News for hackers. Stuff that matters.
Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @02:02PM (#29058909)
God I cannot wait to talk to whales. Their oral histories must be incredible. People have fantasized about communicating with extraterrestrials for ages. I don't understand why we haven't invested significant resources in trying to communicate with the other animals on our own planet. What an incredibly weird and non translatable experience it will be to (finally?!) start this adventure with whales.

Tangent time. If you do a cursory search of how smart whales are, you'll get nonsense about how humans are much smarter because the size of the brain isn't relevant, its the ratio of the brain size to the body size. But somehow that argument doesn't apply to squirrels. Or to a 7 foot human vs. a 4.5 foot human. Whales probably aren't as smart as humans, but its due to the environmental pressures selecting for intelligence, not raw capability. Whales have the capability to far outstrip humans in intelligence (if you accept that neuron count and neuron connections are the raw inputs). Lets get some whale engineers working on the hard problems please.

Re:Attempting to interpret sperm whale clicks with AI, then ... by tsol on Sunday October 31, 2021 @06:09PM (#29060926)
I mean.. hate to be a party pooper, but just because we created a the equivalent of gpt3 for whales, does that mean we can do anything useful? Like talking to whales.. we haven't even established how their language works.

What language even is, is a good question. I read it once demonstrated as this; some species of monkey has a specific call they do when they see a panther, and it results in all the monkeys who hear it to run up their trees. Now what does this call mean? It could mean "jaguar alert!", pointing to a very specific concept-- a certain animal is here and we all know they're dangerous.

It could also mean "I'm scared!", and maybe it's just monkey see monkey do. It could also mean something more abstract, like a blood curdling scream-- there's no one thing that it means, but as humans we instinctively know that people don't scream like that unless something legitimately awful is happening. So maybe the call communicates emotion rather than an intellectual concept-- it's a call of fear that makes other monkeys who hear it also scared.

Just breaking down what animal language even _is_, is a challenge. I'm not optimistic on hearing any oral histories of whales, or even that they record history. I mean humans only started recording history for its own sake like 2000 years ago with herodotus. Before then we have tablets to keep track of stock, letters, and murals which were often made to depict the strength of the reigning emperor and the foes he vanquished. So maybe if we talk to whales it'll be a little like if aliens came to ancient Egypt to talk to the pharaoh; we'll just get a dictator whale telling us about all the other whales he's killed and how he's the greatest.. haha probably not that, though.

Re:Attempting to interpret sperm whale clicks with AI, then ... by secondaryacct on Sunday October 31, 2021 @10:33PM (#29062507)
Recording the feats of the emperor is already miles beyond what whales probably do and I d go as far as to call that recorded History,since the goal of writing is to allow the dead to speak to the future, a megalomaniac endeavour only overly inflated brains could think about.

After seeing that poor monkey spending his life learning sign language just to ask for more bananas and never asking a tough question, Im a bit pesimistic.

Re:Attempting to interpret sperm whale clicks with AI, then ... by easygenes on Sunday October 31, 2021 @10:56PM (#29062630)
Alex the African Grey parrot is I think the most developed example we have of animals learning human language and concepts. Alex is notable for being the first non-human to ask an existential question, asking, "What color?" when seeing himself in the mirror. He quickly learned he was grey.
Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @08:59PM (#29062032)
I think your entire post is hugely optimistic despite the negative overtone! The fact that other people see there are these important and fundamental questions that can be addressed by this nascent field gives me hope that this won't be another one off attempt like so many before. I'm very hopeful that it can snowball due to the hype around machine learning even if I'm personally skeptical that machine learning is a particularly fruitful avenue.
Re:Attempting to interpret sperm whale clicks with AI, then ... by senectus1 on Sunday October 31, 2021 @09:07PM (#29062076)
I would expect the vast majority of their comms to be :

Nice water over here

Loads of food here

I'm down to F, anyone else?

Do we really have any evidence that suggests it'll be anything else?

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @09:33PM (#29062206)
The best example I know offhand is the custom hunting styles of different orca pods. Given that orcas also play I find it more likely that there is rich communication than not.
Re:Attempting to interpret sperm whale clicks with AI, then ... by yalok on Monday November 01, 2021 @01:28AM (#29063282)
Orcas are mostly silent, especially when hunting, iirc.
Re:Attempting to interpret sperm whale clicks with AI, then ... by cetalingua on Wednesday November 03, 2021 @08:32AM (#29093709)
Actually no, orcas echolocate a lot and also exchange quite a few calls while hunting fish (based on acoustic tag data).
Re:Attempting to interpret sperm whale clicks with AI, then ... by jtbayly on Sunday October 31, 2021 @10:10PM (#29062414)
“He’s lonely.”

