bigger, above linear share of that 50% of curation rewards, under the concept of having contributed more voting power to the post, and those voters that are not the top voting power contributor get a sublinear share of the 50% rewards.

This means that to maximize their curation rewards, "small" stakeholders will be incentivized to vote posts that are or are expected to be at $6 or more at payout or posts in which they are the biggest voter. Conversely this means that "big" stakeholders will be incentivized to vote posts that are not and are unlikely to be at $6 or above at payout, in order to maximize their curation rewards.

Let's Go Deeper

Iceberg20150522.jpg
Source

"Small" stakeholders will be incentivized, in order to maximize their curation rewards, to vote for posts that are, or even better, will be at $6 or more in payout.
How to achieve this you ask?

  • Voting for already discovered authors, popular authors.
  • Voting for accounts that systematically bidbot their posts above $6.
  • Voting for posts that already have $6
  • Joining curation guilds' trails that vote posts to $6 or more.
  • Voting posts that are below $6 in which they are confident they will be the largest voter, as in for example low quality (aka sh**posts).

Conversely, "large" stakeholders will be incentivized -in order to maximize their curation rewards- to:

  • Vote content that they feel confident wont exceed $6 at payout in which they reckon they will be the largest voter.

How to best achieve this you ask?

  • Voting borderline abuse aka subpar content aka sh**posts that never exceed $6, you need predictability, so you are incentivized to vote accounts committed to posting subpar content.

Your Voice is Worth Something, But Money Talks and You Listen.

Frankly speaking, this is not looking good for the content discovery system or for the antiabuse community. These proposed changes create more problems than they solve. The EIP incentivizes practically all the behaviours that a healthy content discovery system should disincentivize, and disincentivizes practically all the behaviours a healthy content discovery system should incentivize. In short: it is a clusterfuc*

On The Other Hand...

On the other hand this will greatly facilitate, it will make it easy for most stakeholders to maximize their curation rewards and will incentivize major stakeholders not to dump their SP. It is, we reckon, a last resort measure to prevent a massive steem price drop due to big stakeholders dumping. Yes, at any cost to all other factors.

Subversion

In this context, the subversion of the system seems to be the only possible solution to make the content discovery system appear to work. In this proposed system to be implemented in HF21 -and make no mistake, it will pass, for as said above: your voice is worth something, but money talks and you, listen- the only apparent solution to keep quality content authors in the rewards loop and to achieve the so much longed for trending page with quality content is to subvert the system, turn it over, make it do what it is not supposed to, hack it so to speak.

One of many forms to do this is the following:

For curation guilds to partner with certain bidbots, so that:

  • Many small delegators that care about quality content join their guild's trail and assure themselves by joining forces that they will vote quality content and posts over $6 at payout, and therefore a linear share of the curation rewards of the posts.
  • Posts that are curated by the guild are promoted for free by the guild via bidbots, improving the visibility of quality content beyond their organic SP, in that way making trending great and also further improving the curation rewards of the organic SP in their trail via the posterior large promotion votes.

This could be achieved by a process in which the authors posting for those guilds set an account owned by the guild and the bidbot as beneficiary and then using a rewards liquification system such as @likwid. In this manner the guild first curates the posts with the organic SP of the trail members, then buys a juicy vote from the bidbot to promote it to trending. At payout all rewards are liquified, the guild gets back what it invested in the promotion, and shares the promotion profits with the author, who in turn also gets the author rewards corresponding to the organic SP of the curation trail.

The Way Of The Future

downloadfuture.jpg
Source

The reason these issues regarding the EIP were dealt with in this post is because as an antiabuse community we need to not only know where abuse was in the past and where it is in the present, but also where it will be in the future in order to prepare for it.

BidBot, Reward Pool and Content Discovery System abuse are categories that are somewhat borderline and controversial, but nonetheless we believe it is healthy to begin talking about them and about how they can be approached, engaged, minimized.

As denoted by the subversive possible solution explained above, we believe the solution lies in antiabuse initiatives, antiabuse friendly bidbots and curation guilds joining forces.

This is Jaguar Force,
Reporting directly from the Jungle.
Onwards!

Sort:  

Howdy @jaguar.force,

This is quite a voluminous post and I can you tell that a decent amount of time has been spent. I really appreciate your perspective about a lot of these issues but there are a number of claims here that are not very easy to verify. This made it difficult indeed for me to judge.

This is largely due to what appears a somewhat speculative nature of the claims but the technical / analytical nature of this post. This tends to make the post esoteric and not easily understood by the layperson. I agree with much of your premises but I am not sure about the conclusion particularly when it comes to curves and math. I just may look into this on the HF21 testnet though.

It may be advisable to produce a separate post for an analysis submission and then reference it in the antiabuse post in the future. I believe there is a HF21 testnet already up and that is how you could refine the data as you know It will be subjected to scrutiny when the witnesses look at it.

