• 4 Posts
  • 1.49K Comments
Joined 3 years ago
cake
Cake day: June 15th, 2023

help-circle








  • I think when people talk about fiction being too “political” nowadays, they usually mean one of three things:

    1. “I hate minorities.”
    2. “I am only now savvy enough to notice the political undertones that have been a defining characteristic of the genre for a century or more.”
    3. “The writing sucks.”

    Personally, I can get behind #3 because boy howdy there’s some bad writing out there.

    And I’ll admit I occasionally fall prey to #2. Some things I watch today feel too “on the nose” in regard to current events, and sometimes it’s hard to tell if it’s truly worse writing than my old favorites or if I was just too ignorant, naïve, or credulous to pick up on it when I was younger. It’s also sometimes hard to tell when re-watching old stuff because I don’t feel the zeitgeist of 30 years ago in my bones the same way I do today’s (naturally).

    For example, I can watch an episode of Futurama that’s literally about the world being nearly destroyed by a giant ball of 20th-century garbage, and somehow it doesn’t feel overtly topical, while the new episodes about bitcoin and AI feel more like a sermon than a sitcom. Maybe the writing is worse, or maybe I’m just old now. Who’s to say?



  • I don’t want AI in my browser even if I can turn it off for the same reason I don’t want my refrigerator door booby-trapped with an explosive even if I can turn it off.

    Bugs happen. Configuration changes happen. User error happens. Software is complex, and I shouldn’t need an intimate knowledge of every goddamn app I run to be sure it’s not siphoning all my data off to god-knows-where. I use hundreds of programs on a daily basis. It is completely untenable to carefully configure every single one, stay abreast of constant updates and changes, and spend 76 full working days reading every TOS I am subject to. And of course, all their policies and defaults are subject to change without notice, so nothing I learn today will necessarily apply tomorrow anyway.

    I want to be confident that my web browser is not — either by design, due to a misconfiguration, or due to a bug — sending my data to OpenAI. I do not want a booby-trapped browser, even if I can turn off the booby-traps. I do not want my fridge to explode, so I don’t buy fridges with built-in explosives. Seems pretty simple to me.

    I also want to be confident in the same for others. If I deploy a browser to 100 employees’ machines, or even just my mom’s, a little opt-out checkbox under Settings will not give me any peace of mind.




  • Share pictures of yourself, or your children, only with actual friends and not for the whole world to find

    Good advice but let’s be real: in practice, this means having no social media profile, and even that is a half-measure.

    Even if I carefully curate my friends list (most people don’t), and share my photos with only my inner circle (most people won’t), I have no control over what my friends do. If my cousin posts a photo he took at Thanksgiving, it’s probably going to be visible to all his friends, and even friends-of-friends. That’s thousands of people I’ve never met and there’s not much I can do about it.

    There are pictures of me on Facebook, and I do not use Facebook. The social cost of getting on everyone’s ass about taking/posting pictures with me is too high even for a grumpy old fart like me. At least I’m not tagged (since I don’t have a profile), so it’s not neatly pre-sorted for potential attackers. But that’s at best security through obscurity, and it isn’t even very obscure. Anyone targeting me specifically would have no trouble finding pictures of me, and none of that is realistically within my control.

    It’s more like “beater bike security”. Any bike lock can be thwarted by a dedicated thief, so the best strategy is simply to be a less attractive target than the other bikes around.

    This is a systemic problem. It goes beyond individual choices and even beyond social media policies.