In the web interface on our instance all posts are flagged nsfw by default, but you can uncheck it on a per post basis. I think that various apps have different settings that don’t default that way, nor will posts coming from other instances.
This is inaccurate. It did not say “childlike” at the time that Ada complained. After we got defederated, I asked a new mod of that community to update the sidebar because it was very light on rules, purpose, etc. and I thought that maybe our conflict with blahaj could have been avoided if the sidebar was more explicit about what the community was about. As part of his revision, on his first pass he copied and pasted a dictionary definition of “adorable” which included the word “childlike”, then went back and re-edited it to remove that word on a later editing pass. I want to say it was in there for about two hours? During that time, a couple of people spotted it and made some unwarranted assumptions.
Here’s a post I got one of these reports on. Can someone from .world check? NSFW, obvi.
It certainly could be a client thing, but I’m betting it’s a federation problem. Some kind of bug that’s hard to trace.
That’s upsettingly plausible. Ugh.
There is definitely a feature request already in github for tagging. I’m sure they would love the help
Those are both my leading theories, though again that doesn’t tell me why I’m getting these reports almost exclusively from lemmy.world