This is puzzling me:

• Why were victims gathered from multiple settlements?
• Why is there such a sex disparity; or, what happened to the men?
• Why were they buried with food and belongings, if the deaths were all violent and some victims show signs of trying to run away or fight back?

I couldn’t find a single good explanation for all three things. Specially the last one, it seems contradictory.

Fun to see this while messing with subtitle translation!

I picked the three seasons of Nights with a Cat to watch. Thought “my mum would love this”, so I’m translating subtitles into Portuguese. (She loved the first two seasons, by the way.) And wow, it made me see the effort the group (pspspsps) poured into it — replicating the cat and Moon logo, animating everything, a thousand gradients, so goes on. I had to sacrifice a wee bit of the aesthetics (not much though) for the sake of making the translation viable, plus because I was merging the files, but it was clear the group did that as a work of love, I’ve seen plenty professional translators (incl. myself) doing a sloppier job.

My hypothesis is that, if you look hard enough, you’ll find almost all systems necessary for Language in other animals. Specially basic communication and basic logic. With human-exclusive development were mostly “wiring” them together into Language.

I’ll give you guys two examples, both with bonobos.

They seem to have simple syntax. They’re able for example to vocalise “A” then “B” to communicate something that is both “A” and “B” at the same time; e.g. “pay attention to me” + "I’m excited/“worried” = “come help me quick!”.

And at least one of them was able to play pretend. That implies rather strong ability to abstract and handle logic; something like “object exists, object is not here, but I’ll pretend the object is here because it’s fun”.

But you don’t see them combining both things together. Their vocalisations are mostly practical; and they can’t abstract the process of conjoining vocalisations, that would allow recursiveness. That recursiveness would allow also smaller (in both time and meaning) vocalisation units, since it they could be used together.

If that hypothesis is correct it solves a big chicken-and-egg problem we currently have: if Language appeared first for communication, then who was the first Language-able human communicating with? The answer is that they could still communicate with other humans, because even if they didn’t have Language, they’d already have non-linguistic systems that are superficially similar enough. All the advantages associated with Language would appear gradually, as more people are able to use it.

And of course, HN comments gotta make me cringe and facepalm.

When the topic is large models, those bloody HN muppets go out of their way to cherry pick definitions of intelligence so they can sealion their idea that those systems are “intelligent”. But when it comes to Language development, suddenly fine-grained distinctions don’t matter, right? Naaah, then screw definitions. Let’s conflate “communication” with “language”, what could go wrong? And if someone points out the difference, disregard it, with the same disgusting stubbornness of a dog insistently eating their own vomit!

I won’t quote the comment chain, but it’s rather easy to find for anyone with masochistic tendencies. Instead I’ll address something here.

“Language” is not simply “communication”. Language has a specific structure, something like, where small and contrastive elements are combined into higher order elements, recursively. For example:

• articulations → phonemes → morphemes (spoken)
• gestual movement → gestemes → morphemes (sign)
• morphemes → words → phrases → utterances (both)

Like playing with legos but using the blocks to build even bigger blocks, you know.

You see this structure in spoken languages, sign languages, even when writing. But it’s pretty much exclusive to human beings. The nearest of that we see among other animals are composed vocalisations in chimps and bonobos, but even then it’s quite a few steps sort. (They can combine A+B, but not A+B+C+… indefinitely, or use the set as an element for a new combination.)

And humans don’t even use Language all the time to communicate, dammit. Non-linguistic communication is a thing, and extremely common. Rolling your eyes to convey “this makes me cringe”, that “mmmh” to convey “I’m listening, go on”, raising your hands to convey “that’s it!”, those aren’t language, but they’re things humans do. And they resemble rather closely non-human communication, by the way.

Oh noes! Anyway… (spamterest is the site that prompted me to install uBlacklist. It’s mostly garbage but search engine optimised as fuck, so it’s highly visible garbage.)

