Live data from Hacker News

The new rules of context engineering for Claude 5 generation models

claude.com

291–300 of 434 posts

Re: The new rules of context engineering for Claude 5 generation models

#291
post #193

Earlier quoted context omitted.

A new default. That wasn't the default a few months ago. Silently deleting your user's data is so stupid on so many levels. They have no idea what they're doing.

They may also realize that these chats are super sensitive especially around ppl constantly pasting secrets into their chats

Can you imagine ransomware that targets these logs?

Re: The new rules of context engineering for Claude 5 generation models

#292
post #58

Earlier quoted context omitted.

> I guess part of it is also that I don't mind doing 'hand-edits' like for example LLMs love to say "// so and so removed" I just go and remove that manually later rather than being like "don't comment about what you removed!11" cause you're really fighting deep grooves in the model's behavior at that point. I feel like this is the way. There are surely things where it's faster; certainly it's more pleasant to do sim…

I’m ashamed to admit that I once suddenly realized what I had done: I asked the LLM to rename a variable. It took it some minutes or so. With IDE by hand it would have been seconds.

The other day I asked it to update a handful of files using a new pattern I built. It was tedious work, and probably would have taken me 10 minutes, but I figure why not just let Claude do it. Well, it spun up multiple agents, each taking a different approach (one writing a Perl script, the other just doing fancy regex, etc) and 30 minutes later had the result. Granted, it wasn't my 30 minutes it was using, but it felt wasteful nonetheless.

Now, if it's something minor, I'll ask Claude about it, and get the gist of what I need to do, and just do it myself.

Re: The new rules of context engineering for Claude 5 generation models

#293

Earlier quoted context omitted.

It's language and speech patterns that seem designed to trick readers into believing that claims are correct, even when the claims aren't based on anything and are possibly wrong. It was rewarded for this during training for some reason. Alternative theory: The LLMs only way to "think" about abstract concepts is through language, and this leaks into into conversation it has with humans. But humans generally prefer to…

"There is a depth of thought untouched by words, and deeper still a depth of formless feeling untouched by thought." - Rilke Your assertion that we don't think in language is questionable. It runs counter to the lived experience of developing thoughts through writing ("writing isn't capturing thinking -- it is thinking"). I believe there is more to thought than language alone, but I also feel quite sure that language…

> Your assertion that we don't think in language is questionable.

It's true though; we routinely see people get stuck for a word that they know but can't quite recall at that moment in time. It happens daily across billions of people, yourself included.

If we thought in language, it is impossible to be stuck for a specific word. But we all experience this at some point in our lives, hence we aren't thinking in language.

Re: The new rules of context engineering for Claude 5 generation models

#294

Earlier quoted context omitted.

It's amazing that you can predict a counterexample to an open math problem, all without thinking.

Yet they do.

"If only we had a word for that process that happens before the words come out."

"Artist Formerly Known as Thinking"

Re: The new rules of context engineering for Claude 5 generation models

#295
post #185

Earlier quoted context omitted.

the phrase load-bearing caveat makes me irate

Only that one? Not belt-and-suspenders enough

It's justa caveat is side to the main point. No human would say that. It's like a load bearing utility shed

Re: The new rules of context engineering for Claude 5 generation models

#296
post #53

Earlier quoted context omitted.

Now I have the full picture. You're right to push back, and that's on me. The load-bearing seams of language are the smoking gun I should have been aware of.

Until recently I thought "load-bearing seam" was a satirical exaggeration - I'd seen both claudisms independently but never combined. But a couple of days ago it hit me with "The key structural point first: the only load-bearing seam is [...]"

Fable has a Chomsky grammar.

Re: The new rules of context engineering for Claude 5 generation models

#298

Earlier quoted context omitted.

> We simply have better magic wands and more powerful spells now. Wouldn't it be nice though if the incantation of the same spell would always do the same thing every time ? You see that's how my old wand and spells worked.

Look at it from the perspective of a manager above individual developers. Their instructions to those devs are already a series of non-deterministic incantations. They ask for things in English and get something back in Git a week later! We've just pushed that indirection down a level from managers to ICs. The ICs are shocked and surprised that this level of imprecision is allowed. Their managers are not shocked at a…

You're applying the doorman fallacy here. This has been discussed a lot. Honestly, the sooner the industry accepts that we aren't there yet nor does the current architecture enable us do get there, the sooner we can make progress towards that ideal state.

Re: The new rules of context engineering for Claude 5 generation models

#299

Earlier quoted context omitted.

> We simply have better magic wands and more powerful spells now. Wouldn't it be nice though if the incantation of the same spell would always do the same thing every time ? You see that's how my old wand and spells worked.

One day it will.

Not with LLMs, it won't. (The probabilistic prediction property prevents proper persistent programmable predictability)

Re: The new rules of context engineering for Claude 5 generation models

#300
post #230

Earlier quoted context omitted.

When I was young, a friend asked me, "Hey, you speak three languages, which one do you think in?" I paused, confused, and replied, "People think in words?" Fast forward a decade or so, in my twenties, I had lost most of the inner visual sense I had previously used, and developed an overreliance, in my opinion, on language. (I think my dominant sense was some "non visual abstract sense of ideas", but the visual was al…

You just lack (or lacked) introspection, nothing special. Many people claim to have no inner monologue. When pressed, it always comes out. Simply put, it's impossible to function as a human without it. But many people are not aware, and think it means something like "hearing actual voices".

"Thought is language" is a folk intuition that has been proven wrong over and over. People with severe global or agrammatic aphasia still can score higher on intelligence tasks than you'd expect - even though this data generally comes from people who suffered a stroke, which generally isn't neatly contained to one's language capability. Also, more importantly, even in people with full brain function, the language network appears remarkably unresponsive under neuroimaging when performing arithmetic, logic puzzles, or theory-of-mind tasks. Also famously music tasks: Vissarion Shebalin lost most of his language capabilities to a stroke and still completed his fifth symphony.

There's correlation, and your language network likely augments your intelligence, but as proven by millions of animals, unfortunate humans and also some less-unfortunate human infants, you really don't need language for intelligence.

Where language helps most strongly is metacognition: evidence suggests that it is severely limited without language, and I suppose that is where the common belief that thought is language comes from: the moment you try to think about your thoughts, you use language!

My theory (and this with literally no evidence) is that we use the language network for metacognition precisely because it is not that involved in primary thought. Important to note that "we" here means humans: animals appear to demonstrate metacognition even without language.

Post reply on HN