I've been thinking about this idea for a couple of weeks now. I'm not entirely sure what would be the best way to implement it, but I hope a developer steals the idea, because I believe it could do wonders to curve abuse on this platform. Granted not all bot owners are developers and even among those who are, not all will care, but It's my sincere hope that they see the value in what I'm trying to propose.
Identifying Patient 0
There are plenty of different ways Abusers do their pillaging of Steem. However a big element of the abuse is the automatic creation of sock puppet accounts. If the accounts are made to follow trails to upvote spam, or if they are made to simply spam themselves to be voted by a main accounts are not as important as identifying where it all got started. At least in the sense that this way we can try to identify the owner and abuser.
One form of abuse that has spawned recently is detailed on this post by @therealwolf and it made me think that right now is about as good a time as any to bring this up. Hopefully a developer will take this idea and run with it, or a whale will choose to fund it, either way I think it would be incredibly valuable.
In theory
The software would be given a starting point, a username. The software in turn will scan that user's interaction with the blockchain and look for transfers above 0.001 to ignore memo spams, delegations from and to, and account creations. It would generate a report with all that information.
This report could show not only the name of the interacting accounts but also the frequency of interactions as well, thus allowing a would be investigator to get closer to patient 0 (main account). When the list of abuse accounts is generated this in turn could be downloaded to be provided to bot owners for blacklist updating.
In no particular order, this is some of the information that the report could generate that should be extremely helpful to the investigations:
- Accounts of Note - Rep above 50, High SP Holdings, Age of Account, Etc
- Number of Comments per Account
- Number of Posts per Account
- Number of Accounts Created from account
- Transfers to Exchanges
- Memo searches for Exchange Transfers
- If anonsteem was used - Report the 10 accounts created by anonsteem after the Steem transfer (this would only work if Steem was used of course)
- Cross referencing Systems (Many combinations)
What's the Goal?
Honestly to deter people from even trying. In other words, if we have a SteemDetective service (please steal that name). Then bad actors would think twice before doing anything. The effort it would take to abuse the system might be so much, they might just think it better to go elsewhere.
Conclusion
I'm not entirely sure how feasible this project would be or if they would be enough interest to tackle it, but I hope that someone steals the idea and runs with it. I know that it's impossible to eliminate all abuse, and this would not aim to do so. It just aims to help those who are doing the best to keep our ecosystem clean a little more.
Everytime I read one of @themarkymark's hunting ventures, it usually reads. "I spent 10 hours investigating a spam farm" or something of the sort. Who has that much time to spend doing that? But this could be a game changer for him, for @steemcleaners and others.
Hope everyone is having a beautiful day
Other posts by yours truly
• Another way to look at it - HODLing assistance series
• The Advantage of Steem - HODLing Assistance Series
• Distorted Wisdom - A sure approach to burning out
• helpienaut Meeting 7/16/2018
• Could this be the bull market? If it is, I'll do things a little bit different this time...
Hi Meno. That would be a great tool to have for the Steem Detective. There are lots of accounts operating it seems for other accounts ie the same person.
I'm crossing my fingers someone takes me up on it. I'm willing to brainstorm the idea with someone.
I like the idea, but this will prob. require an intelligent system.
The biggest problem with finding the patient 0 was that most main accounts were created via anonstem through an exchange.
Now in my case - dreamryder007 made the mistake of transfering the money directly to the main-creator, which showed who patient 0 was.
But in any case - I do believe that people make mistakes (owners upvote the posts, writing style etc.) and having an intelligent system which tracks all of these data, might be able to find out who patient 0 truly is.
A developer that directly comes to my mind would be: @andybets.
Haha, yes but that's true. It takes hours to research and deal with that stuff. And in the end, those abusers even try to act as if they didn't do anything wrong.
I'm willing to have a convo with anyone that wants to tackle this and brainstorm. Obviously some of the inner workings should not be disclosed as to not warn the abusers.
But, even on the exchange part, I have an idea to narrow it down. Meaning, maybe have 20, 30 suspicious accounts to look into.
:)
.
Maybe the tool could be broken down to parts... meaning, specific functions.
Something like that. Just make it easier. It will still require manual operation sort of speak. But if it takes 10 hours we are less likely to do it.
.
you bring some important points. I guess, its just about making it a bit easier to do the research. Maybe, in simplified terms, that's the goal.
oh, hope this post will get a couple of thousands of upvotes and someone actually will develop the tool. I am new to Steemit but I find it quite terrifying to look for the real people to follow and share my opinions with.
Especially all those trending posts - most of the time crowded with some sort of bots. Sad in a way. Fingers crossed for the brilliant devs of the community to put this idea into action.
@meno very nice article about abuse things
@meno it's a fantastic platform where we can trange anyone's idea.