Hacker News: julian37

New comment by julian37 in "Claude Code is locking people out for hours"

julian37 — Tue, 07 Apr 2026 17:21:29 +0000

Use planning+execution rather than one-shotting, it'll let you push back on stuff like this. I recommend brainstorming everything with https://github.com/obra/superpowers, at least to start with.

Then work on making sure the LLM has all the info it needs. In this example it sounds like perhaps your hypothetical data model would need to be better typed and/or documented.

But yeah as of today it won't pick up on smells as you do, at least not without extra skills/prompting. You'll find that comforting or annoying depending on where you stand...

New comment by julian37 in "Show HN: Tetris in a PDF"

julian37 — Fri, 10 Jan 2025 16:52:50 +0000

Works fine here (134.0 on macOS)

New comment by julian37 in "OrbStack: The fast, light, and easy way to run Docker containers and Linux"

julian37 — Mon, 02 Sep 2024 14:55:28 +0000

kdrag0n's first post about this on HN, afaict: https://news.ycombinator.com/item?id=34100779

Amazing how far they've got since, in just two years. As others have pointed out, it's already "boring" software in that it just works. And that's no small feat because this kind of tool requires all kinds of low-level hackery to make work, and make work fast. Hats off!

(Happy user here if you couldn't tell.)

New comment by julian37 in "Docker Performance Evaluation Across Operating Systems"

julian37 — Tue, 27 Aug 2024 20:22:49 +0000

You mean what Docker for Desktop can't do? It does have a few unique features (for example automatic HTTPS for containers) but that's not the point. The point is that it's much faster, the fan runs quieter, the battery drains less, memory gets hogged less, it crashes less, the UI is smoother, the updates actually work, and it keeps improving at a steady pace. The fundamentals just work a lot better.

New comment by julian37 in "Docker Performance Evaluation Across Operating Systems"

julian37 — Tue, 27 Aug 2024 20:08:04 +0000

OrbStack is where it's at on macOS these days. It's so much better than Docker Desktop, it's not even funny. It's in a whole different league.

I'd take this article with a grain of salt when it doesn't even mention it, or any of the other alternatives.

New comment by julian37 in "After 6 years, I'm over GraphQL"

julian37 — Thu, 30 May 2024 10:25:32 +0000

This. For graphql-ruby: https://graphql-ruby.org/operation_store/overview

New comment by julian37 in "Applesauce pouches may have been contaminated on purpose, FDA foods chief says"

julian37 — Fri, 15 Dec 2023 16:11:26 +0000

I don't how hard it is, but part of the problem is the FDA's chronic understaffing issues:

https://www.washingtonpost.com/opinions/2023/01/10/baby-form...

https://abcnews.go.com/Business/story?id=3944507&page=1 (2007)

New comment by julian37 in "What if an SQL statement returned a database?"

julian37 — Mon, 11 Dec 2023 23:11:42 +0000

Right, but fetching rows along with a bunch of related rows from other tables isn't niche, it's literally the _raison d'etre_ of GraphQL (simplifying a bit because GraphQL types don't necessarily map onto db tables.)

It's something lots of apps need all the time.

New comment by julian37 in "Minify and Gzip (2022)"

julian37 — Mon, 17 Jul 2023 11:01:53 +0000

Closure Compiler follows the same line of thinking:

https://github.com/google/closure-compiler/wiki/FAQ#closure-...

New comment by julian37 in "Why is there a drink called 手打柠檬鸭屎香 = “hand-made lemon duck-feces fragrance”?"

julian37 — Fri, 30 Jun 2023 20:00:06 +0000

Reminds me of Hand Cheese (Handkäse), a regional specialty popular around the Frankfurt area. It's delicious, quite pungent but with a deep flavor. It gets its name because it was traditionally formed by hand (from Quark, a harmless dairy product similar to cottage cheese). Around these parts one can expect clean hands as well, so there really is nothing untoward about the final product -- except perhaps for its name. Give it a try if you happen to be passing through!

New comment by julian37 in "1Password to Add Telemetry"

julian37 — Mon, 24 Apr 2023 21:17:40 +0000

Surely the UI code is what responds to clicking "reveal" and therefore, if compromised, could fetch the secret even without a click?