No. You’re lonely!

Re:Attempting to interpret sperm whale clicks with AI, then ... by throwaway2331 on Sunday October 31, 2021 @09:57PM (#29062356)
My less pessimistic (dare I say realistic, ha!) take is that whales will have little of note to say. Complex communication (in contrast to "simple" communication like the evocation of emotions, e.g. fear, anger, sadness, etc.) does seem to require the biological need for it, otherwise the genes that control factors that manifest it ("depth" and richness of sense, proclivity for specific mental models, etc.) do not get reinforced and "bred" into the animal.

The other side to this is that communication is mostly non-verbal. For example, chimpanzees, perhaps the most communicationally-advanced animals, besides us, do have a few different "verbal" signals they use. But there is also a richness of non-verbal communication (body language, facial expression, using their hands, etc.). And then these signals have to be also interpreted by other chimpanzees, thus training them to develop new mental models about these signals (and leading way to complex communication).

Whereas all whales really have are clicks, whistles, and "pulsed calls." Along with a tail with much less dexterity than a hand (but is still used in communicating amongst themselves). The richness of their communication is much less than the chimp, so so far it seems they can only use it for rudimentary purposes (mating, hunting/intelligent organization and navigation, and figuring out who is part of the "tribe"/pod).

But in that vein, they're still closer to a chimp than they are to us. Humans have evolved such complex mental models for the world and other humans within it, that all animals are basically on the same, much lower level in comparison.

On a less relevant topic, whales are social animals that stick together, make close bonds with family, and do have a certain "culture" that gets passed down, and is constantly morphing a la "whale songs." It almost pains me to think about what emotions a whale would feel (if it even does) about friends and family that have been hunted to death.

Re:Attempting to interpret sperm whale clicks with AI, then ... by franky47 on Sunday October 31, 2021 @05:10PM (#29060473)
> Lets get some whale engineers working on the hard problems please.

Somehow I pictured whales as the engineers in this sentence. It makes it even better.

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @05:27PM (#29060616)
That was the intent!
Re:Attempting to interpret sperm whale clicks with AI, then ... by ceejayoz on Monday November 01, 2021 @07:02AM (#29065174)
Re:Attempting to interpret sperm whale clicks with AI, then ... by Jxl180 on Wednesday November 03, 2021 @01:44AM (#29091456)
“You’re telling me a shrimp fried this rice?!”
Re:Attempting to interpret sperm whale clicks with AI, then ... by JackFr on Wednesday November 03, 2021 @03:52PM (#29099471)
Never gets old.
Re:Attempting to interpret sperm whale clicks with AI, then ... by forgotmypw17 on Sunday October 31, 2021 @04:19PM (#29060024)
You can do it today with other species. We are lucky to still be surrounded by all kinds of animals: mammals, birds, and even insects.

Of course, in order to talk, you have to spend a lot of time listening first. And what they say cannot often be translated to human talk.

In order to make friends, you have to give first. Our society teaches us to stay away from nature and leave it be, so you have to break past that.

The rewards are breathtaking and totally worth the effort.

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @04:34PM (#29060149)
Your point stands, but I want something wholly different from that. I want the equivalent of the LHC for cross species communication. You're telling me to go do some communing with physics, because we've already got some good textbooks on quantum mechanics.
Re:Attempting to interpret sperm whale clicks with AI, then ... by thaumasiotes on Sunday October 31, 2021 @07:10PM (#29061386)
> I want the equivalent of the LHC for cross species communication.

This is not a valid concept. What would the "equivalent of the LHC for communication between ordinary Americans and the Twitter API" look like? What uses would it have?

There's nothing out there that can receive arbitrary messages. Communication -- of any kind, with anything -- is limited to messages that the other party is capable of receiving.

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @09:04PM (#29062063)
I'm not sure I understand your complaint or why you've caught the Twitter API into it?

I imagine a center for cross species communication research would coordinate scientists across a couple important fields also interested in the question and allocate their resources to the research which seems most promising to make progress. Just like any other institute.

I imagine the fields would be biology zoology linguistics computation and probably since experts for the specific target species.

