Did you do all of this analysis?! Holy smokes dude. Also, how did you get all of this data? Everything on here is that easily searchable? Also is anomaly a person or a bot... I have gotten a few votes from that account but they were a super low percentage and worth 0, so I looked at the steemnow or whatever its called and the voting power was at like 20% but it looked like a persons page, so now Im confused haha.
Hi @kilbride, I queried the data from the SteemData.com MongoDB instance, that's a database containing all blockchain operations. The database queriy for cases like this is typically much faster than accessing the blockchain directly. So yes, it is searchable, if it's "easy" probably depends on your database/query skills.
The anomaly votes are certainly automated, the posts probably not. You're right, that account is utterly draining it's VP at the moment...
My database/query skills are pretty weak atm, so I doubt it would be easy to me to find;) but its fascinating, and important to take the time to really analyze what the stats mean, which you did a great, in depth-job of. Good to see all of the self-correction, people realizing quickly that they will kill the project by being myopic and self-focused. Maybe there is hope for the human race after all haha
I understand that upvote bots are meant as a marketing tool so that minnows or authors with low steem power can get noticed. As your article indicated that the 3.5 days as well as the blacklist is to really ensure that the bots are put to the right use rather than being abused. Thanks for sharing such detail info. Little upvote from me!
If I had any complaint it would be in the repeated phrasing after a given graph; after all, we know the graph is here because we can see it. But that's a minor stylistic issue compared to the actual analysis that you've done, and that's fabulous.
I suppose it was obvious that most of the bot transactions would migrate to the least strict bot available at any time. There's probably a way to determine how much additional income @randowhale has taken on board as a result of the others becoming more strict, probably by a fairly straightforward check of transfer operations on the blockchain.
Thank you @lextenebris, also for the stylistic advises ;)
It's an interesting idea to check how the income changed/moved across the various bots! I was looking at a rather narrow aspect in this analysis and hit a niche with more or less static contribution rates over the given time period that made the move to randowhale nicely visible. I'm curious to see if this is visible as well on a broader scale!
Believe me, after seeing any of my code – you'll appreciate my writing advice a lot more than anything I could say about how to implement things. I certainly wouldn't want to be maintaining the code I write these days. (Crap, I have been maintaining it. I need to find a saner hobby.)
I did a bit of analysis earlier this week/last week about the transactions which transfer funds between the whales at the top of the blockchain. That was fairly interesting, in its way. My intuition suggests that "following the money," as it were, is always going to be a revelatory process.
Following the SP pool in a given bot across time along with watching the account for resource transactions could lead to some interesting insights about which ones are actually seeing better traction – and from what direction. The things that you've already done with looking at where bots are actually putting their votes is a good chunk toward that solution.
HI @paulag, good point! The reason zappl doesn't appear here is that zappl uses HTML instead of markdown for the posts. So even if there's only a single URL in a zappl contribution, its content actually starts like
The filters I used to find the URL-only posts only match plain and markdown links. Full HTML parsing to see if it is only an image/video/link would probably have required to pull all html posts and parse them one after the other.
It's great that they all started to implement at least some restrictions for cheaters. And we see that this really works. But why each of them acts alone?
It seems to me that the introduction of one common Blacklist for all voting bots (including free ones like Busy, eSteem, Dsound and others) would greatly facilitate the task for owners of such bots. They could synchronize information and save a lot of time each other.
This list can also be linked in some way with the Steemcleaners' and Cheetah's blacklist to avoid crossvoting (like Cheetah against bid-bots). How do you think?
From my understanding there is already some exchange of blacklists between steemcleaners and bot owners. I think this is a controversial topic, not all bots apply the same rules, and I assume there's a lot of "grey space" of user behavior that may be OK for some and considered abuse by others. Being on one list would then also mean to be on every list. This would need very clear rules what is accepted by the community, how the lists are synchronized and how to get off the list again.
