I see so many people making a huge deal out of linux stuff adding support for the california age thing, and I'm like. you know basically every online service has been required to ask for your age since 1998? this is literally just "at account creation, the device owner can set an age field. to whatever they want. and then apps can query that instead of asking themselves."
you can set it to the unix epoch if you want
@rcombs yeah i really don't get it
@rcombs everyone knows young people will lie about their age. this enables them to lie about their age very easily. why the fuck would you put up a fuss about it
@whitequark @rcombs I think there are two aspects here. One is the matter of unknown and untrusted folks inserting themselves into projects eagar to comply in advance. This expresses a disempowerment of communities and a surrender of editorial control to a fascist government. From this perspective, it's the statement of values that matters more than any practical effect.
@dalias @rcombs how is it "in advance" if there are laws about it today?
@whitequark @rcombs Is that a serious question? There is no law that says that we as the authors of a program have to modify it to behave in a certain way. If there is, it's unconstitutional. Even if it weren't, we wouldn't comply. It's complying in advance if nobody is holding a gun to your head or putting handcuffs on your wrists. Governments do not decide what we can write.
@dalias @whitequark @rcombs Brazil's law requires OSes to implement age APIs
https://www.planalto.gov.br/ccivil_03/_ato2023-2026/2025/lei/L15211.htm (^F "Application Programming Interface")
@natanbc @dalias @rcombs if this is not done, which party is the liable one?
@whitequark @natanbc @rcombs This isn't an answer in terms of case law there, but in general, we have long held that writing FOSS is expression, and that while building products (physical things, preinstalled systems, maybe automated installations? etc.) out of it might be subject to laws and regulations, nobody can tell us what we can or cannot write upstream. This is a principle we should continue to fight for. If we lose it we will regret that.
@dalias @whitequark @natanbc okay? but clearly a lot of downstream consumers are going to need the facility, so providing a standard (optional!) interface for it is prudent as an upstream infrastructure maintainer, to avoid obvious painful fragmentation?
@rcombs @dalias @whitequark @natanbc why do I, as a user of Linux, want to make that easy? That's actually the opposite of what I think is the best idea.