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

It is perhaps a blessing that the climate of the Papua lowlands makes the potato impossible to grow. While the potato would grow in the cooler highlands, this village to village transmission mechanism would not get the potato past the hot lowlands where plants would rot in the ground. The potato yields a lot more calories per hectare than the sweet potato, and when introduced to New Zealand, fueled massive wars.

Nominative determinism- first author's name "Mutlu Yıldız" means "happy star"

There's a well-known issue of "construct validity" in neuroscience.


Once a horse gets a taste for baby chicks, it's hard to convince them they aren't omnivorous.


When cows have phosphorus deficiency, they occasionally eat small birds, rodents, insects, or bones


It's not uncommon for cows to get a taste for chicken eggs and start seeking them out


Is this just based on that one video or is there something more to this


Its probably true of most animals. I raised a couple of pigs and we intentionally never fed them any meat, even though they'd probably eat it, because we didnt want them to get a taste for it


For sure they'd eat it - pigs have a taste for everything.

You don't want to be the farmer who falls over in his pig sty.


> You don't want to be the farmer who falls over in his pig sty.

This is the real reason for the panic when Dorothy falls into the pig sty at the beginning of the Wizard of Oz. The pigs would crush her and eat her. (But in the movie, her dress doesn't even get dirty!)

https://www.youtube.com/watch?v=-r6cF5lyqEo


Finally a time when this obscure Wikipedia article is relevant:

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


Apparently it's still a thing in places like rural India for people to squat to poop out in the fields, and then risk being bowled over by a pig rushing over to eat it.

They truly will eat anything.

As far as meat goes, they are not just scavengers - I've seen videos of wild pigs chasing after rabbits, actively hunting them down.

I'd never head of those Chinese "pig toilets", but it reminds me of a Chinese friend who said that growing up (maybe still?) it was common practice in the villages to feed infant poop to the pet/village dogs.


Yup. Lots of stories of old timers dying of a heart attack in the pen and only finding glasses and watches left.


Same here, first time I had balut in Vietnam it unlocked some kind of primal level in me I didn’t know I had.


Horses are clearly and very much into the whole cowboy thing, they LIKE chasing down cows and catching them, some do it all by themseves and catch cows by grabbing there tails and hauling them around that way. Horses are panoptic motherfuckers and are startlingly good stalkers, which I learned living in close quarters with a horse needing to chase them away while doing stuff in the padock, but have them reapear, it took very precise sneeky peeking to catch them moving towards me, as they only move when I am not looking, which led me to the only viable deterance, which is to pretend that they are invisible/of zero importance which un nerves them, but so much as one eyelash of a look at them is taken as an invitation to partisipate, fuckers. Horses also work well with dogs, and for arabians in Arabia, dates and dried fish were used as feed for both. note, live with an arab/percheron cross far enough out that there is a wide representation of wild critters, and most everything is on most everythings menue, depending. And one of my more radicle, radicle friends would say, "we will eat the vegans first, they are slow and weak", refering to what to do durring an opocaliptic event, and until they could get out to my place in the country.


> And one of my more radicle, radicle friends would say, "we will eat the vegans first, they are slow and weak"...

A radicle is an early plant growth stage. No wonder they would hate the vegans most!


> Well they have to destroy them after scanning, else it is a copyright violation.

I think this is the "an undercover cop has to tell you they are police if you ask them" of AI law


So push back and explain how that user is wrong?


I wouldn't characterise it as 'undercover cop...' but perhaps it might be fairer to say that until it's tested in court, no-one knows, so they (Amazon etc) are taking the 'safe' path of destroying the books?


Well, trying anything in court is definitely what they want to avoid. I mean, what is your suggestion here - should they buy the book, scan it, then sell or give it to someone else? That IS a copyright violation in itself, no need to try anything in court.

Whether or not AI training or replication of the text through the model is a copyright violation, is a whole different but unrelated question.


I don’t think it’s a sure thing that training a model on a work is considered a duplication in the sense that ripping a CD might be for example - unless I’ve missed something, that’s what hasn’t been decided in court. I think they see destroying the book as the safest course of action for them.


But they do not delete the scan after training, they keep all the training data for future use and future training.


Yes, and I imagine that’s why - in the minds of their lawyers - doing anything with the book (other than destroying it) is arguably less safe than destroying it


> The water thing

It’s amazing how Karen Hao managed to graduate from MIT without learning the difference between m^3 and L


Maybe she just liked giving people mysteries


Would this imply that the matrices should have dimensions of multiples of 40?


About 15% of the UK's land parcels are not in any land registry database, because recording there only became compulsory in 1990, so if an event like a sale/morgtage didn't happen to a property since then, it might not be registered. Regardless of the land registry, the ownership can still easily be proven.


Clearly they are batching reasoning inference in a few multiples of 512 tokens as a throughput optimization


My first thought would be an adjustment to a reasoning budget parameter (using llama.cpp as my reference) which would lead to these results. But no way to know precisely without an OpenAI statement.

It could be a very dishonest way of scaling to demand during peak hours. I know that some people already scoff in this topic about the subjective nature of perceived performance of models. But the model seemed less smart when US comes online (at least from my testing over the month of May).

On my company blog post from a few weeks ago I felt the need to point this out because it had a perceptively more consistent pattern during those overlap times. Should have saved the session logs for further analysis https://webesque.agency/blog/2026-06-19-llms.html


Isn't the standard to use continuous batching? If they are using continuous batching -- I'm curious why generated token length matters, and why they might be clustering them. If not -- I'm curious why they aren't and what is the tradeoff here.


This "~512 batching" makes me think of things like diffusion or prefill.

If they managed to put together some dirty hack that lets them generate about 512 tokens worth of reasoning in parallel instead of in sequence? That would explain it.


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

Search: