It's got good content but also a massive amount of slop. You don't see much straight up AI generated video, but it's incredibly common to see youtubers just sitting there reading a ChatGPT script.
Being all in on AI doesnt mean they have to be all in on bottom of the barrel slop videos. Youtube actually does a great job of not serving you slop if it gets to know you well enough. And it does notify you if it detects the video is AI generated. I would hope tech companies are awake to the fact that people’s feeds being deluged by the absolute uncanniest and disturbing visuals you’ve ever seen is feeding into anti-AI sentiment, and they probably shouldn’t be further enabling that.
I'm not who you're replying to, but I tuned out big movies and TV long ago, and just recently, YouTube became a hostile enough environment for me to start limiting it. When I do go there, it's to watch specific channels I already know. Clickbait titles and slop suggestions just aren't worth the psycholigical hassle when I want to relax.
I had been replacing it with podcasts, but they are so ad-infested now that it's been pushing me more and more to audiobooks. I also watch a fair bit of content on smaller special interest platforms, like CrunchyRoll for Anime or Pepperbox for guns. They're easy to sub and unsub depending on what I'm interested in. I often rent old movies. And more and more... I read.
I had used Prime's videos for a while when they had something I liked, but the ads have gotten so bad that I now go out of my way not to emotionally invest in anything they produce. The experience has left me skeptical of other platform exclusive shows and I generally just go do something else with my time. I used to be a big fan of both Star Wars and Star Trek, but at some point they both started producing Platform Content and in response, I haven't watched anything from either for about a decade now.
> Just hours after the $1,000 arrived, it was divided into smaller amounts and sent to a network of other accounts
Go after all these network of other accounts and cashapp must be held liable. They didn't do proper identity verification... And, these police seems incapable of finding real criminal. They are fool to assume people will use their own account to scam others.
Kinda funny that logo says "Trust Delivered". They don't seem to have 0 problem with such a big scale damage on public internet and simply trying to shrug off the problem.
This is problem with Indian companies, their founder and management always try to hire cheap labor and penny pinch on everything. And, this is the result we see again and again.
PHP team has always been like this. Instead of creating better software they have always tried to add features with no consideration of backward compatibility.
> they have always tried to add features with no consideration of backward compatibility
This could not be further from the truth. Backwards compatibility is a big reason why PHP doesn't have . Odds are if you look at their RFCs and internals mailing list you'll likely find proposals and even implementations for any BC-causing feature you can think of.
In fact, trying to add features while maintaining backward compatibility with existing messy code is exactly the reason why PHP is so messy.
I hope they clean up some of that mess in PHP 9, but no, they're just going to add $shiny_new_feature and make the attack surface even larger. Like, why does the session handler even care about measuring upload progress?
> 8.1 and earlier Not patched: these versions have reached end of life and will not receive an official fix
I would not recommend using PHP. The PHP team keeps breaking backward compatibility and trying to introduce new features instead of focusing on fixing bugs and hardening the compiler.
i don't look at code anymore. I do write great specs and run whole app/program in my head. And I do read code where i need to read which is probably 5-10%.
It is similar to how i don't look into assembly code anymore and trust llvm.
Yeah, and the noise is also per model, run the same model five or ten times and put the confidence around it. Then try different models and different workflows. Thats actually how I do it when I build workflows and automations: test, test, and test, collect the statistic on where and what works better, then combine.
reply