From the stuff I’ve looked at, it’s often nonsensical. Doing stuff in unconventional ways for little apparent reason. Like, in BMWs example, you have to use a lift or jack up the car level to just check the transmission fluid; or the mostly-plastic cooling system which fails and results in the engine being ruined often. I’ve seen other weird engineering choices in electronics too (have done contract work for a German company).
Hmm, I’m having trouble just landing interviews. Probably 50-100 applications before an interview. The possibility of losing my home is all that keeps me motivated. I do like software engineering though, and am always working on personal projects that probably won’t make me money.
But yeah, I do find it hard to motivate myself for practicing leetcode problems, and I do suck at them, since most stuff is just about remembering algorithms I learned in a class more than a decade ago, and the few times I’ve needed to implement them myself in my career, I’d just look them up.
I’m not sure it’s “AI” causing lack of demand though. It’s just most large companies have given up “innovating,” and are in rent-seeking mode now. In the US, at least, I’ve heard that many companies are only hiring devs in India to backfill the devs they’ve laid-off or quit. I’ve interviewed with an agentic AI startup, and it was pretty obvious it was “Actually Indians” (underpayed, overworked devs would fix the shitty code their AI agent would write). Most of the interviews I’ve gotten were to lead a team of all-remote Indian devs.
I’ve always found needing to manually add a class instance parameter (i.e. self) to every object method really weird. And the constructors being named __init__. Not having multiple dispatch is kinda annoying too. Needing to use decorators for class methods, static methods, and abstract classes is also annoying. Now that I think about it, Python kinda sucks (even though it’s the language I use the most, lol).
Haven’t used it yet, but venice.ai looks interesting; they have a good privacy policy. Right now, I just use ChatGPT with the “improve models” setting turned off, and use “temporary chat” mode. I don’t really trust OpenAI to be doing the right thing though. I’ve used 14B models locally, but they aren’t as good as 72B+ models.
I like ranger (https://github.com/ranger/ranger).
Yes, and it’s very popular, at least with younger people. Asmongold, Rogan, Shapiro, Fresh and Fit, Charlie Kirk, etc. They’re usually further right than mainstream too. The algorithms boost them, they typically sell snake oil/crypto, and are suspiciously well-funded (last year, many were found to be taking Russian money).
Privacy policy changes are worrying. They also implemented Privacy Preserving Attribution, which sends anonymized data to advertisers, and enable it by default. I personally like LibreWolf.
If you’re implying you want to spend as much time away from your family as possible, you could just get a divorce. Wouldn’t have to sell all your life to shareholders then.
I don’t think the next generation of the US is looking over at the Foxconn dorms and their suicide nets with envy. Even the average manufacturing job still in the US doesn’t pay a living wage (due to unions being destroyed). Not that these tariffs will bring manufacturing back anyway.
current large models like ChatGPT model more “neurons” than are in the human brain
I don’t think that’s true. Parameter counts are more akin to neural connections, and the human brain has something like 100 trillion connections.
I have done that a few times, but editing it so it doesn’t sound like AI, and to remove characters AI likes to use that aren’t on standard keyboards (like the em dash). Not sure if companies care about cover letters or not either.
I think shortages will be short-lived as companies and retailers just have to suck it up and pay more. People won’t be able to buy as much stuff, so layoffs and a recession or depression are likely, but there’s not much I can think of doing to prepare for that.
Interesting. Didn’t know Canis Lupus wasn’t native in South America. Also interesting that the Pampas Fox isn’t even in the same genus as dogs, yet interbreeding was successful (by happenstance?).
Years ago, I used to just ignore letters. Nothing happened for years. Moved and switched ISPs (smaller, regional), then was banned by that ISP after a few months, lol. Copyright holders can sue you, so it’s just an odds thing.
I just use a VPN now. Used Tribler before I set up my Arr stack, because Tribler didn’t work well with it. Tribler is faster than i2p if you use 1 hop, but theoretically, the Movie Picture Association could run an exit node and still sue you, if you only use 1 hop. (Actually, it’d probably be a legal gray area for them to run an exit node because they would be facilitating piracy).
Yeah, I have much more time in FreeCAD than PhotoShop. I don’t think it’s either intuitive or efficient, lol. But, I was surprised that I was able to both create model to both 3D printing, and create paths for my plywood router. In my experience KiCAD is near peer, because Altium also sucks.
Yeah, for all I know, GIMP is as intuitive and efficient as Photoshop. I.e. they’re both as un-intuitive to me, and I’m not efficient in either (I probably haven’t use PS in a decade). But, I guess being intuitive is a good “hook” in regards to market share. I am a software engineer, and do prefer (neo)vim, but I guess I started out on NetBeans. I kind of see that as a level-up in regards to efficiency; not sure if that exists in other industries.
Not saying GIMP’s UI is great (I only use it occasionally), but efficient UI isn’t necessarily an “intuitive” UI. I.E. an intuitive UI may not be efficient for a professional that takes the time to learn it and works with the UI ~40 hours/week.
Fuck grass, especially non-native grasses.
(Sorry, I have a hatred for lawns that probably stems from being forced to mow as a kid.)
Also, I think children are sociopaths naturally. I know I was. I think there are studies that show empathy doesn’t fully develop until later in life. It’s also one reason child soldiers are used.
Haven’t played it in years, but really liked it. Probably have hundreds of hours in it. I really liked a Star Trek total conversion too.
Guess I just prefer languages that do it this way:
class AClass { var aProp = 0 fun aMethod() { aProp++ } }Though I suppose confusion and bugs can happen when you do something like:
class AClass { var aProp = 0 fun aMethod(aProp: Int) { // `this.aProp` is needed to access the property } }