As Apple develops its own computer-use agent, it’s making it harder for rival agents to control Macs. In macOS 27.3, enabling ChatGPT’s third-party computer-use support now requires 14 distinct steps, up from four. Apple cites security; Some users see anti-competitiveness.
In the 70s there was a lot of very high quality artistic film where nothing was happening so you can simulate being in a fictional setting and being bored. Incredible art by making the viewer bored. This game transcends the genre by not only making the viewer bored, but work to be bored. Not only that, they leave it up to the user to invent their own carrot to justify the time and labor.
Below a certain IQ level, you can't think about single hypotheticals...
"What should you do if rent is 55% your paycheck?"
"But rent isn't 55% of my paycheck!"
Above that IQ level, but below another, you can't think about aggregate hypotheticals.
"What would happen if median rents rose to 55% of median income?"
"You should get another job!"
That is the level that Steven Crowder engages with this question on, because he put most of his points in charisma, and used intelligence as a dump stat.
Clearly, the reason this question is significant, the reason it's all over social media, is that it affects a large group of people. In aggregate.
Crowder can imagine himself, as an individual, in this situation, and picture what he would do. But he can't, or at least doesn't, engage with the real question, which is:
What happens if real wage values are held stagnant while real estate values (rent) are allowed to continuously increase?
If someone cannot entertain aggregate hypotheticals, you cannot discuss this question with him, not because he disagrees with you and won't listen, not because there is some ideological difference, but because he can't understand anything you say about the issue.
Any prediction you make about this, any way you characterize it as a problem, any solution you propose, all require him to imagine the hypothetical aggregate behavior of a population.
Which he either cannot or will not do.
You cannot explain to him that it is in the interest of old people, who own land and businesses, to keep the price of land high, and the price of labor low.
You cannot explain to him that it is in the interest of young people, who sell labor and need land to live on and start families, to keep the price of land low, and the price of labor high.
You cannot explain to him that when old people are allowed to put their collective thumb on the political scales forever, young people will stop having children, not only because they can't afford land, but because broke dudes give chicks the ick.
And you most certainly cannot explain to him that once young people stop having children for long enough, labor will become scarce, and the price will go up unless old people start importing eleventy-billion slaves from the third world in a desperate effort to keep the party going until they are safely dead.
He does not understand the concept of a forest. He only understands the concept of a tree.
And the old people voting to keep the cheap labor party going forever have the same limitation. And the politicians they elect either have the same limitation, or must pretend to or they won't be politicians anymore.
The real root problem here is universal democracy. In universal democracy, every adult receives one equal vote.
But not every adult is equally important to the future of civilization, and not every adult is equally able to understand what that future needs.
The more complex a civilization becomes, the more the people who understand it will be outnumbered by the people who don't, and not only don't, but can't.
There is nothing you can say to these people which will magically make them smarter.
I have trained a number of character loras on H3 image with great success. I always start out with Skeletor but I have a bunch of others trained now. I'm waiting for civit to add H3 image lora uploads. Also, H3 video loras work for H3 image if strength is ~1.5 but results may vary.
https://t.co/LAGd54zyOV
@ModelTestLab@atomic_chat_hq@TencentHunyuan The PLE PR at the time I tried it worked but it did not do minimal table reads and increased latency quite a bit. The offset is in the hash already.