jdw64

11 ms·

181.
by jdw64 2mo ago
The homepage design is nice, and I think anonymous chat is fun too. Even though the programming topic probably has a pretty clear target audience, I'm surprised that kind of hateful chat still shows up.
182.
by jdw64 2mo ago
My blog posts don't get any replies even when I post them on Hacker News, but everyone wants to communicate with others. I actually think anonymous chat isn't that bad. I think the problem is people who abuse the goodwill of writi
183.
by jdw64 2mo ago
I think so too, but the problem is that this was a 'serious' project. Some tech leaders genuinely believe that AI writes better code than humans, and the sad part is that they're somewhat right. So these days, I'm not su
184.
by jdw64 2mo ago
These days, I see a lot of people using AI for everything except hobby projects. Recently, I joined a maintenance project where the team leader even instructed us to use AI exclusively. So I just keep one hobby project that I write by hand,
185.
by jdw64 2mo ago
>I’ve created a full-time job for myself talking to Claude Code. Maybe this person and I could become friends.
186.
Cookies Became the Web's Memory(makonea.com)
4 points by jdw64 2mo ago|1 comments
187.
by jdw64 2mo ago
I'm Korean, but I often think that the so-called virtues of Koreans are actually just vulgarity. Even when it comes to social stratification, they don't do it with any sophistication. it's just crude and vulgar. They could do
188.
by jdw64 2mo ago
Ah, if only I were from Helsinki, I would have signed up.
189.
by jdw64 2mo ago
To be precise, it's because of Korea's corporate structure. Korea operates in a way that favors large conglomerates. Since the media is also owned by corporations, it tends to protect corporate interests rather than amplify citize
190.
by jdw64 2mo ago
English(lang) is the problem. Korean and English have very different grammatical structures. I can read English in real time, but writing, listening, and speaking are difficult. (Maybe it'd be different if I were more proficient, but I
191.
by jdw64 2mo ago
As a Korean, the reason people rush into stocks is simple: labor value has been completely destroyed. Realistically, it's nearly impossible for high income young people in their 20s and 30s to buy a house in Seoul. That forces them to
192.
by jdw64 2mo ago
To be honest, the optimization the author talks about isn't really high level optimization. And Claude's pattern actually calls for a more extensible design. Strictly speaking, the author's instruction could be considered inc
193.
by jdw64 2mo ago
I read through a bit of it, and it seems like the author assumes training was done only on open source data, but I doubt that's actually the case. There are plenty of paid, proprietary source codes being sold out there, hacked code, th
194.
by jdw64 2mo ago
There's a statistic similar to this showing that socially successful people tend to think they're above the law. I think that once people achieve a certain level of success, they start to think, 'I can get away with this much
195.
by jdw64 2mo ago
Reading through this, the Apple side of the story is actually the more fascinating one. It's frankly bizarre that a company of that scale had to rely on a former employee's (Chang Liu) personal iCloud access. It exposes a massive
196.
by jdw64 2mo ago
I have a similar problem. I'm not sure about architecture. When I first implemented the slot system in my language, I designed it around that implementation, and in the end, the architecture suffered. There were quite a few cases where
197.
by jdw64 2mo ago
I've noticed that when only AI writes the code, there are cases where it ends up being unusable by humans. For small code snippets or function-level units, AI is often much better than me. But when I look at the whole picture, AI tends
198.
by jdw64 2mo ago
I'm curious about that too. I'm building a language with AI, and when I leave the AI running on its own, it keeps generating code in strange directions. I wonder what kind of structure AI-only code would have.
199.
by jdw64 2mo ago
But it's true that you learn a lot by explaining working code to others. That's also why I write on my homepage. Most of the code isn't entirely written by me. I bring it in from open source and other places, and as I review
200.
by jdw64 2mo ago
800,000 test??? wow So if I can use Kriad, does that mean I can build AI without having to use NVIDIA chips
201.
by jdw64 2mo ago
I got curious and searched a bit on Chinese sites, Zhihu, and Baidu. The relationship seems to be correct, but there are some complex points. I don't think it's at the point of collapse yet. It's true that third party apps li
202.
by jdw64 2mo ago
Usually, scholars don't bother formatting spreadsheets. they just need to produce statistics from the data. But for office workers, formatting Excel is another skill that gets evaluated. It's like how I put meaningless animated UI
203.
by jdw64 2mo ago
I think it's meaningless because the baseline RAM has changed. These days, the minimum is 8GB. The reality is that once you use frameworks and standard libraries, there's a point where 'standardization' leads to generali
204.
by jdw64 2mo ago
I wish Gemini's performance were as good as Claude Code's. But I deeply agree with the content of this post.
205.
by jdw64 2mo ago
Being able to explain things simply and being deeply skilled in a specific domain seem like similar but different abilities. Being deeply skilled in a domain means understanding its specific terminology and the dominant modeling approaches
206.
by jdw64 2mo ago
On one hand, there are posts saying LLMs are terrible. On the other, there are reports that LLMs are solving century-old problems. I'm not sure where to find the balance. Seeing this, it seems like it all depends on how you use them
207.
LLM Writing and Me(makonea.com)
1 points by jdw64 2mo ago|0 comments
208.
by jdw64 2mo ago
I've had opportunities to look at code from Korean and Chinese companies, not just academia. From what I've seen, AI generated code is often better than the average code in those codebases. When I talk about PhD level code, I&#x27
209.
by jdw64 2mo ago
I see it 'very often.' Most research paper code has no reusability—they just implement the algorithm. So I can see why it might be considered low-quality. They don't use things like Result or policy types like we do in produc
210.
by jdw64 2mo ago
I'm not really sure. I'm mostly thrown into existing codebases and just modify things to fit the existing style, so I'm almost always evaluated purely by metrics. I do think your (TheOtherHobbes) standards are really good, th

More ›