sungho_

Just out of curiosity. What makes startups different? Is the lack of structure the key difference?

> This is still true - a compiler can never win this battle. All a human programmer has to do is take the output of the compiler and make a single optimisation, and he/she wins. This is the advantage that the human has - they can use any of a wide variety of tools at their disposal (including the compiler), whilst the compiler can only do what it was programmed. The best the compiler can hope for is a tie.

This is awkward

I really love this project and want to keep rooting for its success. But the presentation is currently very confusing. The biggest issue is that I can’t tell what is fictional 'lore' and what is an actual feature of it. I don’t think you have to sacrifice the concept to make things clearer; it’s definitely possible to maintain a strong concept while ensuring visitors aren't lost. Anyway, I hope this goes well!

How about adding these texts and reactions to LLM's context and iterating to improve performance? Keep doing it until a real person says, 'Yes, you're good enough now, please stop...' That should work.

It's fun

Honestly, it must have been annoying yet fun. If I'd gotten something like that, it would have amused me all day.

"....what you think counts. Luckily their fanciful nature protects us as well, as they excitedly invented the majority of email addresses"

It went well, right?

If a reader gets angry simply because the author used ChatGPT instead of Claude, then the reader is an idiot.

They say no childbirth means no children. But must children be such an inconvenience? Even if Korea ceases to produce human infants, AI children may be born in their stead. Or, through reverse-aging, the elderly could become the new children.

Orion's arm

The links you provided need a control group to be considered proof. The key is how it compares to when counseling was provided by just a friend, not an expert.

What surprised me most was how large a single neuron is.

You are confusing ‘thinking it is possible’ with 'being certain of it.'

Contrary to what one might think, if you were to live forever, you might end up taking insane risks and trying anything and everything.

It's completely different to say that the government took the lead in an industry that didn't exist before and to say that it provided support to an already successful industry. Of course, what I said was wrong refers to the former. In fact, the government supports all industries to some extent, so that can't be a label.

I am Korean, and what you are saying is a lie created by anti-Korean people in Japan. Do you really think it makes sense for a government experiencing an economic crisis to desperately seek revenue sources and hope to overcome the crisis by funding a cultural industry that hasn't even succeeded yet?

It's not true at all

> It would be an interesting test to hand the unfixed revision of the code to an LLM while also giving it the docs, and say “make any fixes to make this conform to standards of the framework and libraries”.

For your info, a few days ago, I was trying to migrate Tailwind 3 to 4 in a codebase where I had just set up the boilerplate, and I went through hell and eventually gave up. I used the Tailwind 4 documentation and Claude 3.5 sonnet in Windsurf IDE, and the part related to migration in my codebase was probably less than 50 lines. All of those tasks would take just a few minutes if a person did them directly (excluding reading the documents)

I think it would be appropriate to translate and share one of Korea's memes here.

Foreign Guy: I heard that there are refrigerators specifically for kimchi in Korea. Is that true? Me: Yeah, not everyone in Korea has one, but I do. Foreign Guy: I've asked this for the 10th time now, and everyone has answered the same way.

Sam Altman and OpenAI aim to become the gods of a new world. Compared to that goal, it makes sense that money feels trivial to them.

It may not be a complete solution, but if they enforce remote work in both the public and private sectors and switch to a four-day workweek, the situation will be better a lot. Research shows that even doing that won't decrease productivity; in fact, it might even increase it, so there's no sacrifice involved. What do you think?

Doesn't the fact that it can only accomplish tasks with benchmarks imply that it has limitations in intelligence?

I've already started not to notice the quality differences in the photo-like images produced by each image generation model.

Now, examples of image or video generation models showing off how great they are should be stickman drawings or stickman videos. As far as I know, no model has been able to do that properly yet. If a model can do it well, it will be a huge breakthrough.

What if anti-aging technology is developed and you can live up to 1000 years?

People with incurable diseases or just strange individuals can volunteer for such experiments. After that, if they write about their experiences, it could be featured first on Hacker News.

Oh then his words were valid

This can be adjusted by moderation. Why not have the chatbot be compliant at first, and then increase the moderation level as the relationship grows and the lock-in effect occurs?

> There's no female loneliness epidemic because some men have multiple female partners (unlikely)

Some men definitely have multiple female partners, but it's only not at the same time. This isn't just wordplay. A few men date many women, so many men are single while fewer women are.

Thank you for saying that. And be warned: the below part of this comment is self-promotion, so don't read it if you don't want to.

- - - - - - - - - - - -

I have a post[1] about how I use my mouse. From that, you can get a sense of how I use tools. I think it will be interesting because while there are many articles about how to use keyboards, I couldn't find articles about mice. I'll keep posting interesting writings, so I'd be happy if you subscribe. I think when I submit a post introducing my room to HN, they will be at a loss for words.

[1]: https://www.sungho.blog/p/maximizing-mouse-not-keyboard-pote...

Analysis paralysis is not inevitable. It's just a case of lacking tool intelligence