Odds are they used whatever material they had at hand, but IIRC the most common ones were bones and antlers.

Time to be[an] yourself!

Link to the paper, and excerpt from the conclusion:

In our study, we showed that baby domestic chickens (Gallus gallus) respond to sound-symbolic associations similar to humans, preferentially matching the sounds ‘Bouba’ and ‘Kiki’ to a round and a spiky shape, respectively.

This could be convergent development between mammals and aves, but honestly I think it’s inherited. (I wonder if lizards show evidence of the effect? Some can be quite smart, like green iguanas.)

If inherited then we’ve been subjected to the bouba-kiki effect for more than 300M years — kind of funny how something so old would have an impact on Language later on, for some weird clade of hairless apes.

“Now that we don’t do that, you see these things on the internet where, ‘Don’t use ChatGPT, it’s 17 gallons of water for each query’ or whatever,” Altman said. “This is completely untrue, totally insane, no connection to reality.”

He knows he’s a con artist, he knows people know he’s a con artist, and yet he’s talking as if we were supposed to trust him to not be a con artist. That’s basically to call everyone stupid/gullible/trash by proxy.

He added that it’s “fair” to be concerned about “the energy consumption — not per query, but in total, because the world is now using so much AI.” In his view, this means the world needs to “move towards nuclear or wind and solar very quickly.”

Even before those huge datacentres, “don’t reduce consumption, increase production” is how we’re cooking the planet.

There’s no legal requirement for tech companies to disclose how much energy and water they use,

That’s something that could be fixed. At least in Europe, China, Japan; probably here in Latin America, too.

Altman also complained that many discussions about ChatGPT’s energy usage are “unfair,” especially when they focus on “how much energy it takes to train an AI model, relative to how much it costs a human to do one inference query.”

Whataboutism at its grossest.

The coin evidence dates the sanctuary to between about 80 and 171 AD, during the period of the Roman cohort fort in Kumpfmühl and the associated Danube settlement, before the establishment of the legionary camp at Regensburg.

This is interesting because it shows how widespread the “mystery cults” (like Mithraism) were back then.

Originally the Romans built a small fort in the place, near the Celtic settlement of Radasbona. But then by 171 Marcus Aurelius had it rebuilt to host the Italic Third Legion. And given legions back then had 5200 soldiers, this means the fort was considerably smaller than the necessary to hold 5k people; if it was just a bit smaller, they’d extend, not rebuild it.

For reference: in the 1st century it’s believed the city of Rome had ~1M inhabitants, and Alexandria had ~500k. The empire as a whole had, like, 60M? 75M? inhabitants. So even for the standards of back then, this sanctuary was found in the middle of nowhere, and yet there was social pressure to build a shrine for Mithras there.

the sanctuary provides valuable new evidence for the rituals and material culture of the enigmatic Mithras cult across the Roman world.

That’s important because we know practically nothing about the cult. The initiates swore an oath of secrecy, so written info from those times is rather scarce.

Who knows? Definitively worth trying! (…I’m not trying it.)

indications.

Yes but the text omits what is being indicated: directions. In Spanish I bet this is clear by context, but since I was trying to parse it from other Romance languages, the result was weird:

• IT - so Seymour got some instructions? (“Indicazioni” sounds way more abstract, as “indicating what you need to do”)
• PT - so they were there despite the recommendations? Then why is bald guy looking at Seymour with a “you dun goofed” face? (“Indicações” would imply “indicating a place as worth visiting”)

Just small language differences, you know.

“it’s what he did”/“it was her heart what he stole” - works like this.

Is it common to say that “que” in Spanish without any verb, like they did there?

I’ve seen people preparing sweet cakes with Parmesan cheese. I’ve eaten it. It’s actually really tasty, so… I hope the person who took this pic tasted their tea/chocolate/whatever before throwing it away.