Re:Attempting to interpret sperm whale clicks with AI, then ... by thaumasiotes on Monday November 01, 2021 @12:36AM (#29063028)
> I'm not sure I understand your complaint or why you've caught the Twitter API into it?

Because I want you to tell me what you think communication between a human and the Twitter API would look like. I have taken the position that you've produced a bunch of words without bothering to think about what those words are supposed to mean, with the result that in fact they have no meaning and you're attempting to refer to a concept that cannot actually exist.

A general theory of cross-species communication is significantly more of an ill-defined problem than a specific theory of human-Twitter API communication, but it must include the human-Twitter API case, so it's a useful example to think about.

> I imagine a center for cross species communication research would coordinate scientists across a couple important fields also interested in the question and allocate their resources to the research which seems most promising to make progress. Just like any other institute.

Ah, just like any other institute.

Certainly you can have people dress up as scientists and act like they're scientists -- just like any other institute! -- but this is only a good idea if you think they can accomplish something by doing it. What you're describing is a pure cargo cult. Their only goal is to appear to be similar to other, more useful people.

We already have many experts in cross-species communication. We call them animal trainers.

Re:Attempting to interpret sperm whale clicks with AI, then ... by azeirah on Sunday October 31, 2021 @07:25PM (#29061460)
This is Wittgenstein's argument about why we'll never communicate with lions (iirc) right?
Re:Attempting to interpret sperm whale clicks with AI, then ... by WalterBright on Sunday October 31, 2021 @04:29PM (#29060106)
Brain size may not necessarily correlate with intelligence because it isn't the number of neurons, it's the organization and optimization of them.

Just like we make much more powerful CPU chips in the same volume of silicon as before.

And it could be that it just isn't necessary for whales to optimize brain density, like it is for humans and crows.

Re:Attempting to interpret sperm whale clicks with AI, then ... by taneq on Sunday October 31, 2021 @06:27PM (#29061076)
I’ve read various stats about the human brain using a surprising percentage of our overall energy budget. I wonder how much energy a whale brain uses (overall and per kg)? While it’s not the only relevant stat, TDP does provide a clue as to processing power.
Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @04:38PM (#29060188)
Agreed! But afaik the neuron process "node" across species is not so different. The relevant metaphor I would think is humans have pcie and a large register count while many (all?) others are still on pata and register starved. Maybe a real biologist can come in with more facts and less bad metaphors
Re:Attempting to interpret sperm whale clicks with AI, then ... by stubish on Sunday October 31, 2021 @08:44PM (#29061955)
So far it is looking like the number of neurons in the cerebral cortex is the best indicator of intelligence.

http://mitp.nautil.us/feature/227/the-paradox-of-the-elephan...

We started as basic primates, started cooking our food, getting more nutrition, allowing our guts to shrink, allowing us to spend more energy on our brains, growing more and more neurons in the important bits, at which point things spiraled away using our new language and technology to make better technology.

(Not a real biologist)

Re:Attempting to interpret sperm whale clicks with AI, then ... by kevinmchugh on Monday November 01, 2021 @12:21AM (#29062966)
We've invested quite a lot in trying to communicate with other ape species. There's very little to show for it. Koko the signing gorilla is the biggest success and 1) that was conducted hugely unethically 2) had a closely-involved researcher/caretaker/interpreter. Without the lead researcher communication was very limited.
Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Monday November 01, 2021 @12:40AM (#29063052)
I think the actual investment is shockingly little. Even compared to a moderately sized zoo the investment is tiny. It's like a dozen individual people trying over the past five decades.
Re:Attempting to interpret sperm whale clicks with AI, then ... by tablespoon on Sunday October 31, 2021 @10:01PM (#29062379)
> God I cannot wait to talk to whales. Their oral histories must be incredible.

What makes you think they'd have oral histories?

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @10:23PM (#29062473)
I think it's a decent enough guess and I'd give better then even odds on it.

Whales have the capacity to communicate. Stories are more effective than unadorned facts at being remembered (for humans anyways). Remembering important facts is a huge advantage in almost any context. The biggest jump imo is the second one, but I find it unlikely that whales would have a radically different memory storage and retrieval model than humans

Re:Attempting to interpret sperm whale clicks with AI, then ... by lm28469 on Monday November 01, 2021 @08:03AM (#29065619)
There is nothing at all that hints to that though. It makes me think of people who believe their dogs can "smile" the same way humans do.

Humans are very good at two things, seeing patterns and projecting their own feelings/experiences on other living creatures (or inanimate objects even). We're the exception, not the norm.

