Re: to @dantheman "notice-to-bot-spammers"

in #steemit8 years ago

I have suggested this in the witness channel, and now I am bringing it up on Steemit to see if this has any potential at solving the spam bot issues.

First I would like to note that fighting bots with bots is potentially an unending war and will cause more spam TX on the blockchain than it would without. Hiding the spam bot posts, is merely sweeping your problem under the rug.

If we add a command that allows the active witnesses to flag a certain account for spam and cause a suspension of the account posting authority for 30 days by making that account's bandwidth zero for that time period.

We would require at least 15 of the 19 active witnesses to execute the flag command for the suspension to occur. This will prevent accidental suspensions as more witnesses will have to investigate these accounts before issuing the flag.

This will be a blockchain based solution, and I am not sure if it can be done. But there is very little chance of abuse. On the other hand, this can be done with a reputation system as suggested, but is this reputation system blockchain based? or is it the equivalent of appointing mods on Steemit?

I would also like to know what is Steemit's policy towards bots in general, not only spam bots. Are all bots considered villains?

There are spam bots, anti spam bots, plagiarism detection bots, curation bots, comment bots, and soon stats bots and many more. What is the determining factor, with spam bots it's easy to distinguish, but what about the rest?

Joseph

Sort:  

I'm afraid that this will not work. It is very easy to create bots and have them spam the chain with posts. OTOH it is a lot of work to determine if an account is a spam bot. Witnesses couldn't keep up with that.

@joseph
I can not comment much about spam bots, I think this is the case not only at the site steemit.com, but I'm sure if the team steemit also acting in it and willing to listen to ideas from the community

I am sure the community will help with this, if that ever becomes an option. Witnesses would not have to look for these spam accounts, a list would be compiled and investigated. It would take less than a minute to ban 50 spam accounts in the cli wallet.

I don't think it should matter if they're a bot or not. If they're behaving like one, seeya.
Just because it might turn out to be an actual person, why should their bot-like behavior be any more tolerated?

I really do wish it was possible tho cause i absoutely hate them defintely kill joys

To all my homies that spam.... here is a drive-by for you!

HEY @calva look here what a beautiful gift from MR. BEN.
To share with your spam friends! GOGOGO

Nice picture

Hey @joseph , thank you for your suggestion. Even if it's not something that can be done, or if would even be effective (you guys would know way better than I would), the fact that we've got people here racking their brains to find solutions to these kinds of problem is awesome.
THIS is true community activity right here. Thanks for all your efforts.

This might be a solution. I would lower the number of flags to 10 or eleven.

But ... new bots can appear every second. They can be created / duplicated much quicker that discovered and neutralized.

A very unpopular solution comes to mind - captcha :(

You make some great points, I was for the rep system. Now I need to rethink it. Your flag command suggestion definitely a good alternative. ​The bot issues is going to get worse. Great points, Thanks Upvoted

Great idea! I like very few bots. @wang is great and @weenis can be helpful as well. @wang only comments to welcome new users and give them a few tips and answers. However if all a bot is doing is repeating the same thing over and over with no benefit to the user, it is spam. Plagiarism should have a warning around it for potential plagiarism that can be removed with people voting on it. Curation bots should be limited to being used by whales and dolphins who have a vested interest in good content and allow instant upvotes of certain users if they wish. comment bots other than @wang are worthless in my opinion and should be removed. Comments here are meant to be meaningful and provide feedback or ask questions. Simply saying "great post" is not valuable to anyone in the community. It does not show you read the article or care about anything besides trying to get a few pennies. That's my two cents. Bots should be able to be voted on and made invisible if the community feels that way.

Wang seems to be the only bot who makes really bank here. Would be interesting to know, who actually made it.

In fact, many boats are earning more money than regular users , who wrote the article! I think if a person was caught in the spam - ​​it is necessary to block an account can not be restored

Good proposal! It makes us look bad when every comment in a new users post is a bot. Why would they continue to use the site if they think it's only bots!?

Thanks for your effort joseph. I'm not sure either who should look into that, but considering the witnesses might not have the time to do then there are not many options left. And then again it comes back to trust and is a lot of work, compared to the time someone needs, to set up a new bot.

It seems most of them are basically trolling or just don't understand anything about what they are doing. They don't even getting any rewards from what I've seen, but they destroy the visibility of almost every post and showing false info with their nonsense votes on every thread.

Yeah hard to find a border, from what point is a bot disturbing or not. Surely depends on the personal point of view. But anyways, there's no way to get rid of all bots, so there has to be found a solution of some kind.

I think it's a great idea. I am excited to see how Steemit continues to grow and solve issues.

Suspension! I'm on board with that!

This is a issue that needs to stop , thank you for your post

I like this solution. No one individual can pull the switch. Also thought of flag (downvote) power decreasing by larger margin (than an upvote) and going to full over longer period. Think I like this better

The 'helpful'info bots are just as bad as the 'great post' bots ,they throw up a wall of text in most new submissions. It's discouraging users from commenting on content they like.

this post shows how fast something can go viral. The topic is more up to date and it will be a challenge for the developer of steemit. Will see and time will show. Good luck.

Just wait for few days these bots won't be alive here anymore

Maybe use the incentives for finding great post for identifier bots. If you flag something as a bot and many people also flag it then reward them. These will only be paid after 30 days. This will help give time to discover other bots who may be using the flag to get rewards.

As I understood from article two variants considered:
1 War of bots with bots
2 War of human curators with bots
I've never flagged anybody, but maybe it's better to let users do it?
Probably the best one will be mix of all above

The idea with downvoting autoposting bots now is to discourage them now and into the future. If autoposting bots only get hidden when they post, hopefully the owners will stop trying.

I think the efforts so far have prevented even worse comment spam, right now.

Thank you for the post and for the excellent ideas, I am a newbie and had no idea on my first post that the first 6 replies were bots, thanks to the community I was informed that this was the case. It would be great to have a quick reference sheet to better understand the steemit process when you first open an account...just my dos centavos