If they only knew... - Bidbot abuse

in #steem6 years ago (edited)
So yesterday we had a great conversation on the MSP Discord server. I was invited to discuss this along with @clayboyn, @r0ndon, @isaria and other prominent leaders of Palnet including Steem Witness @lukestokes. I often think that the main reason why people abuse bots, why they bot up in insane numbers is because they don't really understand what they are doing, and maybe the solution is not that complicated.



img src

Maybe, saying it's not complicated is not entirely correct, but hear me out. The way that things seem to work currently, is that we have an uninformed part of the steem population overusing bots to promote content with very little social value. Trending posts of the color challenge types are hardly something that edifices the value proposition of an immutable blockchain. However, I think I know why these people use the bots, why they think this is the way to "game the system".

Mathematical Enigma


More than once just this week, I've had the opportunity to explain to some Steemians how much money they are losing using bidbots. In other words, they think wrongfully that they are actually making money on top of their investment, but this is far from the truth.

It may not be clear to most people, specially when they start out, how this whole thing works. Curation, early voting penalties, flags, etc. So, many of the new users see the payout on the post, on the bottom of their publication and simply say - "Well, that says more than I spent... so I'm good".

If you think I'm exagerating, it might be because you've not tried to teach someone how to analyze the numbers. Even Steem Supply is not 100% reliable, specially at the moment when the blockchain is attempting to curve the circulating SBDs by printing liquid Steem.

I guess, it might be insulting of me to say that some people are mathematically impaired. But then again, I know a few that admit it, and simply say almost with pride "Math is not my thing"- To those I would advice, to seek out someone they trust and verify the information I'm sharing.

What would happen if they knew?


I think they would tone it down, and maybe considerably so. In other words, if I knew I was going to lose 10% , 20% of the money I was about to use to promote a post, I would probably not be so careless of the content. After all, the idea is to gain organic support and increase the number of followers.

In this sense, the bidbots become not a money printing machine, as many Steemians seem to think they are, but a promotional tool, as they should be. You don't send to trending a post to triple your money, you send it to promote your absolute best attempt to make the best content. Maybe, I can get behind that idea.

How do we teach them?


Here is the biggest challenge of them all. My approach is not very effective if I'm to be honest. I've reached out to them on Steem, later on discord and I've taken the time to explain the math in private. Why would I do this? - Well, because I think its more effective than just flagging and much more effective than insults.

But, I've been thinking for the past few days on a new bot. Yes, I know what you are thinking and I also happen to know most people hate the transparency bot out there. But give me two more seconds to explain:

What if we had a bot, that after 7 days, after the post had finished paying out, it would go to a ex-trending post and tell the user X something like this:

"Hello there @userwhousedbotslikeamaniac I'm the investment adviser bot, I noticed you used 978 SBD's to promote your post. I just wanted to let you know, that you lost 27% of your investment on this promotion after payout"

And take its leave....

You see, it would not be on every post, specially not on every post that used bots, but just the one's that made it to the top spots of trending.

Development


I'm not a programmer, I simply dabble a bit with the scripts, but I wonder if the task is a lot easier than I think it is. After all, since the bot would be posting results, the only calculation would have to be done comparing the initial transfers against the total payouts.

I wonder if this would be something a developer could get behind. I'm sure there are a few programmers out there that could pull this off in a day or two, and I'm willing to bet most if not all, also wish they could help clean up the trending page a little bit.

So tell me, What are you thoughts on this?


Other posts by yours truly

• What's the deal with Ulogs?
• A Question for the Bid-Bot Owners - Dlike
• Dlike solves nothing - Making socks easier to put on
• Fakenewsd by my ISP provider
• Join me tonight for Songwriter Shop Talk Featuring @shookriya

Sort:  

I'm conflicted about educating people about vote-buying, because there's clearly a much larger market for buyers than sellers, and in some sense it's a zero-sum game; part of why I'm able to make money is that so many people are willing to lose it. If they were more sophisticated it would be harder for me to do as well.

