uuldika

joined 3 months ago
[–] [email protected] 4 points 4 hours ago

I'm looking forward to fucktons of individual suits absolutely slamming the courts every time an EO is issued. crowdfund the filing fees. turn petitions into copypasta. DDoS the Court system. they literally asked for this.

[–] [email protected] 1 points 2 days ago

friendship ended with Raytheon. now Rhinemetall is my best friend.

[–] [email protected] 3 points 2 days ago (1 children)

Why would he push NATO countries to increase their defense spending target by 150% if he was planning on actually invading?

Trump isn't planning to invade NATO (except maybe Greenland.) he's planning to leave NATO. NATO exists to protect the Western world, and its highly interconnected economies. it's a globalist project, and the US's role as de-facto leader of NATO gave us the ability to project tremendous soft military power, the same way the dollar lets us project soft economic power.

MAGA are isolationists. NATO protects "the West," but Trump doesn't care about the West, only America. so what if Putin carves up Europe? has Europe even said "thank you?" they're screwing us over, just look at the trade deficit! that's the logic.

[–] [email protected] 9 points 2 days ago

the neutrino was estimated to at 220 PeV. goodness!

[–] [email protected] 64 points 3 days ago* (last edited 3 days ago) (1 children)

the amount of consumer goodwill that statement bought easily pays for any losses from the hotdogs. companies seem hellbent on torching their brand reputation for short-term gains these days. little gestures go a long way.

[–] [email protected] 4 points 3 days ago

if they existed they'd be killer for RL. RL is insanely unstable when the distribution shifts as the policy starts exploring different parts of the state space. you'd think there'd be some clean approach to learning P(Xs|Ys) that can handle continuous shift of the Ys distribution in the training data, but there doesn't seem to be. just replay buffers and other kludges.

[–] [email protected] 2 points 5 days ago (1 children)

what's the difference between a single neural network and multiple neural networks? are BERTs multiple neural networks working together in concert, if you count the encoder and decoder separately? the different layers of a MoE? certainly VLMs count.

[–] [email protected] 50 points 6 days ago* (last edited 6 days ago) (4 children)

I have no clue what to do. I'm American. I'm trans. I voted for Harris. I live in a blue state. I've gone to protests. I've talked to state senators about legislation. I've canvassed. I've talked to Trump supporters and tried to get them to change their minds.

it all feels totally useless. nothing I've done has had any effect. my own mom voted for Trump three times. it's like being chained to an anvil and thrown overboard. I'm swimming but I'm still drowning.

[–] [email protected] 17 points 1 week ago

I'll just have a cup of rain, thanks.

[–] [email protected] 12 points 1 week ago (4 children)

Technology Connections did a video on this rule.

regular US outlets are 120V. regular EU outlets are 240V. P=VI, so to produce the same amount of power as a 240V kettle, a 120V kettle needs to draw twice as much current.

the gauge of a wire determines how much current it can carry without setting insulation on fire. home outlets are typically wired for 15A, around the world. so in EU, 15A service can deliver twice as much power since that's 15A of current at 240V = 3.6kW, while in the US at 120V = 1.8kW.

so EU kettles are twice as powerful, typically.

[–] [email protected] 1 points 1 week ago

I actually don't. :) it's the first thing I disable on my phone. I only capitalize "I" specifically, and proper nouns.

again though, my issue is one with producing the Unicode character of the specific em-dash glyph. maybe some phones automatically convert dashes to em-dashes based on context but I personally wouldn't know how it's done.

[–] [email protected] 1 points 1 week ago

no no, you misunderstand. I'm not talking about the verbal flourish of dashes for interjection - I use those all the time, in this very sentence - I'm specifically talking about producing the specific Unicode character (—) instead of just using the normal ASCII dash (-). the only way I can make the actual em-dash character is by long-pressing. if I do -- or ---, it's just a sequence of normal dash glyphs for me.

 

I frequently see removed in a lot of posts and comments. I'm guessing there's some kind of filter at work? I checked my Voyager settings and I don't see that I've blocked any words, is this happening server side?

From context, it looks like the censored words are pretty common, and not especially potent, curse words. Is this really necessary?

 

Noob question, but are any major Lemmy instances blocked by lemmy.ml? I checked the blocklist at https://lemmy.ml/instances, but I can't easily tell how big those instances are, or what they're blocked for. I'm guessing we don't federate with NSFW servers and spammers but are there any large communities I'm not seeing?

 

Trump's lawyers are petitioning the Court to restrict the Temporary Restraining Orders (TROs) imposed by the lower courts to the named plaintiffs only, which they frame as a "modest" request. This is absurd. The subject of the TRO is a policy change, and it's the policy itself that's likely illegal. TROs are supposed to preserve the status quo while litigation proceeds. Grandfathering a handful of litigants in while letting the change go through in no way preserves the status quo. Striking down the power of lower courts to issue so-called "universal injunctions" would likely be an Equal Protection violation, or it would at least flood the lower courts with tens of thousands of plaintiffs individually filing cases to get the same relief.

How do you think the Court will rule on this? I'm inclined to think (hope?) we'll have the same 5-4 majority uphold this TRO as we had in USAID v. AVAC last week.

 

From a few days back, but I'm surprised this didn't hit harder on the news. Perkins Coie was the law firm contracted to do oppo research on Trump during his 2016 campaign, which sourced the infamous Steele Dossier from FusionGPS. The EO strips the firm of all its contracts, directs EEOC to investigate their (past!) DEI initiatives and hiring, forces all contractors to inform the government if they do any business with Perkins Coie (even as part of other contracts), blacklists every employee from ever being hired by the government, strips them of their security clearances, prohibits Federal employees from interacting with anyone from Perkins on official business, and even bans employees from setting foot in any Federal building! It also threatens other large law firms.

Trump's administration has done plenty of heinous things so far, but this blatant exacting of revenge against entities external to the executive branch feels like a new low.

view more: next ›