Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
maleldil
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
3 ms
·
1.
▲
by
maleldil
5d ago
The character limit thing is real. Another problem is that even if you manage to address all of their points, they're not required to engage with that. They can just ask the LLM to write a basic response back and say they won't ch
2.
▲
by
maleldil
7d ago
This technique is so obvious to anyone who spends more than a minute with multiple choice tasks. It's wild they're claiming it as a feature.
3.
▲
by
maleldil
7d ago
This already happens. It's clear when a reviewer used an LLM, and it's very annoying for the authors that have to respond to what are usually low quality, superficial reviews.
4.
▲
by
maleldil
7d ago
There is. NLP conferences, like the ACL family (and the ARR) require you to disclose in the paper if you used LLMs for writing and coding, which ones and how. Whether every author is honest is a different matter.
5.
▲
by
maleldil
7d ago
Comptime is a big one too. You can achieve similar things in Rust with generics and macros, but comptime makes certain things easier (and other things harder).
6.
▲
by
maleldil
8d ago
I like the "strong opinions, loosely held" approach myself.
7.
▲
by
maleldil
9d ago
You don't need to use fzf. Zoxide has an interactive function too.
8.
▲
by
maleldil
17d ago
Oooh, that is very useful. Thanks for sharing!
9.
▲
by
maleldil
18d ago
Why do they show different information (temperature, rain etc.) for the same location?
10.
▲
by
maleldil
18d ago
I'm missing a good way to turn on/off plugins and settings depending on the directory. Eg turning off rust-analyzer on some projects, or using pyright instead of mypy as the LSP.
11.
▲
by
maleldil
19d ago
You have to indent the text by two or more spaces to count as code. Hacker News doesn't use Markdown, so backticks don't work. FWIW, this is clear in the "help" link by the textarea.
12.
▲
by
maleldil
20d ago
No, you don't. There have been technological advances beyond just "more random Internet text", and they lead to more powerful models with more refined emergent behaviours.
13.
▲
by
maleldil
21d ago
I don't see how this is proof, and not just that the model got the better. You're comparing models a year and a half apart; this is a lifetime in LLM development.
14.
▲
by
maleldil
23d ago
You can still have extensive "compile-time" checking with linters and static type checkers. This makes Polars even better in typed Python, as it type-checks much better than the dynamic mess that is Pandas.
15.
▲
by
maleldil
26d ago
It's quite an effective strategy, sadly. People look at Cuba and see a failed state without considering the material conditions.
16.
▲
by
maleldil
26d ago
Why would a colonialist power ever be benevolent towards its colonies?
17.
▲
by
maleldil
27d ago
WSL2 is also a virtual machine.
18.
▲
by
maleldil
27d ago
The UK is banning the sale of tobacco to people born on or after 1 January 2009 entirely. That will eventually kill the industry (which is good).
19.
▲
by
maleldil
29d ago
No mention to Mistral?
20.
▲
by
maleldil
1mo ago
It's not the same as running on an x86 Mac, though. Asahi's efforts are herculean, but it will always be an uphill battle. Many things are still unsupported, even on M1.
21.
▲
by
maleldil
1mo ago
You're definitely right about that. I couldn't tell you where I found out about this keybind.
22.
▲
by
maleldil
1mo ago
I don't see how these are incompatible. The PhD prepares you to be an independent researcher. The way you prove this is by studying the same topics for a few years, and by the time you reach your viva, you'd be expected to know a
23.
▲
by
maleldil
1mo ago
There's no pretense. That's always been the whole point.
24.
▲
by
maleldil
1mo ago
In an ideal world, sure. In reality, a PhD[1] is about publishing papers in the hope that you learn something along the way. [1] My experience in ML in 2026.
25.
▲
by
maleldil
1mo ago
I took GP as asking for evidence that the reduced-cost Sol is actually a distillation of the previous-cost Sol. AFAIK, providers distilling or quantising models and offering them as the same model have not been proven.
26.
▲
by
maleldil
1mo ago
> not "23 lines pasted" If you mean what Claude Code does when you paste longer text, you can just press Cmd-V again to show the full text.
27.
▲
by
maleldil
1mo ago
That's not the same as smart pointers like std::unique_ptr.
28.
▲
by
maleldil
1mo ago
> Moving ownership between containers to avoid allocations was possible That's not the biggest upside of move semantics. You can't safely express smart pointers without move semantics. They tried with std::auto_ptr, and it didn
29.
▲
by
maleldil
1mo ago
By "web features" you mean things Google decided they wanted and neglected to put through the actual process. This is why Firefox is important; the alternative is letting Google control the Web like Microsoft did.
30.
▲
by
maleldil
1mo ago
Yeah, publishing in ML can be annoying because of that. What's wrong with "I thought this problem was interesting"?
More ›