I have a somewhat philosophical justification to want to clean up our trending a bit. It has to do with the challenge all cryptocurrencies have to face. Which is the convergence of wants. How does a crypto/community increase its user base. More people wanted to use, to trade, to buy with the crypto.

The almost obvious answer seems to be, by making that ecosystem pleasant to participate of. As much as possible that is.

So lately, I've been thinking that it's not really about guarding the reward pool itself, the same amount of steem will be printed regardless. Its about attempting to keep the convergence moving in the right direction.

I hope that makes sense, the concept is somewhat abstract, but I can see it clearly in my head, at least after three cups of coffee.

It does make sense, I'm thinking very much the same thing. Personally I don't think Trending has much of a role, if any, in the enjoyable ecosystem, though. Its equivalent on every other social media site is also complete crap, so I blame the core concept, not the bots.

Heh I distinctly had you in mind when I decided not to mention it. ;) It's quite tricky!

I'm curious now, do you have broad data on what user percentiles are profitable? Intuitively I think it's probably better than poker where the top 3% are winners, but probably not by a whole lot.

If you do, putting up a post with a graph that doesn't identify anyone individually but shows the profitability of users by percentile might be really well-received. In fact if you don't want to I might go and find that data myself.

Ah I don't have this kind of data, would likely want to do that with SQL instead of my primitive per user blockchain data. Hmm and come to think of it, I don't compute the effects of the skewed curation early vote penalty. Silly. I should make the adjustment and see how that changes my previous botting figures!

It would be interesting to see in aggregate. Go for it, I don't have this ability with my current methods. If I did it I probably would want to tap into steem SQL.

I should make the adjustment and see how that changes my previous botting figures!

If you do I'm really curious to see how much it changes my profitability, since I use it very intentionally.

Go for it, I don't have this ability with my current methods.

Oh, I'm even farther from being able to do that than you are. But I'm pretty interested, and we'll see where that gets me.

We hope this traduction helps to bring this information to more people and educate then, and meaby catch a couple of developers:

https://steemit.com/witness-update/@castellano/si-tan-solo-supieran-el-abuso-de-los-bid-bot

good stuff... i'll resteem it too... :)

They play with the system and deform this social network, however not everything is pessimism, there is excellent content and excellent communities that provide good content ... Anyway, we will see how this social network evolves in the future until now we are doing well.

Sin duda alguna... sin embargo mi punto, es que creo que hay muchos usuarios que piensan que estan ganando dinero al abusar de los bots, y la verdad es muy distinta.

Asi que me la idea es darles esa informacion que les hace falta.

Excelente amigo, bueno la información es muy útil, yo uso el robot para traducir mi ingles es 30% aceptable... jejejeeje muchas gracias por escribir en español. jajaja

te entendi sin problemas. :)

Sounds good. It's really just information. You can choose to ignore information, so it's not like anything's being imposed on anyone.
I'd be interested to know how many are labouring under this impression.
I guess if they've followed some step by step instructions somewhere it's certainly possible that they are using bidbots without first having a real understanding of the basics.
I like it.

I distinctly remember someone complaining to @themarkymark because they had a negative ROI, as if that meant the bidbot was a scam.

It shows you the expectation people have when they use them. Meaning, in the mind of that particular Steemian, if he was not printing more money for himself, the service was not provided.

It makes you wonder how that came to be.

What bidbot operator sees himself/herself as a money printer? I'm seriously wondering.

Yeah, I've always been in two minds about bidbots.
If we wanted to kill them overnight a return to quadratic rewards would do the trick. Since there's apparently no broad appetite for that, I'm happy to default to 'code is law'.
Ultimately if the game has dumb rules, you don't get upset at the guy winning the game with dumb rules, you change the rules. I can't do that, so I'm not losing any sleep over bidbots.
They were the completely foreseeable consequence of linear rewards.
Play stupid games, win stupid prizes.
Happily there's enough competition in the space that they're not lucrative any more, at least not for the customers; and other options like steemhunt etc will offer new and exciting possibilities for SP hodlers, so it'll solve itself in time.

