With regard to the Shinigami Eyes plugin issues, I want to stress how important it is to Trust, But Verify.
A tool can be wrong. This one in particular is curated by a team of human people who can and will have biases. Any easy blacklist curated by real actual humans is going to be open to their biases. AI or Bot driven lists* aren't affected by biases, but they also aren't smart enough to tell "Rowling" from "growling" or "terf" from "butterfly". Honestly, the bot/program that identifies words is more reliable in the long-run vs the one that relies on human input. It still needs checked, obviously, but it's largely free from human bias. A little guy that just points to words and goes 'it has these letters in this order!! Be Aware!' but doesn't assign Good or Bad to it.
It's truly unfortunate that Shinigami Eyes has been made unreliable through bias. This sucks! I wish it was not the case! It was a useful tool for a lot of people for a very long time, and the fact it's being weaponized now against my community is incredibly saddening.
Much in the way you cannot trust a callout post blindly, you shouldn't trust any labeling or blocking tool blindly. It's unfortunate that many people will be learning this through this debacle. :(
WHY ARE YOU MARKED RED ON SHINIGAMI, WTF DID U DO
i cant even post moths anymore. because of woke
Reviews and ratings for Shinigami Eyes. Find out what other users think about Shinigami Eyes and add it to your Firefox Browser.
As an addendum, because I did do some cursory lookin' and the link above to firefox reviews does reflect this: Things seem to have taken a downward turn in the last 5 months, though this problem of false flagging was absolutely present before. A tool that relies on user reports is not a reliable tool. Any number of factors can lead to false flagging. Humans and programs are equally fallible (for different reasons) and both can give you a wrong result.
Stay safe, friends. It's truly unfortunate that a tool trusted to protect us from constantly seeing rampant transphobia online has turned against the community it was built to protect.
*systemic biases do influence machine learning, ai, and bots. for the purposes of this, I used generalized words to mean 'an algorithm that blocks things based on a word or string of words without consideration for context', which is how tumblr's blacklist function works.
