The idea is to countervote massive selfvoting and "educate" people to "give back" to the community.
Almost. The idea is to countervote optimal selfvoting and oppose it with self votes and a conspicuous comment for awareness. Optimal means self voting highly very frequently (pretty much has to be automated) and not voting for really anyone else. We don't presume to arrogantly educate people, it's just a statement.
@sadkitten is community funded, so it's the stake of several people who agree with the bot activity and who thing it's good enough to contribute their stake towards the flagging behavior.
The bot is not an AI in a grand sense but just for clarity it does use a completely automated algorithm to identify optimal self voters and the flag them, which works on a weekly basis as the comments imply.
I know you say it for dramatic effect, but there's nothing evil about self voting, it's just selfish by definition, and counter to the idea of curating content. Your use case is a common one among projects: self vote to fund your activities. Many projects do this, even projects we are ostenisbly in line with, like Steem Cleaners. Still, this is not something we agree with.
As you pointed out, on Steem we can agree or disagree in the form of votes in proportion to our stake. @sadkitten is the voice for people who disagree with self directing rewards completely towards the voter.
If you self vote less optimally, by voting on a few other accounts or not self voting as frequently, you will not be regarded as such as high self voter. The list is updated every week. Note however that an upcoming algorithm update will take into account the overall voting pattern so vote cliques (popularly referred to as "circle jerk") will also be considered.
Personally I don't like to see projects getting flagged but then again I don't like to see project accounts voting so selfishly. I'm glad you posted about this, thanks for engaging in the topic.
thank you for the clarification, @personz. what do you think, is there a way to meet each other without dogmatically expecting some behaviour from only one side? i mean, you understand what the problem is. i'm just asking if there can be a way that both sides are satisfactorily with. would you like to support the @steembay project somehow? pls, let's think about something constructive. you try to do something good and @pollux tries the same. it should be possible to reach an agreement...
meanwhile the downvoting goes on and on. and because of the way higher reputation the flagging accounts have, it hurts the reputation of @steembay more and more. how much time does he have to react to this? seriously, this certain mercilessness can't do good here in this special case. negotiate this, please.
The bot is due in a few hours to analyse the previous seven days' activity and update. If @steembay was not self voting very optimally in the last seven days then it will not be on the updated list.
Otherwise I don't see a reason to make an exception. The whole point of using an algorithm is that it is fair. People have accused us (without looking) of favoring whales and so on, but the bot does not play any favorites. If we do that the project loses credibility.
I am open to discussion about it, obviously as here I am. But if @steembay is committed to self voting we are at cross-purposes unfortunately.