And anyone can point out a thousand examples arguing the opposite: that interfaces are becoming “worse”, downgraded. Or argue that neither is true, interfaces are the same quality-wise, and all we saw was “sidegrades”.

It’s still treating what’s a subjective matter of features as if it was an objective matter of age. It’s all an “it depends”; you can’t simply assume “newer thus better lol lmao”, as the usage of “dated” implies.

At most what a designer can do is to say “doing things this new way will improve usability for $cohort1 but decrease it for $cohort2. Since $cohort2 is not part of our core target audience anyway, so let’s implement it”.

(I was going to include a bunch of examples but I feel like they’d make people miss the point.)

I can kind of parse Spanish from other Romance languages, but some stuff throws me off track. Like that indicationes for “directions” (I had to check it in a dictionary), or the verbless que’s in the third panel.

But yeah, the joke is really clear.

My issue here is specifically with the usage of the word “dated”.

Typically, when people use this word to refer to interface, they don’t highlight the differences in features or trends; and they don’ty take into account that different users might do different things, and thus require different designs. The whole thing gets oversimplified as “its newer so gooder lol lmao”, or as if the old design had an expiration date.

Okay. Gotta explain this shit because I’m in the mood to do so. And because if I get something wrong people will point it out. (Implied: take what I say with a grain of salt, it’s potentially inaccurate. I’m mostly commenting for funzies, so caveat lector.)

Have you seen a chatbot? One using generative models, like ChatGPT and the likes? They’re unreliable, and prone to output bullshit¹, have a meltdown if asked about the seahorse emoji, or if you ask them about a tag with a reversed slash². (Did you eat a rock today?

Those generative models typically output media (text, images, audio) and call it a day, so they’re pretty contained; not exactly safe because they can generate misinformation, but still. But what if you allowed them, for example, to interact with files? Create file, delete file, edit file? Then you have what people call nowadays an “AI agent”, like Claude Code. And given the “AI agent” is built upon a generative model, and a generative model is unreliable, guess what: once allowed, it will misbehave. For example, wiping out the contents of your hard drive. Oopsie.

But let’s go further. What if that “AI agent” runs on its own, without being prompted to do so? Then you got a claw, like "Open"Claw. (no, it is not open source. Just tech bros misleading you, as usual.) It’s actively trying¹ to do things. You build an unreliable system, make its unreliability matter more than just misinforming people, and then you let act unsupervised. What could go wrong?

Perhaps it wipes the hard drive because it concluded¹ it had to restart afresh. (Play stupid games, win stupid prizes.) Or its behaviour actually affects other people; the claw" could try¹ to contribute with an open source project, gets its code change request denied, and writes a hit piece on a developer. Or perhaps the merge is actually accepted, and the software in question affects a lot of people, and since it’s now misbehaving the life of those people are worse. (Or gone, if the software handles something serious.)

Do I need to stress how much of a bad idea this is?

Notes:

1 - “Bullshit”, “try”, and “conclude” are being used metaphorically here.

This reminds me hobo code, sets of symbols allegedly used by some types of travellers (mostly hobos and beggars) to tell their peers “if you ask food here they’ll give you some”, or “don’t come close, danger”. I say “allegedly” because I’m not sure if those systems are actually used, or if someone made this up.

Perhaps this is a pre-writing take of the same idea? As in: “good place to rest”, “be careful of the local fauna”, etc.

“Natiform” (butt-shaped), “trifurcation” and “boustrophedic”?

[HN comment excerpt] LibreOffice has a great mission, but their UI feels dated

Emphasis mine. I think people should stop treating interfaces like bread, as if they were going to get stale and mouldy over time.

Relevant to note I don’t speak Nahuatl. I parsed this info from Wiktionary + Wikipedia, it’s surprisingly easy to follow.

(For the non-possessed form, as in “a penis is an organ”, use “tepolli” instead. Wiktionary also mentions “tototl” bird being used with that meaning, kind of like English “cock”.)