You can't "guess" or rely on intuition for these topics, that's how we got religions, the geocentric system, &c.

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Monday November 01, 2021 @11:18AM (#29067802)
Of course you can guess and rely on intuition. The key aspect of science is the feedback loop with data to refine/reject those guesses and intuition.

Its also hilarious to me that you call out my guess and then make your own unsubstantiated claims, which I'm sure feel "intuitive" to you.

Re:Attempting to interpret sperm whale clicks with AI, then ... by thaumasiotes on Sunday October 31, 2021 @07:06PM (#29061354)
> Whales have the capability to far outstrip humans in intelligence (if you accept that neuron count and neuron connections are the raw inputs).

Understood normally, the raw input also includes the software installed in the whale's brain. This limits the potential capability of whales.

If you were to overwrite that with software of your own design, you'd have a robot in the body of a whale, but not an actual whale.

Re:Attempting to interpret sperm whale clicks with AI, then ... by PaulDavisThe1st on Monday November 01, 2021 @12:39AM (#29063047)
There's no software in a whale's brain, or in a human's brain.

As has been discussed in a couple of articles linked on HN recently, it might not even be useful to think about brain function as computation.

Re:Attempting to interpret sperm whale clicks with AI, then ... by thaumasiotes on Monday November 01, 2021 @12:45AM (#29063083)
> There's no software in a whale's brain, or in a human's brain.

Sure, and man is the only animal with no instincts, ensouled cells are metaphysically different from soulless cells, and fossils were placed in the ground as a test of your faith.

Re:Attempting to interpret sperm whale clicks with AI, then ... by 01100011 on Sunday October 31, 2021 @09:31PM (#29062193)
Do we even know that whales communicate with conscious intent? What if the sounds they make were mere reflexes driven by some internal state like adrenaline, sex or hunger hormones?
Re:Attempting to interpret sperm whale clicks with AI, then ... by cetalingua on Sunday October 31, 2021 @10:35PM (#29062521)
We do, actually, for dolphins at least. Researchers did an experiment where dolphins had to coordinate their actions (push 2 buttons together) without seeing each other. So they coordinated their actions acoustically by communicating when they should push the buttons.
Re:Attempting to interpret sperm whale clicks with AI, then ... by aahortwwy on Sunday October 31, 2021 @06:47PM (#29061227)
> If a lion could speak, we could not understand him.
Re:Attempting to interpret sperm whale clicks with AI, then ... by PicassoCTs on Sunday October 31, 2021 @04:50PM (#29060289)
They will sing us the song of the great holocaust of the 1800s, of the day the metal surface whales turned the skies red and nothing could save them.
Re:Attempting to interpret sperm whale clicks with AI, then ... by lovecg on Sunday October 31, 2021 @06:16PM (#29060986)
Personally I want to know if they still tell tales of the hero whale who killed one of the ships in 1820: https://en.wikipedia.org/wiki/Essex_(whaleship)
Re:Attempting to interpret sperm whale clicks with AI, then ... by klipt on Sunday October 31, 2021 @02:33PM (#29059166)
> Their oral histories must be incredible.

Everything was idyllic in the before times. Then the human ships arrived and ruined everything. Some of them killed us. Others ignored us but polluted the ocean with noise so we couldn't hear each other's whale songs anymore.

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @02:59PM (#29059377)
Maybe! But I find that an awfully self centered view. I imagine humans (boats?) will play a (perhaps significant) aspect in their vision of the world. But I'd be surprised if it was any bigger than e.g. malaria is for humans.
Re:Attempting to interpret sperm whale clicks with AI, then ... by blacksmith_tb on Sunday October 31, 2021 @04:45PM (#29060250)
I would say that's lowballing it considerably[1]. Whaling nearly drove sperm whales to extinction until they were protected in the 1970s.

1: https://en.wikipedia.org/wiki/Sperm_whale#Relationship_with_...

Re:Attempting to interpret sperm whale clicks with AI, then ... by Misdicorl on Sunday October 31, 2021 @05:01PM (#29060391)
Fair enough
Re:Attempting to interpret sperm whale clicks with AI, then ... by TheOtherHobbes on Sunday October 31, 2021 @06:45PM (#29061208)
There have been a number of Orca attacks on yachts in the Mediterranean and the Gulf of Cadiz/Portugal area recently.

Perhaps they're hungry. Perhaps they're pissed. No one knows.

Historically, orca attacks on humans - outside of captivity - have been very rare.