New comment by julian37 in "Pg_jsonschema – JSON Schema Support for Postgres"

julian37 — Wed, 22 Mar 2023 13:29:31 +0000

I've found that this PL/pgSQL implementation also works well when you don't need maximum performance:

https://github.com/gavinwahl/postgres-json-schema

New comment by julian37 in "AI found a bug in my code"

julian37 — Thu, 17 Nov 2022 12:46:39 +0000

That's hardly a convincing example. All of these points can be solved elegantly with a stream abstraction, which can be cheap or free given a sufficiently advanced language and compiler.

As for legal or policy reasons, those still aren't reasons to write boilerplate code. Your reimplementation can be tight and reuse other abstractions or include their own.

New comment by julian37 in "Overzealous Destructuring"

julian37 — Sat, 08 Oct 2022 03:17:06 +0000

All of the examples show a single use but the value of destructuring becomes more obvious when you look at an example with multiple uses. Take this variant of one example:

const Component = (props) => { funcA(props.a.deeply.nested.variable); funcB(); funcC(props.a.deeply.nested.variable); };

Depending on what the functions are doing, neither the VM nor tsc might be able to infer that variable's value stays the same or that props, props.a, props.a.deeply, or props.a.deeply.nested will stay the same for that matter.

The VM will likely have to generate code to dereference the chain all over for the second use, and the compiler might lose narrowing information. Both of these can easily be avoided with destructuring.

(You could use "const variable = props.a.deeply.nested.variable", but then you have many of the same issues the article complains about.)

New comment by julian37 in "Coping with Copilot"

julian37 — Sun, 21 Aug 2022 00:08:15 +0000

Agreed about saturation and the choice of variable name, but the code would trigger a division by zero and not result in NaN: https://go.dev/play/p/vYm4tSNEJ7M

(Also, in--say--Ruby and JavaScript 1.0/0.0 is Infinity and not NaN.)

New comment by julian37 in "Coping with Copilot"

julian37 — Sat, 20 Aug 2022 20:31:03 +0000

This is a good example of why I dread Copilot: even if Go specifically couldn't express this any more concisely, there is a language that can and Copilot's very existence makes it less likely for that other language to be used as much as it deserves.

Besides, the generated example seems to be missing code to gracefully handle the case where len(filtered) is zero. Maybe there's a precondition that prevents that from happening or maybe a division by zero is exactly what you'd want, but at face value it looks like the bot did a rush job.

New comment by julian37 in "Largest airship built in United States since 1930s to take shape soon"

julian37 — Thu, 02 Jun 2022 09:14:46 +0000

Reminds me of Cargolifter, they tried something less ambitious and failed... but that was 20 years ago, at the very height of the dot com boom. Better luck to the Pathfinder 3 team! It would be amazing to see airships revived.

https://en.wikipedia.org/wiki/Cargolifter

New comment by julian37 in "Advanced techniques for reducing Emacs startup time"

julian37 — Mon, 16 May 2022 08:17:40 +0000

Came here to say this. I'm starting Emacs only after a reboot, who cares if it takes 3s or 30s to launch (mine around 5 or so.)

Use emacsclient if you want to use it as $EDITOR.

https://emacsredux.com/blog/2014/12/23/uptime/

New comment by julian37 in "We think this cool study we found is flawed. Help us reproduce it"

julian37 — Sun, 01 May 2022 13:55:00 +0000

You work with crypto according to your profile. I hope that when you see a random generator return a series of a hundred 6s, you go and check what's wrong with it instead of assuming you just got lucky this time ;-)

New comment by julian37 in "Six companies control 90% of what you read, watch, and hear"

julian37 — Mon, 18 Apr 2022 23:40:19 +0000

If you follow the link, it says Americans spend that amount of time daily with "major media", and no mention of news only. So that would presumably include reading/watching fiction, listening to music etc.?

> included are digital (online on desktop and laptop computers, mobile nonvoice and other connected devices), TV, radio, print (offline reading only), newspapers, magazines, radio, and television

Pretty big blunder in an article about media trustworthiness if you ask me.