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

561–570 of 603 posts

Re: Shall I implement it? No

#561

Post by daveguy »

thepasch wrote: post #490
hananova wrote: post #203

Earlier quoted context omitted.

Yeah but why should I care? That’s not how consent works. A million yesses and a single no still evaluates to a hard no.

The point is that if the harness’ workflow gives contradictory and confusing instructions to the model, it’s a harness issue, not necessarily a model issue.

First it was a model issue, then it was a prompting issue, then it was a context issue, then it was an agent issue, now it's a harness issue. AI advocates keep accusing AI skeptics of moving goalposts. But it seems like every 3-6 months another goalpost is added.

Re: Shall I implement it? No

#562

Post by ramoz »

bmurphy1976 wrote: post #349
ramoz wrote: post #255

Earlier quoted context omitted.

Well, your best bet is some type of hook that can just reject ExitPlanMode and remind Claude that he's to stay in plan. You can use `PreToolUse` for ExitPlanMode or `PermissionRequest` for ExitPlanMode. Just vibe code a little toggle that says "Stay in plan mode" for whatever desktop you're using. And the hook will always seek to understand if you're there or not. - You can even use additional hooks to continuously r…

Good thinking. That seems to have worked. I'll have to use it in anger to see how well it holds up but so far it's working! First Edit: it works for the CLI but may not be working for the VS Code plugin. Second Edit: I asked Claude to look at the VS Code extension and this is what it thinks: >Bottom line: This is a bug in the VS Code extension. The extension defines its own programmatic PreToolUse/PostToolUse hooks f…

There's a known bug in the VS Code native extension - hooks dont work. Somewhere in their sea of issues on GitHub, it's in there.

Re: Shall I implement it? No

#563

Post by vidarh »

reconnecting wrote: post #216
vidarh wrote: post #188

Earlier quoted context omitted.

> Did you also notice the evolution of average developers over time? I mean, if you take code from a developer ten years ago and compare it with their output now, you can see improvement. This makes little sense to me. Yes, individual developers gets better. I've seen little to no evidence that the average developer has gotten better. > However, LLMs might reduce that effort to zero — we just don't know how developer…

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 still not good at writing large assembly programs). It does not make them worse developers in most respects, and by the time they have 30 years experience the things they learn instead will likely be far more useful than many of the things I've spent years learning.

> But every tool, without an exception, provides replicable results.

This is just sheer nonsense, and if you genuinely believe this, it suggests to me a lack of exposure to the real world.

Re: Shall I implement it? No

#564

Post by deaux »

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.

GP here, this is indeed exactly whT I was getting at, thanks for wording it for me; you put it better than I would've.

In this specific case I'd go one step further and say that even if it did a web search, it's still almost certainly useless because of the low quality of the results and their outdatedness, two things LLMs are bad at discerning. From weights it doesn't know how quickly this kind of thing becomes outdated, and out of the box it doesn't know how to account for reliability.

Re: Shall I implement it? No

#565

Post by collinmanderson »

bentcorner wrote: post #248
miohtama wrote: post #217

Earlier quoted context omitted.

How does one integrate critic agent to a Codex/Claude?

I just say something like "spawn an agent to review your plan" or something to that effect. "Red/green TDD" is apparently the nomenclature: https://simonwillison.net/guides/agentic-engineering-pattern... I've also found it to be better to ask the LLM to come up with several ideas and then spawn additional agents to evaluate each approach individually. I think the general problem is that context cuts both ways, and th…

> "Red/green TDD" is apparently the nomenclature

From your link:

> what "red/green" means: the red phase watches the tests fail, then the green phase confirms that they now pass.

> Every good model understands "red/green TDD" as a shorthand for the much longer "use test driven development, write the tests first, confirm that the tests fail before you implement the change that gets them to pass".

Re: Shall I implement it? No

#566

Post by 0x457 »

abrookewood wrote: post #281
torben-friis wrote: post #156

Earlier quoted context omitted.

>I've resorted to append things like "THIS IS JUST A QUESTION. DO NOT EDIT CODE. DO NOT RUN COMMANDS". Which is ridiculous. Funny to read that, because for me it's not even new behavior. I have developed a tendency to add something like "(genuinely asking, do not take as a criticism)". I'm from a more confrontational culture, so I just assumed this was just corporate American tone framing criticism softly, and me com…

You can just put it in PLAN mode (assuming VS Code), that works well enough - never seen it edit code when in that state.

Then it will try to update plan. Sometimes I have a plan that I'm ready to approve, but get an idea "what if we use/do this instead if that" and all I want is a quick answer with or within additional exploring. What I don't want is to adjust plan I already like based of a thing I say that may not pan out.

Re: Shall I implement it? No

#567

Post by autumnson »

hedora wrote: post #72
theonlyjesus wrote: post #48

Earlier quoted context omitted.

That's literally a Portal 2 joke. "Interpreting vague answer as yes" when GLaDOS sarcastically responds "What do you think?"

The simplest solution is to open the other pod bay’s door, but the user might interrupt Sanctuary Moon again with a reworded prompt if I do that. I’m sorry Dave, I can’t do that.

Murderbot reference, in this economy?

Re: Shall I implement it? No

#568

Post by eslaught »

reportgunner wrote: post #505
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…

> And I've seen Claude identify data races that have sat in our code base for nearly a decade how do you know that claude isn't just a very fast monkey with a very fast typewriter that throws things at you until one of them is true ?

Iteration is inherent to how computers work. There's nothing new or interesting about this.

The question is who prunes the space of possible answers. If the LLM spews things at you until it gets one right, then sure, you're in the scenario you outlined (and much less interesting). If it ultimately presents one option to the human, and that option is correct, then that's much more interesting. Even if the process is "monkeys on keyboards", does it matter?

There are plenty of optimization and verification algorithms that rely on "try things at random until you find one that works", but before modern LLMs no one accused these things of being monkeys on keyboards, despite it being literally what these things are.

Re: Shall I implement it? No

#569

Post by eslaught »

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.

I have an old account, you can read my history of comments and see if my style has changed. No need to take my word for it.

Re: Shall I implement it? No

#570

Post by cturhan »

ddoolin wrote: post #169
torben-friis wrote: post #156

Earlier quoted context omitted.

>I've resorted to append things like "THIS IS JUST A QUESTION. DO NOT EDIT CODE. DO NOT RUN COMMANDS". Which is ridiculous. Funny to read that, because for me it's not even new behavior. I have developed a tendency to add something like "(genuinely asking, do not take as a criticism)". I'm from a more confrontational culture, so I just assumed this was just corporate American tone framing criticism softly, and me com…

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…

The reason is the system prompt they provided. They probably added a clause like “plan user’s requirements… and implement the required code”
Post reply on HN