With that kind of attention, you need precision with the data backing up your claims. It is good that we are, at least, talking about this in antiabuse.

Your contribution has been evaluated according to Utopian policies and guidelines, as well as a predefined set of questions pertaining to the category.

To view those questions and the relevant answers related to your post, click here.


Need help? Chat with us on Discord.

[utopian-moderator]

Thank you for your review, @anthonyadavisii! Keep up the good work!

Great explanation. First time I’ve fully understood what’s going on with HF21. Hopefully SCOT & Snax tokens will save us. I’m already earning as much in combined other tokens as in Steem for quality posts.

Posted using Partiko iOS

Interesting post, and well-written.
I agree with much of it, absolutely.
Until now I had no idea of the significance of the $6 USD price barrier. I'm a small fry content-creator who has to worry more about the $0.02 cutoff point.
And yes, the way things work on Steem is an Animal Farm (the exact opposite of what it set out to be). Good content fails to be rewarded, most of the time. Shitposts do get rewarded, most of the time.

Bidbots are gross. I skipped on upvoting an otherwise great post last night that would have trended without the bots at all. In the end the most powerful deterrent to their use is people voting with their feet in who they support on here.

I hope the bot owners realize they will be able to make as much money by just letting guilds curate on their behalf instead of being a blight on the system.

Hi @jaguar.force!

Your post was upvoted by @steem-ua, new Steem dApp, using UserAuthority for algorithmic post curation!
Your post is eligible for our upvote, thanks to our collaboration with @utopian-io!
Feel free to join our @steem-ua Discord server

Been going rogue from the start. Manual curation. Not as profitable. I guess that makes me an idiot for arguing for tighter self-discipline ethics that have a costly price.

I am agree about bidbot but not about self-vote. If i like my post, why not vote myself? For both cases the problem Is that they are permitted from Steem algorithms. Don't forget Steem Is a decentralized system and therefore rules are announced at the beginning from founders and translated in programming code. To change them there is a consensus rule that produce new hardfork. Now, is very simple change the code to not permit the author of a post to vote himself but after 21 hardfork It Is not happened. Why? Perhaps nobody of witnesses et al. think it's an abuse. About bidbots I think Is difficult to fight them through the code and I have not an answer at the moment.

Where is your reference about the 6 dollar statement? The reward and curation functions are smooth increasing functions. Also, it does not make sense to talk about dollars only VESTS matter.

The new reward function satisfies the same fundamental qualitative properties as the current reward function. It satisfies the indifference principle which makes it a viable function. See the posts in this account for the related theory -> https://steemit.com/@theoretical or this post for a recent version -> https://steemit.com/steem/@eonwarped/curves-did-you-know-zx412ln2

You need to have mathematical rigour when making statements about an economic system. Otherwise, the validity cannot be guaranteed and statements become opinions.

The source for the 6$ statement is this: https://steemit.com/steemit/@steemitblog/steemit-update-hf21-testnet-sps-eip-rewards-api-smts
That being said it is true that these issues are mathematically complex and so we are in the process of onboarding a Math advisor to Jaguar Force in order to understand and cover these issues in a more rigorous manner.

I cannot find the statement anywhere. Maybe I am missing it. Could you cite the paragraph?

Posted using Partiko Android

Link corrected.
" As shown in the charts, 2e12 approaches our current linear rewards at roughly 16 STEEM, which is why we feel it will provide a meaningful and balanced change to Steem’s economic distribution."

The labeling on that graph is a bit unclear but I think it measures the reward in relation to the current reward curve. The property that something will become more rewarding in relation to the current reward function does not imply anything since the new economic system will be different.

Posted using Partiko Android

See the comments in that post, smooth, ats-david and josephsavage appear to be interpreting it in the same way, as in: The rewards curve goes linear for posts around $6 and above.

I don't see them saying that they will use the 2e12 curve:

The above chart demonstrates 3 different constants fed into the convergent linear rewards curve and their effects on comment payouts. As shown in the charts, 2e12 approaches our current linear rewards at roughly 16 STEEM, which is why we feel it will provide a meaningful and balanced change to Steem’s economic distribution.

They say they think 2e12 is good. But, it doesn't seem like a done deal. They do say that they really want this type of reward function. But the constants will determine when the new function is worse or better than the current linear function.

Yes the information provided is as usual ambiguous, but by saying "which is why we feel it will provide a meaningful and balanced change to Steem’s economic distribution." we feel it is being implied that is the curve that is included in the proposal and even in the testnet, and although It is not a done deal as you say, it seems to be the curve to be implemented if the HF21 passes.

Coin Marketplace

STEEM 0.18
TRX 0.13
JST 0.029
BTC 57400.65
ETH 3108.60
USDT 1.00
SBD 2.42