Weird. I happen to be watching Malcolm in the Middle and I find a link to Malcolm in the Middle

I feel like this tweet suggests that the PDF is a polyglot or an embedded second PDF.

https://x.com/gynvael/status/2024180784064598134

Initial impressions says no about being that file a polyglot.

If you like polyglot files, see https://www.alchemistowl.org/pocorgtfo/

PoC||GTFO is the GOAT

Oh yeah. I have the paperback 'bible'. I don't think that that one is a polyglot, though.

Can’t you use the tome as a cluebat?

I believe it’s a dual use tool, hence a polyglot.

Ah, no, sorry, no polyglots there yet. We'll get there one day, but so far our tooling doesn't allow for it yt.

Ah! I thought your wording was a hint (it's the viewer that thinks it's only 92 pages).

I literally came to post the exact same line as my indicator that this was AI-generated. I ctrl-f'd first and sure enough I'm not alone in using 'key insight' as the canary.

I wish they supported Janet

Can you run Fennel in Neovim? It's a Lisp running on Lua. https://fennel-lang.org/

Yep! and I already do: https://github.com/Grazfather/dotfiles/tree/master/nvim/fnl

Yes.

I try to 'punish' for these and tip less than I would normally.

At an airport once the coffee stand had a girl behind the counter that would just hand you an empty disposable cup. She didn’t even handle cash, it was card only using a self checkout POS system. And it still asked for a tip with the default of 20%.

There should be a way to specify a negative tip.

If you really want to make a deal out of it, you hand it back to the server and ask them to select 0 for you. Mention that you don't think the service deserves a tip, and that you resent being asked.

If enough of us start doing this, it has a chance of applying some back pressure, although unfortunately that force acts through the poorly-paid checkout operator who doesn't directly have the option to influence decisions.

The server isn't the one who configured the UI, and probably needs the money. You're punishing the wrong person.

Is Jessie Frazelle still there? She is very impressive.

no, but still being super impressive. CEO of a company rebuilding a CAD rendering engine because they put an LLM on top of it. So you describe the mechanical specs of the part you want and it models it. Takes all the tedium out of modeling stuff. Super cool and many applications.

Oh cool! That looks like a super interesting product.

https://zoo.dev

They had to do CAD while working on Oxide and realized that it sucked. So she went off to solve that.

that's taking yak shaving to another level!

i tested sonnet 4.5 just last week on a zig codebase and it has to be instructed the std.ArrayList syntax every time.

I made a Zig agent skill yesterday if interested: https://github.com/rudedogg/zig-skills/

Claude getting the ArrayList API wrong every time was a major reason why

It’s AI generated but should help. I need to test and review it more (noticed it mentions async which isn’t in 0.15.x :| )

The linked blog post about making this is an excellent read.

Thanks! I think I spent as much time writing the post as I did making the skill, so I’m happy someone got some value out of it.

Fighting fire with fire

A little bit! I wrote a long blog post about how I made it, I think the strategy of having an LLM look at individual std modules one by one make it actually pretty accurate. Not perfect, but better than I expected

Try it again. This time do something different with CLAUDE.md. By the way it's happy to edit its own CLAUDE.md files (don't have an agent edit another agent's CLAUDE.md files though [0])

0: https://news.ycombinator.com/item?id=46723384

Are you using an agent? It can quickly notice the issue and fix it. Obviously if it's trained on an older version it won't know the new APIs.

What did the function that called into it do with the result? If it was a simple "if rv != 0xabcd goto fail" the patch could probably be simplified to just... nop a few bytes.

In a very strange coincidence, I happened to read that first story in a book[1] I'm reading, just last night! What are the chances?

[1] https://debuggingrules.com/

I don't get it?

    707000 = 1.46x
    x = 707000/1.46
    x = 484247

707000-464272 is more than 200k

He probably did the calculation that 707 - 200 = 507 and 200/507 = 0.39 which is different from 0.46.