Hacker News: internet_points

New comment by internet_points in "A Privacy Analysis of Web and Mobile Conversational AI Agents [pdf]"

internet_points — Tue, 29 Sep 2026 13:12:32 +0000

hyperleaking

New comment by internet_points in "Neal Stephenson responds with wit and humor (2004)"

internet_points — Tue, 29 Sep 2026 09:57:59 +0000

Huh, I hated Snow Crash, but loved Cryptonomicon. Haven't read any of his other fiction.

I liked this interview, very nuanced and well-reasoned answers.

New comment by internet_points in "Neal Stephenson responds with wit and humor (2004)"

internet_points — Mon, 28 Sep 2026 21:35:38 +0000

https://federate.social/@mattblaze

New comment by internet_points in "When did Google get so weird?"

internet_points — Mon, 28 Sep 2026 09:28:33 +0000

omg thank you! I would not have guessed they included an option to stick with "classic", that's actually quite decent. This made my day, thank you again, random internet stranger.

New comment by internet_points in "Replacing the old battery on rechargeable bike lights"

internet_points — Mon, 28 Sep 2026 07:47:14 +0000

I would say $15 is cheap for the learning, growth and experience. Interacted with humans too along the way. People pay much more than this for schooling which teaches you less.

New comment by internet_points in "When did Google get so weird?"

internet_points — Mon, 28 Sep 2026 07:37:15 +0000

I miss the old Google Translate. I use it for learning languages (translating little snippets, lyrics etc.) and these days it "interprets" way too much, confabulating like an LLM and inserting stuff that's way beyond what's in the text. I have to remove words and rewrite and look for other sources to be sure it's actually translating what I'm inputting and not something it just made up to make it sound good.

New comment by internet_points in "Opus 5.5 is good at explainer videos"

internet_points — Sat, 26 Sep 2026 06:32:43 +0000

> a website without a message is meaningless more or less.

meh, I don't care about the message of your business, I just want the ADDRESS and OPENING HOURS. Why is that so hard to put up front and center and instead I've got to scroll past videos and hero images and chatbots and dickovers and embedded interactive maps without the actual address showing argh

sorry that website was probably not your fault

New comment by internet_points in "Writing Parquet files using Haskell"

internet_points — Fri, 25 Sep 2026 13:55:14 +0000

Ooh, you're saying you need to ask it twice to get the same real-world result. I thought you meant the tokenization was less efficient (as if `Endofunctor` were tokenized `Endo`+`Functor` while `AbstractFactory` was a single token or something like that)

New comment by internet_points in "Writing Parquet files using Haskell"

internet_points — Fri, 25 Sep 2026 09:05:36 +0000

there are a lot of strong claims here, but to just pick one:

> inefficient token wise to use functional languages

where does that idea come from?

New comment by internet_points in "Writing Parquet files using Haskell"

internet_points — Fri, 25 Sep 2026 09:01:39 +0000

not the okasaki of Purely Functional Data Structures fame I guess?

New comment by internet_points in "Parsing JSON Objects Without Intermediate ASTs in Haskell"

internet_points — Thu, 24 Sep 2026 09:36:17 +0000

interesting!

There's a library hw-json that uses succinct data structures and seems to have about half the memory usage of Aeson: https://github.com/haskell-works/hw-json#parsing-large-json-... How does your approach compare?

AI video won Nikon Image Competition, microscope user needs to talk about it

internet_points — Wed, 23 Sep 2026 13:57:39 +0000

Article URL: https://reddit.com/r/labrats/comments/1wo0j5e/an_aigenerated_video_won_this_year_nikon_image/

Comments URL: https://news.ycombinator.com/item?id=49816278

Points: 2

# Comments: 0

New comment by internet_points in "Small programming tricks"

internet_points — Thu, 17 Sep 2026 12:51:44 +0000

The code I check in at work is nearly all hand-written. I'll often send files and functions and snippets and questions and `git diff`'s to an LLM. In particular the code review is very helpful for catching silly mistakes. But if it generates code, even if it looks fine, I always end up manually moving it around and rewriting and renaming things. I need to understand the code I maintain, and those "prosaic" changes and refactorings that I do while incorporating the code help ground me deeply in how it works.

Also, I'm mostly not very happy with the architectural choices the LLM makes. It can be good at little details, and sometimes I learn about language features or idioms or libraries by reading its code, but I find it often uses wrong patterns. E.g. just this week I saw it creating two huge intermediate data structures for parsing something in a "cleaner" way, completely missing how the library it had suggested actually had something builtin (and prominently featured in the tutorial) that obviated the need for those intermediate structures and was both safer and more readable.

When I've tried full agent coding, I kept getting distracted while waiting. I also have the feeling that my questions to the LLM were getting "dumber" as I lost contact with the code. And I don't have the hardware for local LLM's, so giving full repo access is out of the question for work.

Altogether I feel like I've found a fairly good balance in how to use them in such a way that I can avoid a lot of the downsides and still have a tool that gives me much value that I didn't have before: I check in fewer bugs, I waste less time picking libraries/methods, I have a better rubber duck, I learn little tricks all the time, I never send LLM-generated text to humans.

New comment by internet_points in "Small programming tricks"

internet_points — Thu, 17 Sep 2026 12:24:01 +0000

I tried fd but it seems to ignore so much that it's basically useless. If I wanted `git ls-files|grep` I would `git ls-files|grep`

yells at cloud

..it's probably ok actually and I should give it a real go, but I'm too used to find and even have a keybinding in my ~/.inputrc to insert a find + where loop

    "\e\C-f": "find . -type f -print0|while read -rd '' f; do ; done\e-b\C-b\C-b\C-b\C-b\C-b\C-b"

(yes I know about -exec; I find the loop easier to work with)

New comment by internet_points in "The Return of Sail Power: Cargo Ships Are Turning Back to the Wind"

internet_points — Thu, 17 Sep 2026 07:30:46 +0000

Ooh what's your favourite sea shanty, or chanty gang?

New comment by internet_points in "The Return of Sail Power: Cargo Ships Are Turning Back to the Wind"

internet_points — Thu, 17 Sep 2026 07:23:45 +0000

s/start/buy

ftfy

New comment by internet_points in "Anecdotally, programmers dislike "reduce""

internet_points — Thu, 17 Sep 2026 07:21:29 +0000

https://news.ycombinator.com/item?id=49736033 may help :)

New comment by internet_points in "Anecdotally, programmers dislike "reduce""

internet_points — Thu, 17 Sep 2026 07:18:07 +0000

That's what I tend to do, but since foldr/foldl' is so ubiquitous in Haskell it would be nice if I could just remember the argument order of the callback. kccqzy's explanation (in particular "it replaces the comma") might just help me do that :)

New comment by internet_points in "Monsanto's Cruel, and Dangerous, Monopolization on American Farming (2008)"

internet_points — Thu, 17 Sep 2026 07:13:32 +0000

Speaking of poison, does washing fruit and vegs help? Or is it like the https://en.wikipedia.org/wiki/Five-second_rule

New comment by internet_points in "I resigned from Anthropic today"

internet_points — Wed, 09 Sep 2026 13:33:54 +0000

Oh, I agree. I don't think LLMs will end civilization. I'm guessing climate change will cause many more deaths than LLMs. But I think LLMs will cause some too. (And not in one of those impressively smart ways as it's presented from the OpenAI fearmongers, just in a stupid "we vibe-coded the sandbox for our bruteforce SCADA-hacking agent experiment and oh no it did what we told it to do" way.)