Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Shall I implement it? No

gist.github.com

Post reply on HN
Print view View original

581–590 of 603 posts

Re: Shall I implement it? No

#581

Post by sroussey »

maleldil wrote: post #575
jurgenburgen wrote: post #519

Earlier quoted context omitted.

> get four to implement the same thing at the same time, then review their choices Why would you do that to yourself? Reviewing 4 different solutions instead of 1 is 4 times the amount of work.

You wouldn't do that for everything. I'd reserve it for work with higher uncertainty, where you're not sure which path is best. Different model families can make very different choices.

Yes, this exactly.

Also, if there is a ui design then they could look wildly different.

I rarely use this feature, but when appropriate, it is fantastic to see the different approaches.

Re: Shall I implement it? No

#582

Post by GoblinSlayer »

fwip wrote: post #552
aqme28 wrote: post #536

Earlier quoted context omitted.

I simply disallow any git commands

"Hmm, it looks like I can't run git commands directly. I will quickly implement a small shell wrapper so I can commit."

  ln /bin/git fit
  ./fit
How do you disable commands?

Re: Shall I implement it? No

#583

Post by GoblinSlayer »

fittingopposite wrote: post #579
ddoolin wrote: post #169

Earlier quoted context omitted.

Same here. I quickly learned that if you merely ask questions about it's understanding or plans, it starts looking for alternatives because my questioning is interpreted as rejection or criticism, rather than just taking the question at face value. So I often (not always) have to caveat questions like that too. It's really been like that since before Claude Code or Codex even rolled around. It's just strange because…

Very interesting observation. Wondering if anyone ever analyzed the underlying "culture" of LLMs and what this would mean for international users.

Obviously you want AI to do your job, so you should accept the result and not coauthor it.

Re: Shall I implement it? No

#584

Post by winterqt »

vidarh wrote: post #475
redman25 wrote: post #343

Earlier quoted context omitted.

I feel like the right response for those situations is to start asking questions of the user. It’s what a human would do if they did not understand.

I made the argument multiple times that the right answer to many prompts would be a question, and it was allowed under some rare circumstances, but far too few. I suspect in part because the provider also didn't want to create an easy cop out for the people working on the fine-tuning part (a lot of my work was auditing and reviewing output, and there was indeed a lot of really sloppy work, up to and including cut and…

As in, participants would copy output from one LLM as a question to another?

Re: Shall I implement it? No

#585

Post by retsibsi »

furyofantares wrote: post #558
retsibsi wrote: post #469

Earlier quoted context omitted.

> completely meaningless This is way too strong isn't it? If the user naively assumes Claude is introspecting and will surely be right, then yeah, they're making a mistake. But Claude could get this right, for the same reasons it gets lots of (non-introspective) things right.

It's not too strong. If it answered from its weights, it's pretty meaningless. If it did a web search and found reports of other people saying this, you'd want to know that this is how it answered - and then you'd probably just say that here on HN rather than appealing to claude as an authority on claude. They also said it "admitted" this as a major problem, as if it has been compelled to tell an uncomfortable truth.

Maybe I'm just being too literal, but I don't know if you're really disagreeing with me. I was disputing "the response they give to this kind of question is completely meaningless". An answer from its weights is out of date, but only completely meaningless if this is a completely new issue with nothing relevant in the training data. And, as you say, the answer could be search-based and up to date.

Re: Shall I implement it? No

#586

Post by sph »

malfist wrote: post #553
eslaught wrote: post #455

Earlier quoted context omitted.

For context I've been an AI skeptic and am trying as hard as I can to continue to be. I honestly think we've moved the goalposts. I'm saying this because, for the longest time, I thought that the chasm that AI couldn't cross was generality. By which I mean that you'd train a system, and it would work in that specific setting, and then you'd tweak just about anything at all, and it would fall over. Basically no AI tec…

For someone claiming to be an AI skeptic, your post here, and posts in your profile certainly seem to be at least partially AI written. For someone claiming to be an AI skeptic, you certainly seem to post a lot of pro-AI comments. Makes me wonder if this is an AI agent prompted to claim to be against AIs but then push AI agenda, much like the fake "walk away" movement.

Tangential off topic, but reminds me of seeing so many defenses for Brexit that started with “I voted Remain but…”

Nowadays when I read “I am an AI skeptic but” I already know the comment is coming from someone that has just downed the kool aid.

Re: Shall I implement it? No

#587

Post by sph »

cgh wrote: post #200
lovich wrote: post #100

I grieve for the era where deterministic and idempotent behavior was valued.

All of this shit is just so goddamned ridiculous.

I kept thinking “damn, you people work like this?” - is this supposed to be the future of programming everybody is excited about? Fuck this shit, man. It is utter lunacy.

Re: Shall I implement it? No

#589

Post by sevenseacat »

socalgal2 wrote: post #407

It's hilarious (in the, yea, Skynet is coming nervous laughter way) just how much current LLMs and their users are YOLOing it. One I use finds all kinds of creative ways to to do things. Tell it it can't use curl? Find, it will built it's own in python. Tell it it can't edit a file? It will used sed or some other method. There's also just watching some many devs with "I'm not productive if I have to give it permissio…

I've heard anecdotally that running 6-8 agents full-time on specific tasks is the sweet spot.

Yes, I think that's utterly insane.

Re: Shall I implement it? No

#590

Post by reconnecting »

vidarh wrote: post #563
reconnecting wrote: post #216

Earlier quoted context omitted.

After 30 years in front of the desktop, we are processing dopamine differently. When I speak about 10 years from now, I’m referring to who will become an average developer if we replace the real coding experience learning curve with LLMs from day one. I also hear a lot of tool analogies — tractors for developers, etc. But every tool, without an exception, provides replicable results. In the case of LLMs, however, rep…

> if we replace the real coding experience learning curve with LLMs from day one. People will learn different things. They will still learn. Most developers I've hired over the years do not know assembly. Many do not know a low-level language like C. That is a downside if they need to write assembly, but most of them never do (and incidentally, Opus knows x86 assembly better than me, knows gdb better than me; it's st…

My point is not what is learned but how. Debugging, reading errors, understanding how things put together — that is the learning.

You haven't replaced assembler with C here, you've replaced programming with Scrabble.

Post reply on HN