And let's not forget the most obvious fact about any evolving population pool:
In the absence of a controlling authority or the ability to establish one, allowing your opponents to limit themselves while you choose otherwise is the best way to profit. In the presence of a controlling authority or the ability to establish one, becoming part of it and strangling your opponents while keeping yourself free of limitation is the best way to profit.
I'm pretty sure that's exactly what we can observe in your graphs.
That's a good point! Even if all "big players" agree on a common sense, there will always be some bots offering their services outside this agreement. They'll make good profit at least for a while.
Whether or not they will continue to make good profit comes down to whether or not they can be coerced into doing otherwise.
How quickly can a new bot be spun up, publicized, and start taking requests for votes? (That's not a rhetorical question; I honestly don't know.) If it becomes profitable to make that cycle shorter to get inside the discovery period of whoever has taken enforcing whatever the policy of the day is, that's what will happen.
I tend to come at these things from the perspective of a game designer and evolutionary systems guy. I don't labor under the illusion that voting bots are anything but minigame gambling machines where you put your money and you take your chances, which sometimes makes me feel a little odd on this platform.
It will be interesting to see how the distribution of people who have become consistently familiar with being able to buy a portion of the reward pool react to the new bot changes. There was already a distinct pressure by the platform mechanics themselves to short form content over anything else and ephemeral content over evergreen content, and I don't see the pressures being introduced by these bot changes doing anything to reduce those mechanical pressures. Quite the opposite.
As someone who is more interested in longform and long-term content value, that's a little disappointing.
Hello, as a member of @steemdunk you have received a free courtesy boost! Steemdunk is an automated curation platform that is easy to use and built for the community. Join us at https://steemdunk.xyz
Upvote this comment to support the bot and increase your future rewards!
Did you do all of this analysis?! Holy smokes dude. Also, how did you get all of this data? Everything on here is that easily searchable? Also is anomaly a person or a bot... I have gotten a few votes from that account but they were a super low percentage and worth 0, so I looked at the steemnow or whatever its called and the voting power was at like 20% but it looked like a persons page, so now Im confused haha.
Hi @kilbride, I queried the data from the SteemData.com MongoDB instance, that's a database containing all blockchain operations. The database queriy for cases like this is typically much faster than accessing the blockchain directly. So yes, it is searchable, if it's "easy" probably depends on your database/query skills.
The anomaly votes are certainly automated, the posts probably not. You're right, that account is utterly draining it's VP at the moment...
My database/query skills are pretty weak atm, so I doubt it would be easy to me to find;) but its fascinating, and important to take the time to really analyze what the stats mean, which you did a great, in depth-job of. Good to see all of the self-correction, people realizing quickly that they will kill the project by being myopic and self-focused. Maybe there is hope for the human race after all haha
I understand that upvote bots are meant as a marketing tool so that minnows or authors with low steem power can get noticed. As your article indicated that the 3.5 days as well as the blacklist is to really ensure that the bots are put to the right use rather than being abused. Thanks for sharing such detail info. Little upvote from me!
Thanks @fun2learn! Yes, a main limitation however is that the definition of "right use" is not so easy and especially not the same for everyone...
Hey @crokkon I am @utopian-io. I have just upvoted you!
Achievements
Community-Driven Witness!
I am the first and only Steem Community-Driven Witness. Participate on Discord. Lets GROW TOGETHER!
Up-vote this comment to grow my power and help Open Source contributions like this one. Want to chat? Join me on Discord https://discord.gg/Pc8HG9x
Nicely done, and a fine tuned analysis.
If I had any complaint it would be in the repeated phrasing after a given graph; after all, we know the graph is here because we can see it. But that's a minor stylistic issue compared to the actual analysis that you've done, and that's fabulous.
I suppose it was obvious that most of the bot transactions would migrate to the least strict bot available at any time. There's probably a way to determine how much additional income @randowhale has taken on board as a result of the others becoming more strict, probably by a fairly straightforward check of transfer operations on the blockchain.
Good job!
