Hacker Newsnew | past | comments | ask | show | jobs | submit | zjp's commentslogin

No there's an alternative to countersignalling LLMs by pretending to be dumb. Since LLMs reproduce all of the worst of English modernity, overrelying on the PMC mush register, the business casual of English, with all the agreed upon casualisms that have rotted out the language like "there's" before plurals and “because noun”, there's an easy way to ensure no one mistakes your writing for an LLM's: use the constructions of the formal register or early modern English.

Roll thine eyes all thou wish’t but this I promise thee, if thou art lucky there will come a day on which thou shalt speak the single most important sentence in thy life, and that sentence will contain the word “thee”: “With this ring I thee wed”.

Feed that into a slop detector.


At least detect my country by IP range and put "US" at the top of the country dropdown.

Always a great story. It's worrying that people are starting to turn to LLMs for indeterminate problems like interpersonal and relationship advice.

A big difference between the whispering earring and LLMs is that ostensibly the whispering earring has an actual model of the people around you that it can draw on to influence your decision making, whereas LLMs only know you.

So it's a bit like "talking" to a Cosmopolitan article.


Imagine thinking 47 degrees in July is a bad thing? Do you like, like getting summer swampass in New York or wherever you live?

I want to be spared, so I am.

My project depends on over 150 external packages. We just do integration testing or manual testing. Update your dependency, check the feature that depends on it.

It depends on whether you mean programming (typing your solution into your text editor) or programming (formalizing your solution to a problem in a way that can be programmed).

I think it's better to say 'the artifact is not the product'. Code is actually very hard, because it is the act of formalizing vague problems. That's thinking. It can be separated from literally typing the code in, and some problems are solvable in so many isomorphic ways you might as well describe them to an LLM and let it pick one out of a hat (embarrassingly solved problems), but it's like saying "prose is the easy part".

Yes. Agents are good at solving densely represented (embarrassingly solved) problems, and a surprising and disturbing number of problems we have are, at least at the decomposed level, well represented. They can even compose them in new ways. But for the same reason they would be unable to derive general relativity, they cannot use insight to reformulate problems. I base this statement on my experience trying to get them to implement Flying Edges, a parallel isosurface extraction algorithm. It’s a reformulation of marching cubes, a serial algorithm that works over voxels, that works over edges instead. If they’re not shown known good code, models will try and implement marching cubes superficially shaped like flying edges.

You are still necessary to push the frontier forward. Though, given the way some models will catch themselves making a conceptual error and correct in real time, we should be nervous.


I've had the same experience. I do a lot of automation of two engineering software packages through python and java APIs which are not terribly well documented and existing discussion of them on the greater web is practically nonexistent.

They are completely, 100% useless, no matter what I do. Add on another layer of abstraction like "give me a function to calculate <engineering value>" and they get even worse. I had a small amount of luck getting it to refactor some really terrible code I wrote while under the gun, but they made tons of errors I had to go back and fix. Luckily I had a pretty comprehensive test suite by that point and finding the mistakes wasn't too hard.

(I've tried all of the "just point them at the documentation" replies I'm sure are coming. It doesn't help)


"English being my second language, I curse it everyday and wish it could be more like, say, Hungarian, in which such a thing as a spelling bee would be unthinkable."

I love this about English! We are the most prolific word thieves of all time. We even stole an entire grammatically complete sentence from French ("Je ne sais quoi").

If you want English to be more like Hungarian, start inserting Hungarian words into sentences otherwise written in English and I guarantee people will adopt them as loanwords in short order. Never define them, we'll figure it out from context and vibes, and we'll never pronounce them correctly, which might make it grating to listen to them spoken back to you. But you can absolutely just incept words into English. We'll take them. We're hoarders. We all love that shit.

My favorite thing about it is the register system that developed from all this theft. There are at least three: German, French, and Latin. German is less formal, and French and Latin are often equal but differ in that French is less bureaucratic than Latin. The start, commencement, and initiation of something are different. And an initiation is different from an inauguration. You ask your friend, question a witness, and interrogate a suspect. Greek is more abstract than Latin. A moral question is nearer to the heart than an ethical question. You diagnose a disease, you judge a person. You have compassion, you merely feel sympathy.

Though, I would hate to learn it as a second language for the exact same reasons.


> If you want English to be more like Hungarian, start inserting Hungarian words into sentences otherwise written in English

I've been doing something like this with Finnish (which is in the same language family as Hungary) - I use Finnish colloquialism but directly translated into English. Things like "going ass first up a tree" (meaning doing something in a sub-optimal way) or "better on the ground than in the devil's mouth" (when you spill something). I find it amusing.

The author is right though, the English language is dreadful; In Finnish the words are written and pronounced the same way. Try that with some names of cities or towns in England.


Going ass first up a tree is funny enough to catch on if you keep using it. It fills a real semantic gap in the idea space of taking great pains to do something the wrong way. I'll never forget reading it just now.

Come on now though, dreadful. There's something beautiful about a language that's a fusion reactor for all other languages on Earth.


A lot of expressions in English started out as calques, outputs that process: you're paving the way!


> "Try that with some names of cities or towns in England."

Are you suggesting it's not intuitively obvious that the town 'Towcester' should be pronounced the same as a "toaster" for toasting bread?


I think a lot of what the author has an issue with is related to the Great Vowel Shift, not necessarily loanwords.


> English being my second language, I curse it everyday

"every day"

Perhaps the author should check grammar while he checks spelling :D (Not the only issue I noticed, and I didn't read the entire article...)


> I love this about English! We are the most prolific word thieves of all time.

It's impressive. English language: ~500,000 words. German language: ~135,000 words.


Where do these numbers come from?


Multiple sources have those approximate numbers. Here's one: https://www.merriam-webster.com/help/faq-how-many-english-wo...

"Webster's Third New International Dictionary, Unabridged, together with its 1993 Addenda Section, includes some 470,000 entries. The Oxford English Dictionary, Second Edition, reports that it includes a similar number."


The way English uses this vast vocabulary is beautiful.


the zeitgeist hungers for loanwords


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: