• 12 Posts
  • 94 Comments
Joined 1 year ago
cake
Cake day: July 19th, 2023

help-circle


  • I wasn’t going to explain my downvote, but it’s been a few days and apparently everybody here is thinking about MRAs when there’s more at stake.

    I see Nixon in Trump: somebody who starts and prolongs wars for their own political gain. Of my three uncles who qualified to go to Vietnam, one was permanently disabled during basic training, one didn’t come back home, and one fell apart before I was born. I had to “voluntarily” register as a potential servicemember in order to access various standard government services as a young man in the 2000s, while the USA was invading Iraq and Afghanistan. Under a sufficiently fascist government, the USA has shown itself capable of sending its men to death. This system is explicitly misandrist; only men are required to register and only my uncles suffered this hate.

    Misandry isn’t equal and opposite to misogyny. Our society was never obligated to hate men and women in ways that are nicely symmetric and amenable to analysis; indeed, critical theory suggests that society deliberately structures itself to obfuscate its hate.




  • I haven’t done a headcount yet and the election’s not fully tallied, but I think that the Senate still has around 70% support for NATO, and historically we can expect to see a “blue dog” phenomenon in the House as a reaction to Republicans gaining seats. Effectively, both the Democrats and Republicans will function as big tents of two distinct parties, and there is usually tripartisan support (everybody but the far-right Republicans) for imperialism. We may well see votes where the legislators override presidential vetoes to force weapons sales and otherwise fulfill NATO obligations.

    And yes, you read that correctly; Democrats move right as a reaction to Republicans doing well. Go back to bed, America…










  • It’s almost completely ineffective, sorry. It’s certainly not as effective as exfiltrating weights via neighborly means.

    On Glaze and Nightshade, my prior rant hasn’t yet been invalidated and there’s no upcoming mathematics which tilt the scales in favor of anti-training techniques. In general, scrapers for training sets are now augmented with alignment models, which test inputs to see how well the tags line up; your example might be rejected as insufficiently normal-cat-like.

    I think that “force-feeding” is probably not the right metaphor. At scale, more effort goes into cleaning and tagging than into scraping; most of that “forced” input is destined to be discarded or retagged.