For this race, I agree. The worst he can be is a Fetterman… Who sucks but is still better than fucking Dr. Oz.
But be careful with this line of thinking. It’s very similar to how the right ended up voting for Trump. It’s actually how a lot of really awful governments gained power. “Anything different” than the current catastrophe does not inherently mean better.
Im pretty wary… i think hes probably got a bit of authoritarian streak and probably got this tat when he was young and dumb and in the military which leans right. I also think people change and he has changed. I think he’s taken a hard left on the economy and class consciousness.
But is his authoritarian streak gone? His reddit comments show that at the time of posting he had little sympathy for sexual assault victims and was at the very least kinda weird about black people. Hes also very pro gun, which I dont find to be an inherent issue and I even agree the left needs to be more prepared and armed, but when lined up with other problematic views it stresses me out. And I dont realllly trust all of his hand waving around the tattoo and reddit comments.
But maybe should we embrance that hes a bit of a bulldog in an era when the left is so toothless? Also its almost impossible to do worse than Susan Collins lol.
Not super common or super niche. I use R. And it completely made up code a year ago. Sometimes I still does, but less. And when I ask it for citations it can make shit up too. I really stand by the assertion that it needs a lot of babysitting.
But, between it getting better and me getting better at asking and some patience, I get what I want. But, it does require a lot of fine tuning and patience. But its still just faster than googling. And I could see the argument that the models haven’t improved but that they just have access to search engines now and that I’m mostly using them and a search engine. And sometimes they’re so whacked out I’ll ask them to search for something but theyll tell me they don’t have access to the internet and they’re so absolutely convinced of that that I have to close that chat and start a new one.
If you feed it in documentation or ask it to search for its answers in substack (or really just whatever search constraints you want) and then tell it to give you the links it used, you might have a better time. This forces it to look up an answer instead of hallucinate one. And when it gives me code, more complicated things usually fail pretty hard at first and I have to feed it the error output for a few rounds and guide it a lot.
I don’t know what to tell you. I have them successfully compiling tables of search outputs to compare different things for method development and generating code, saving me hours of work each week. It all needs to be checked, but the comparison comes with links and the code is proofread and benchmarked. For most of what I do it’s really just a jacked up search engine, but it’s able to scan webpages faster than me and that saves a lot of time.
As a hobby, I also have it reading old documents that are almost illegible and transcribing them pretty well.
I really don’t know what you’re doing that you’re just getting nonsense. I’m not.
Technology these days works in that they always lose money at the start. Its a really stupid feature of modern startups IMO. Get people dependent and they make money later. I don’t agree with it. I don’t really think oir entire economic system is viable though and that’s another conversation.
But LLMs have been improving exponentially. I was on board with everything you’re saying just a year ago about how they suck and they’re going to hit a wall even. But the don’t need more training data or the processing power. They have those and now they’re refining the LLMs. I have a local LLM on my computer that performs better than chat GPT did a year ago and it’s only a few GB. I run it on a shitty laptop.
LLMs with access to the internet are usually about as factually correct as their search results. If it searches someone’s blog, you’re right, the results will suck. But if you tell it to use higher quality resources, it returns better information. They’re good if you know how to use them. And they aren’t good enough to be replacing as many jobs as all these companies are hoping. LLMs are just going to speed up productivity. They need babysitting and validating. But they’re still an extremely useful tool that’s only going to get better and LLMs are here to stay.
I get the desire to say this, but I find them extremely helpful in my line of work. Literally everything they say needs to be validated, but so does Wikipedia and we all know that Wikipedia is extremely useful. It’s just another tool. But its a very useful tool if you know how to apply it.
I get your concern in a sense because kids in Rochester do have it pretty rough. While crime stats in the us and NY are down, violent crime Rochester specifically has stayed pretty steady since the 90s and is higher than national and NY crime levels.
But I don’t think this will be the thing that impacts these kids and school districts the most. The issues Rochester face come from poor incentive management for schools, high teacher burnout and poor employee retention, large classroom sizes, issues with corruption within specific school systems, and concentrations of low socioeconomic people.
I need to admit I left the public education world in NY a while ago. But it used to be that the worse a school performed the less money it got. Because schools only get a small percentage of funding from the federal government that gets pulled from them when they underperform (I very quickly looked at the numbers and it appears Rochester gets less than the national % average in federal funding, about 10% compared to the national 13% depending on what year you’re looking at). So schools with students who needed the most help either lost money or just tried to cover up student issues to hide them instead of address them. As long as this and property tax funding is the model and admin dictates their own salaries, schools aren’t going to improve.
IMO, it’s even specifically DEI policies that briefly helped some schools excel in the us during school bussing. Bussing students forced integration of students from all different backgrounds and socioeconomic groups. Rich and attentive parents with more bandwidth made sure whatever school their kid was at was well funded and well run. Kids from different backgrounds became friends and humanized each other. Kids from poor backgrounds were offered support systems outside of gangs and violence. And these poor kids didn’t poison the well and bring crime to their new schools. Kids are kids. If you offer them opportunities, they don’t choose gangs and violence.
What will really fix schools and improve things for kids
I don’t have cartridges and I’ve had no issue with the ink heads in the years I’ve had the printer.
You’re probably right, but I’m just printing random documents on occasion.
I have an epson eco tank. No more ink cartridges.
I dunno. I feel like a few countries made a similar bet (but to a higher extreme) during the Arab spring and it went extra bad. It’s important to be careful about who and what you’re trying to replace an authoritarian system with.