I don’t see how engaging in malicious compliance is being a useful idiot. Implementing the entire surveillance mechanism free of charge, that I would call being a useful idiot.
Purposefully implementing a broken feature to satisfy the letter of the law, while preserving the user’s ability to avoid the surveillance mechanism is certainly not that.
Block-chain!
The enemy is both weak and strong.
No point in contributing to a hostile organization.
Happened with Tears of the Kingdom too iirc.
It’s just gonna pollute the internet with even more bullshit. Language models don’t really understand topics, they just put together words that are likely to appear with each other. Biases are inherent to this design.
"I just want to browse for god’s sake 
"
If you don’t want to be informed, fine. Nobody’s forcing you to use a different browser either.
Not sure how it is nowadays, but back in 2018 Libreoffice Calc was struggling to handle even a single sheet of data entries, performance-wise, let alone multiple sheets.
I’m not expecting it to have every feature imaginable, but I do expect it to not freeze when processing even a relatively small dataset.
If there were a massive bot problem on Lemmy, you’d be seeing upvotes in the thousands, not the low hundreds.
Only thing you get out of this compared to the alternative of malicious compliance is opening yourself up to attack. You can still fight this without painting a big target on your back.