OK, we're definitely not clones. @Tarazkp, you can rest easy, Matt thinks SteemHunt is a good thing and that quadratic rewards would kill bots.

I haven't looked at steemhunt closely. It may be the devil; but it's another direction in which to delegate one's SP, which can't help the bidbots (unless I'm missing something).
I'm very confident quadratic rewards would hurt the bidbots.
If I can halve your post reward using a quarter of the voting power you just paid for, that's going to suck for you, and you're going to stop doing it.

Hmm, I have a local script right now that analyzes historical bot input and output and computes with all the curation timing bells and whistles how much sbd and sp they got (doesn't handle debt ratio though, and probably should add it given how much we are in there).

I'm wondering how many people will bother with it in this form though. I end up having to some data to spreadsheet and do some processing after (mainly price data).

But anyway. It doesn't even matter. Price goes down. There your money went down the drain. Gambling!

Well, that's another element of the equation there.

The result could be given in STU, BTC valuation and maybe USD valuation too.

So a bitbot abused post could have results like this on a downtrend week:

Loss in STU: 27% Loss in BTC: %31 Loss in USD: 35%

well, I am totally agreed by your saying . Honestly speaking, these bot service can fetch few sbd but not the quality of an article. I have seen people depressing out and leaving the platform because they think the most irresponsible post is earning more than their original content and it is turning point to them. Though, the kind of bot you have structured in your head, is certainly a wonderful idea. At least, this kind of bot will spread the real truth behind some of those highly paid out articles.

I think so, because it would not be how much the user spent in bots alone, like the transparency bot, but how much the user lost.

During the waiting week with the verification of my account, I dedicated my hours to "educate myself" about Steemit through videos on YouTube. Those videos were made in good times (november/december 2017) and you could see how you "invested $ 20 and earned $ 60". My first steps in the blockchain were loaded with a lot of misinformation. Unfortunately the videos are still there, hurting and confusing miles of users and future users that go in with a "make money fast and easy" mentality.

Your idea seems wonderful to me, I would like to give more opinions about it but the language barrier limits me a bit and i dont like use Google Traductor at all.

I will be able to ask for permission @meno, so I can translate it into this publication in Spanish and publish it in the @castellano witness blog. I'm going to talk to @nnnarvaez to see if he is encouraged to make your idea reality but in a Hispanic version. There are many who make it phenomenal in Spanish, and unfortunately there are many who believe that buying votes is the only way to advance in the platform.

By all means miguel, please do... I'm a firm believer in the power of sharing ideas. I put this out there to see if I can get a developer onboard, that's the goal!

:)

I like this idea. Even if the OP/original bidbot user ignores it, other readers of the post will see and learn that you're pretty likely to take a financial loss on a bitbot transaction.

Yeah, I think that is ultimately the best outcome. And here is the question.

Would you be offended, if a post you made 8 days ago had a comment by a bot, giving you the breakdown how you lost money?

Or would you be grateful you learned something new?

I've known several people who were happy to have transparencybot doing the math for them.

It didn't bother me, but then again, some people who I find quite reasonable hated it... so I don't know anymore.

The interesting aspect of this to me is the bidbot owners coming in and starting these conversations in MSP. It feels like advertising in discord, which makes me wonder... are people becoming educated on this? Are their profits going down? I just recently got an upvote from some group thanking me for not using bidbots. I don't know where this data on the profitability of bidbots would be available but I think it would be fascinating to analyze.

Believe it or not a dev actually replied to me, so maybe this will become a reality... i guess we will see... :)

Seems like a good idea to me. Any sane person would second guess abusing the bots again if it was clear they had a negative return on their investment.

I'm assuming most who abuse the bots continually understand this is the case though, but they are hoping to gain enough followers in the process to break even or profit in the long run with future upvotes.

I use to think it was the case too... but maybe out of the 6 people I've spoken to in the last month, 1 suspected he was losing money, the other five had no clue.

Coin Marketplace

STEEM 0.36
TRX 0.12
JST 0.039
BTC 69735.97
ETH 3533.64
USDT 1.00
SBD 4.72