MartianNight
No bio...
User ID: 1244
MartianNight
11mo ago
There is an AI track in the Meta Hacker Cup this year. I don't know exactly how it works, but it might be helpful to check which techniques the more successful participants used.
MartianNight
11mo ago
He also doesn't need to make any more appearances in front of large crowds until the inauguration, which will presumably be covered really well because it involves lots of important people besides Trump. I'm fairly certain he'll survive until January.
MartianNight
11mo ago
I'm not sure that's saying much, statistically.
The premise seems to be: for some store owners the expected amount of damages exceeds the cost of boarding up windows. Going with the implied assumption that damages occur only if Trump wins, the expected damages are the product of (probability Trump wins) × (probability angry Democrats smash in my windows) × (cost if they smash in my windows). (It's a little bit more complicated because there are also probability for smashing in windows, smashing in windows + looting products, smashing in windows + looting products + setting the store on fire, etc., but that's not really important for the argument.)
My point is that (probability Trump wins) is not super variable. It's almost certainly between 40% and 60%. But that means the expected damages can increase by only 50% from the absolute minimum. The expected damages and the costs of boarding up windows have to be really close for a difference in that probability to affect the decision to board up the windows. For most stores the decision is going to be the same whether the probability is 40% or 60%.
In short, while you could argue that there would be more stores boarding up the higher Trump's chance of victory is, the mere fact that some stores board up windows does not tell you much about Trump's odds. Most of those stores would be boarding up if Trump's chance of winning were 40% or even less, too.
MartianNight
11mo ago
I wonder if he got the idea from the Demon Core.
MartianNight
11mo ago
Freddie deBoer recently wrote about this, in Big Mommy is Not Coming to Save Us:
This is the “why has the media gone easy on Trump??” narrative, which has somehow flourished for almost a decade now despite the fact that Donald Trump has been covered more critically by our media than any other figure in my lifetime, seemingly to his advantage.
He proceeds to gives a ton of examples from the New York Times.
It’s incredible that so many people sincerely believe that the Times is a secretly pro-Trump publication, as they don’t even bother to pretend that their op/ed section is a space where actual pro-Trump sentiment is going to be shared, outside of a once-or-twice a year novelty piece.
[..]
If they go so easy on Trump, why can they not scare up a single authentically pro-Trump voice for the Opinion page? This recent NYT piece asks nine members of their editorial team to reflect on who they’re voting for and why. All nine are voting for Democrats. It’s a bunch of plugs for Harris or the Democrats generally and one weird endorsement of an environmentalist who stole his wardrobe from the Lumineers tour bus. They couldn’t even find a single staffer to endorse a Republican for appearance’s sake, to ward off the obvious criticism. Not one!
I'm sure there are people on both sides that claim their guy isn't treated fairly, and the other guy deserves more scrutiny. But I think this is a case where the Democrat voters are simply wrong.
MartianNight
11mo ago
Any art created using Photoshop was low status
I'm not a great art connaisseur, but it seems to me this is still the case? If I go to my local museum almost all work on display is created from real materials, and that includes modern and contemporary art.
You might say that musea favor physical art in general (you don't need to go outside to view digital art, after all), but they do still exhibit photos and even film snippets (e.g. on a projection screen), but these are invariably recorded in the real physical world, not purely digital creations. And especially with photos, photoshopping them seems to detract from their artistic value, not add to it.
There is also stuff like Damien Hirst's spot paintings most of which are just colored circles painted on a white canvas. You could almost auto-generate them. Hirst doesn't even paint them himself; he has nameless employees that do it for him. So the artistic merit of these paintings seems to be in the idea. Yet is there any doubt that if Hirst had released these paintings as a collection of PNG files, nobody would have been impressed?
I can only conclude that digital art still very much doesn't “count”, and I expect that AI-generated “art” (which in its current form is not very original) will remain similarly low-status.
The O. J. Simpson case comes to mind. He was acquitted of murder in a criminal trial, but was successfully sued in civil court. Copying from Wikipedia:
So if actually murdering two people only results in a $64 million fine, how in the hell is Alex Jones liable for $1,1 billion for spreading admittedly-hurtful conspiracy theories? He didn't do direct harm; he just said stupid and hurtful shit on the internet.
More options
Context Copy link