tstrimple

Really puts egg on the face of the anti-LLM folks who claim they only produce garbage code. Turns out they are very good at exposing the garbage code humans have been producing for decades.

Nearest example I can reach is Elon Musk. Somehow he loves the culture novels but ends up being an evil piece of shit which is counter to basically everything the novels espoused.

The traitors had a hand in the dumbing down of the country as well. There's a reason Trump loves the poorly educated. All Republicans do. That's their base. The problem they are running into is the inmates are now running the asylum. The ignorant true believers are being elected in place of the former conservative elites who created the propaganda networks to create those ignorant true believers in the first place.

> I started it as a way to learn things. Mostly picked questions that were on the edge of my understanding, researched the problem for myself, and then tried to provide clear and detailed answers. So I got a lot out of it too.

This is literally what I did to learn node.js at a deeper level. Was a lot of fun for a while.

This is the bullshit I find infuriating. Some conservative bloviating about Charlie Kirk and the reaction to his death from internet randos but they completely fucking ignore the vile shit actually elected representatives in their preferred party say.

"Robert Mueller just died. Good, I'm glad he's dead. He can no longer hurt innocent people! President DONALD J. TRUMP"

Fucking hypocrites.

No it wouldn't. The vast majority of shitty things have been done via executive order. Republicans are completely fine with congress being dysfunctional. That serves their goals.

> You are asking Dems to do something they literally cannot do while minority party. You are asking Dems to do an impossible thing and then complaining they cannot.

It's Murc's law all the way down.

I've found Claude to be the least sycophantic OOTB. But I've also been running custom system prompts for a while now and have lost touch with what the default behavior looks like. I've had to balance "push back when I'm wrong" and "act like a peer" style instructions which have lead to some pretty annoying one-upmanship style dialogue.

More specifically those instructions alone caused the model to narrow in on tangentially incorrect details while completely ignoring the thrust of my argument. I get where this could cause issues with writing code, but I do a lot more than write code with CC and the conversational voice is important to me.

Not sure what harness you use, but I've found you can influence LLM "voice" a lot more by replacing system prompts rather than just adding content to a CLAUDE.md file. I see a lot of people complaining about how Claude "talks" but very little talking about how to adjust it.

https://code.claude.com/docs/en/agent-sdk/modifying-system-p...

The main benefits to modifying the system prompt or output style is that the harness "reminds" the model about those things every turn. So if you've experienced things like Claude Code losing track of instruction from CLAUDE.md in large contexts, it indicates you may need to add the instruction to the system prompt instead which is far more durable.

"People won't play difficult games anymore because the new Wolverine game gave them a blue line to follow everywhere and removed any need for decisions or exploration."

I use Claude for this stuff all the time. The difference is I have Claude create repeatable scripts which do the actual things on a schedule. Claude is never invoked in on a schedule, only by me. One example is job board search filtered for roles I'd be interested in. The script itself includes all the things I care about like being remote. Clear salary ranges. Type of work I want to be doing. Claude built a good script to scrape multiple sources and aggregate them into a sqlite database. It filters these out until a number of relevant roles (or amount of time has passed) that the script sends me a message and adds a reminder to my Apple account to check in on new job listings. At that point I'll invoke Claude myself to summarize and prioritize the listings. If there are gaps in what Claude finds vs the script, that's a point where I iterate again on the script. My goal is for the job alerts and filtering to be fully useful even if Claude and all LLM tools disappear tomorrow. And also not give such powerful tools the ability to run on a schedule on vague instructions.

More often than not these sorts of stories are "I ignored all of Claude's warnings and told it YOLO and I'm totally surprised my production database was deleted!"

I have literally never seen this and it doesn't make much since because the turns are flagged with metadata. It should be impossible for Claude to confuse your text for its own. I know because I use this metadata all the time. I've built hooks and processes against it.

Australia's minimum wage is significantly higher than in the US. An adult working as a burger flipper at McDonalds can expect a minimum of $27.81 / hour.

Nothing about your post indicates you deserve any respect. Slop comment which adds nothing to the conversation. Probably AI.

Switching between headphone output and speaker output on my system is also just one line. I have no idea what that line is because I created a toggle-able icon in waybar which is a really simple script that switches between the two outputs and shows me the correct icon. I didn't need a "tool" to do it, but the packaging and UX are far better than popping up a terminal any time I want to do this.

Chronic liar tells another lie. More news at 11.

There seems to be some fundamental thing in humans which makes a subset of them follow narcissists like messiahs. MAGA and Elon stans aren’t that far apart.

No. The conclusion was not totally different. It recommended one of the explained options and recommended one extra hour as a buffer. I feel like this is a distinct reading comprehension issue if you think the final recommendation came out of left field.

As you’ve described it nothing has changed. Word used to add squiggles and people who didn’t research the meaning blindly accepted the recommendation. Same shit. Different day.

There's nature and nurture and regardless of where the lines fall, the implication from idiocracy is the morons cannot adequately nurture any positive genetic intellectual variance. If they can't nurture it, it doesn't really become an advantage. I'm fully convinced we've lost numerous genius individuals all capable of changing the world because their entire life was spent in a rice field or similar setting. What would Einstein have achieved if he hadn't had an electrical engineer as a father who helped unlock his fascination with the world? I have no idea.

> Equilibrium is one of the more ridiculous representations of a hyper controlling state.

All forgiven by me for bringing us the ridiculous gun katas.

https://www.youtube.com/watch?v=U02E2sjwlLM

Not just gay, but trans. Some people don't just take on the opposite gender for a while just to see how it feels. Some become literal furries. Fluffy tail and all. All provided for free on a whim.

My son had zero interest. Finally ended up getting his license at 18. My middle daughter went through the process to get her learners permit on her own initiative as soon as she was able. Likely a difference in friend groups. He was mostly interacting with his friends online. Some of her friends already have a license so she's seeing the independence of no longer needing to text a parent for every change of location. She also has friends who are working while my son mostly had layabout friends.

> Once again, Gen Z is destroying the [world destroying product] industry.

It's about damn time. This burden has been on us millennials for too long now.

I think it’s just because he’s a hateful bigot. That resonates with other hateful bigots. “Trumps” basically anything else. They don’t care how much they have to pay for gas if they think brown people are suffering more than they are. Doesn’t matter if Trump shits on a golden throne. He hates the same people they hate so they love him for it.

Neither are the many politics threads shitting on California.

The trick to avoid needing a plan B is to have no plan A.

I do these sorts of things all the time for cabin rentals. It's pretty standard there. But the cost typically also reflects it. I'd be fine with the same sort of situation on AirBnB, but the fees and such change the calculations really quickly.

The vast, vast majority of RTO doesn't include any sort of office. You're lucky to get a permanent desk these days. The new hotness are hot swap desks. Just grab the first available work surface in a completely open office environment so you can take your Teams calls and still not interact with anyone in the office.