Bots Help Moderate Stack Overflow
stackoverflow.blog
stackoverflow.blog
They added all these automated ways to schedule service for your car through the app or website, and after they got comfortable with it, they made it the ONLY way to schedule service. [good for tesla, not good for community]
The problem I see is that Stack Overflow tries to be an exclusive community.
I think maybe instead of automated ways to flag bad things, maybe they can figure out automated ways to praise people, be inclusive especially with new users and redirect mistakes into constructive avenues.
The site relies quite a lot on knowledgeable people to review and edit things, and the more grunt work that entails, the more you have to really care about being a part of the site to commit to it. The types who are interested in shouldering that burden will tend to be the ones who are extra hardcore, who are also more likely going to be impatient with mistakes and newbies.
"Natty started to flag answers automatically once it had detected a particular answer as a non-answer with considerable accuracy."
It might be that I don't understand - does Natty flag something directly, or flag them for the moderator?
On the other hand, the Heat Detector bot sounded it was punitive too, but in a way that might strengthen the community. (interesting thought - could it flag moderators?)
Why shouldn't it? Moderators are just like normal users but with some extra privilege regarding editing/deletion.
There seems to be a bit of confusion on how SO moderation works every time SO is discussed here. Most of the moderation is done by other users. I'm very active in editing/flagging/closevoting, and I can assure you the instances I see an actual moderator (the ones with a diamond next to their name) taking matters into their hands in a given month can be counted in single digits.
I spent an hour looking through them today and I've only seen one false positive and of course someone reported it instantly.
It's pretty useful. One of the questions I saw today was from few years ago and no one had answered it. Someone added a "me too" answer that wasn't helpful. I flagged that answer and provided a new answer to the question, and now it may help someone in the future.
You haven't seen arbitrary, shitty moderation unless you were on the internet in the early 2000s.
The craziest thing is, those people are almost all still on the internet, but they've aged out of the worst of the terribleness.
The only major support site I'm aware of with a 'stupid questions' punishment policy. Which isn't a bad thing in and of itself except its implemented in such a draconian and nontransparent way. Demerits never seem to expire, even after years of no 'stupid' posts and theres no explanation where they come from.
The bizarre thing is even with this and newbies banned left and right it seems the admin is on another planet running some other website as all they want to talk about is social justice and how evil heteronormative SO users somehow can psychically tell who is a woman or minority and supposedly harassing them to epidemic proportions but nobody somehow ever sees this.
Maybe get a handle on a few real basic problems before you start crowing about your geewhiz new bot.
You don't get banned for questions closed as duplicate, what are you talking about?
And it's not about questions getting closed, it's about downvotes.
And from that to saying that the procedure is opaque, I'd say you'll have to provide some sources about that.
He posts questions that are duplicates but are only obvious duplicates if you're have a PhD in quantum statistics. And he attracts downvotes and gets demerits. Most SO users don't have PhDs, therefore most of the the downvotes and demerits must be due to something mystifying.
The procedure is not opaque because:
Well, maybe his questions may be regarded differently by other people, and then it makes sense.
There's a rule on meta.stackoverflow.com that each such complaint must point to an example question. A good rule.