They are described as being useful for novel problems, but no matter the vendor or if it is an agentic system I set up, or I watch a reasoning model prattle on, I understand those things as filters that are limiting the conceptual search space, and regardless, it is very easy to bump up against the limits. I understand the use as a rubber duck, that's fine, but this cult like belief that we can't criticize is out of control. My skill issue is that I keep trying to use all of these skills and yet I don't have this default pro-LLM belief which seems to be the requirement. Just today I got multiple models to invent QEMU configuration items that don't exist while trying to solve my problem, which I guess I have to say now is novel by your list here, but it was also something pretty easily found in the documentation I later found out... and even knowing that I wasn't able to get the models to understand that even when I explicitly gave them that information. I've had other experiences like trying to squeeze a watermelon seed. At this point, there is just too much risk of anything they produce giving me a wild goose to hunt down. It is absolutely maddening, but also the people telling me I need to pray about it aren't helpful. These things have not fundamentally improved since they did an impression of D&D games, but I can totally see why people would think that. They approach a database with a natural language query interface but that implies that it knows the context of your language and that they have the data, and when they don't, they make it very difficult to find their errors because they are so adjacent to correct.
That’s what his llm CLI is? I’ve been waiting for this release so I can take my existing notes on best practices coding with LLMs (which I’ve been doing for both work projects and side projects) and try some experiments with rolling a coding agent myself instead of using Claude Code or VS Code’s Agent mode. If it works well then other folks on my team might switch to it too.
I don’t get where you get the idea that people aren’t actually using this stuff and being productive.
I don't know the context that could be cordial when there is so much outright dishonesty about the state of LLM uptake. Can we have some guidance about how to call out bullshit? I know that the fake it to you make it or sell the shovels to the gold rush people are staples on here, but I don't what to do with a technology that is purely for that and seemingly for nothing else. Why do all of the breathless pro-LLM posts get pushed to the top of every LLM story, and you have to go to the second comment aways to see the avalanche of people calling bullshit? Thank you for your time moderating, it would be helpful to understand the guidance in face of the brigrading on here.
I guess I'll need clarification -- getting trapped in a non-productive quagmire is the experience of a lot of people. The author chooses when to use terms in different contexts to get out of arguments ... and also, we still don't have any good way to test these things, so I'm not sure where I'm running afoul of the rules when it seems like everyone else is.
I guess I'm having trouble following this, what is the non-productive quagmire? It's someone's Show HN and you started grumping at them out of nowhere - that's what it looks like externally. There is no right way to do that on HN just like there isn't in most typical social contexts. You can just offer your critique without the overwrought bombast, it's what most other people do. I imagine you don't begin work emails in the style of your flagged comment.
I guess I see the overton window in the other direction from this? If I had a coworker who made a random insult machine that emailed other employees random, but deeply personal insults, and some people thought it was funny, and the CEO said that we should reflect on those insults and learning something, I would most definitely begin my emails with probably a stronger tone. And that is the reality of the fatigue that I see, in my professional circles, system architects, as they constantly have to re-litigate every new AI idea that is fundamentally incompatible with deterministic systems.
When I try to bring up this reality, the entire argument is triangulated "people find the insults funny," "maybe you should learn the history of roasting people," "my friend at the other large corp is insult maxing and it is the future, maybe the insults will get less devastating..." I look at people like Ed Zitron, who point out that even if they were five nines accurate how corrosive that would be, and I guess I see my rhetoric here as tame. It was literally from that context, so it did feel entirely appropriate, and honestly still does. And anyone who has asked the models deep questions that they know the deep answers can attest to, these are insult machines.
It really does feel like an invasion of the body snatchers for a lot of people, and I do believe this is completely invisible to non-practitioners, managers, junior people and others in many spaces. I've been trying to give voice to this, as have others on here in even starker (but perhaps more cordial as to be unnoticed) terms. I have a lot of theories about this, like there are just so many overloaded terms between the different groups of people in technology. A model that has doubled its accuracy is still very far away from even the most basic of traditional systems, like say a 555 timer is something that seems to be a missed idea in many places.
I'm writing this with hopefully a more business tone, but this language sure does seem to understate the extent of the problem, and the triangulation of the discourse now does like you say maybe doesn't have room for honest expression. I also feel the need to be almost insultingly verbose because it also seems like my sentiments should be obvious, because in other circles they are boring and not new.
My own non-productive quagmire is the constant exploration of all of these vendors and these techniques and seeing them make errors that are so hard to find as to be antagonistic, and vendors who think they are in control but can only push hidden bias. There's also a rhetorical problem with complex issues you start to look and sound like the crazy wall meme from It's Always Sunny, so, I understand there is a limit to effectiveness of describing the problems, I guess I think of Ed Zitron being hard to parse by most anyone... or I think of Jeff Goldblum's character in The Lost World being frustrated that he uses plain simple language but yet the danger does not seem to be communicated.
Bless you if you even half skimmed this, and thank you for your time.
Edit: Sorry, specifically with Simon's posts, I believe Simon to be genuinely curious but he's very happy to appropriate this situation on the side of the shovel seller and allude to these problems, but the audience for these posts are directly inspiring people to think of these tools as appropriate in contexts where we already know they fail, and so, I see this posts what like at least once a week now on here? And it always brings the most people triangulating the arguments like stated above, and I've had an ongoing feedback to Simon and it is puzzling, like his most latest post where he first said they were understanding things, and then when I said they don't, he said, essentially yeah I know but believing that is why I like them and I just don't know what to do with that kind of disingenuousness, unless people are literally talking about faith then I think people should talk about the pseudo religious things going on and stop describing those things as rational.
If I had a coworker who made a random insult machine
Nothing of the sort happened, though? Why even make this gigantically escalatory analogy? You don't have to like the work presented and you still have the option of non-yelly critique or saying nothing. You're acting like you've done the work of persuasion for your position and everyone is or should be as incensed as you are. And that's clearly not the case.
I also added another reply to a false assertion by the OP here in this thread. Is that better ? It takes a long time to research all these falsehoods that are leading to the hype.
