@Background Pony #1302
Not necessarily. This site is not that site, and some of the implications and tags on that site are from more than a decade ago. And for those very oldest of all implications that came from when implications first came into existence in BooruOnRails, those implications have had time to evolve and expose a great number of “whoops, that isn’t actually useful” or even “Holy shit that was a mistake how do we fix it?” along the way.
For example, the short definition of
faceless male is “a male character is partially present (physically)”, and to me that sounds more like how
offscreen male is used - even though that really has become a term that means NONE of the male is onscreen, but that’s what we have
implied [foo] tags for.
So … We aren’t Derpi. We inherited their tags because it would have been dumb to abandon the enormous amount of tagging that had already happened on the images we imported, and along with those imports came the whole spider web of ‘how all these tags fit together’. But, over time, we keep finding dumb tags, or dumb implementations, or flawed alaises. Many things that made sense when they were created but over time have ended up being dumb.
Maybe, just maybe, The Old Ones who created the infinite potential and chaos that was the soup from whence Derpi was birthed made mistakes that Derpi could never rectify because The Old Ones still haunted their database to such an extend that their mistakes and “Try this let’s see if it works” ended up becoming crystalized like ancient diamonds.
2LDR: Maybe we can tag intentionally and over time our tags will end up better. Sometimes tagging changes on Derpi make sense and can benefit us. But sometimes they look wrong or cause problems that we don’t want to deal with, or have dealt with in different ways. This specific set of tags? Looks wrong to me.
I would not be comfortable making those implications without a good long chatting with the tagging wonks and an even longer “let’s sleep on it”.
After all, on this site something with 8K tags is 10% of the database, and that’s a great bit old chunk of images to retroactively apply or remove tags from when an implication changes.
