This experiment has been running for almost 2 weeks now and I think it's time to analyze the outcome. If you don't know, what I'm referring to ... :

https://steemit.com/goodbot/@goodbot/goodbot-s-first-job

@goodbot left more than 700 comments in that time.

The reason, why I thought policing that tag would be a good thing to do, was mainly to keep it free for the actual newbies, who are encouraged by the welcome page to make a post there.
There isn't a real rule about this and my bot doesn't try to enforce any rules, but tries to give a friendly hint.
While I had some encouraging comments and votes for goodbot's main post, I also got a bunch of not so happy responses ...

https://steemit.com/@goodbot/recent-replies

First, I tried upvoting the positive comments and whenever someone reacted and removed the tag, I gave them some cents for their fixed post.
However, I also had some people going really mad at me and I started copying the steemcleaner's standard reply.

In the end I felt like Eric Cartman, when he was deputized ...

I don't want to be the next Steemit Police.

I still think tag abuse sucks, but I will not comment on responses to @goodbot.
I will just let it run, until I hear a good argument against it.

What bugs me about this is how people who practice tag discipline will lose out compared to tag abusers.

I just hope people will go back to manual curation. With 10 votes/day it shouldn't really be too hard to do some actual reading instead of showering an entire category with (easily abused) votes. If that was to happen more, @goodbot's comments might actually have more of an effect.