Thank you @lextenebris, also for the stylistic advises ;)
It's an interesting idea to check how the income changed/moved across the various bots! I was looking at a rather narrow aspect in this analysis and hit a niche with more or less static contribution rates over the given time period that made the move to randowhale nicely visible. I'm curious to see if this is visible as well on a broader scale!
Believe me, after seeing any of my code – you'll appreciate my writing advice a lot more than anything I could say about how to implement things. I certainly wouldn't want to be maintaining the code I write these days. (Crap, I have been maintaining it. I need to find a saner hobby.)
I did a bit of analysis earlier this week/last week about the transactions which transfer funds between the whales at the top of the blockchain. That was fairly interesting, in its way. My intuition suggests that "following the money," as it were, is always going to be a revelatory process.
Following the SP pool in a given bot across time along with watching the account for resource transactions could lead to some interesting insights about which ones are actually seeing better traction – and from what direction. The things that you've already done with looking at where bots are actually putting their votes is a good chunk toward that solution.
this is very interesting and very nicely done. Surprised to see that zappl is not on the list of apps
HI @paulag, good point! The reason zappl doesn't appear here is that zappl uses HTML instead of markdown for the posts. So even if there's only a single URL in a zappl contribution, its content actually starts like
The filters I used to find the URL-only posts only match plain and markdown links. Full HTML parsing to see if it is only an image/video/link would probably have required to pull all html posts and parse them one after the other.
Great work again, crokkon. Your contribution has been approved.
You can contact us on Discord.
[utopian-moderator]
It's great that they all started to implement at least some restrictions for cheaters. And we see that this really works. But why each of them acts alone?
It seems to me that the introduction of one common Blacklist for all voting bots (including free ones like Busy, eSteem, Dsound and others) would greatly facilitate the task for owners of such bots. They could synchronize information and save a lot of time each other.
This list can also be linked in some way with the Steemcleaners' and Cheetah's blacklist to avoid crossvoting (like Cheetah against bid-bots). How do you think?
From my understanding there is already some exchange of blacklists between steemcleaners and bot owners. I think this is a controversial topic, not all bots apply the same rules, and I assume there's a lot of "grey space" of user behavior that may be OK for some and considered abuse by others. Being on one list would then also mean to be on every list. This would need very clear rules what is accepted by the community, how the lists are synchronized and how to get off the list again.
And let's not forget the most obvious fact about any evolving population pool:
In the absence of a controlling authority or the ability to establish one, allowing your opponents to limit themselves while you choose otherwise is the best way to profit. In the presence of a controlling authority or the ability to establish one, becoming part of it and strangling your opponents while keeping yourself free of limitation is the best way to profit.
I'm pretty sure that's exactly what we can observe in your graphs.
That's a good point! Even if all "big players" agree on a common sense, there will always be some bots offering their services outside this agreement. They'll make good profit at least for a while.
Whether or not they will continue to make good profit comes down to whether or not they can be coerced into doing otherwise.
How quickly can a new bot be spun up, publicized, and start taking requests for votes? (That's not a rhetorical question; I honestly don't know.) If it becomes profitable to make that cycle shorter to get inside the discovery period of whoever has taken enforcing whatever the policy of the day is, that's what will happen.
I tend to come at these things from the perspective of a game designer and evolutionary systems guy. I don't labor under the illusion that voting bots are anything but minigame gambling machines where you put your money and you take your chances, which sometimes makes me feel a little odd on this platform.
It will be interesting to see how the distribution of people who have become consistently familiar with being able to buy a portion of the reward pool react to the new bot changes. There was already a distinct pressure by the platform mechanics themselves to short form content over anything else and ephemeral content over evergreen content, and I don't see the pressures being introduced by these bot changes doing anything to reduce those mechanical pressures. Quite the opposite.
As someone who is more interested in longform and long-term content value, that's a little disappointing.
Hello, as a member of @steemdunk you have received a free courtesy boost! Steemdunk is an automated curation platform that is easy to use and built for the community. Join us at https://steemdunk.xyz
Upvote this comment to support the bot and increase your future rewards!
This is very interesting and nice post. this information is very important for me. Thanks for create a nice post.