One thing that will have thrown the author off the trail is that he is holding a fossil of the organic parts of the snail and that is essentially a cast of the animal, not the shell. They are known as Steinkerns (stonecore).
The insides get replaced by minerals, which harden, the shell dissolves, then the only fossil remaining is a mould of the inside of what used to be the shell.
So on a fundamental level, the headline is wrong. He did not find any sort of shell...
I've been told by a friend -- a wierd thing -- in many places you can dig a hole and it will fill with water. And at some point in the future fish will be swimming in it.
True, but I actually had no idea that it was the soft parts rather than the hard parts that had been fossilized. (I haven’t verified it yet.) Either way, it didn’t read like a bad faith interpretation/comment.
It wasn't written to be one. If the author went to the trouble of making a 3D space filled with many shells, knowing the actual shell was most likely a different shape would be something they would probably want to know, so the position of their fossil could be placed more accurately in the graphed space.
"He swam at my feet,
Powerful arms in broad strokes
Sweeping the sand.
So I asked this man,
What seas do you swim?
And to this he answered,
'I have seen shells and the like
On this desert floor,
So I swim this land's memory
Thus honouring its past,'
Is the journey far, queried I.
'I cannot say,' he replied,
'For I shall drown long before
I am done.'"
There's a lot more to morphology than just the shape of the shell, and indeed the shape can sometimes be misleading, in that very different species can have somewhat similar shells, and different individuals of the same species can have quite different shell shapes. You've got a gasteropod, so it would be good to pay special attention to the peristome and siphonal canal (based on the bio classes I took in the area, I'm no expert) but of course there's lots of features that could be helpful in an identification.
https://en.wikipedia.org/wiki/Gastropod_shell#Parts_of_the_s... is a good list, and maybe you've already done this but you would want to find a dichotomous key of gasteropod families native to the area to narrow it down. Good luck in figuring out your shell!
I'm a little confused about how significant of information can be derived from a 2d projection of the shell. This sort of mathematical modeling looks like phrenology.
Using PCA on 3d shapes is a proven method for identification. It's nothing like phrenology aside from both involving morphology. Former actually works, latter does not.
(I upvoted, seems a little weird that you were downvoted) So, I spent a little time looking up "Principal Component Analysis" and yeah, this method works well with these forms, and my comment comparing it to phrenology was definitely more out of ignorance!
I was most surprised that you could flatten a 3d structure down to 2d and not lose so much information that it would cause a very high rate of error. Someone else was skeptical enough to do a study as a critique on it, only to have it retracted. (funny in light of this post)
I have quite an early memory of being on the somewhat remote property in Australia that my mother grew up on (Central NSW, near Condobolin).
My Uncle, who then ran the property, walked over to a rock, whacks it with a hammer or similar, shows me a bit of a trilobite (which are totally different to our sort of bytes). He did this a bunch of times. I still have the rocks. No amazing full horizontal cross-sections, but it certainly got my very young mind excited.
There were fossils RIGHT THERE from before there were dinosaurs!
Oh, and that central Australia used to be an Ocean!
These clear demos to young kids, or adults, are great, and the many other examples here in the comments are a testament to that (Vienna? wtf!).
St. Stephens cathedral in Vienna was built with sandstone that contains seashells. It's hundreds of kilometers away from the shore, but ~15 million years ago the area where it stands now was a seabed.
The stones are not from the exact location where it was built, but from close by. The quarry where the stones came from hundreds of years ago is still active, and you can find tons of fossils there. It's practically impossible to get a piece of rock from there without visible seashells.
Almost any surface on earth was once under the water. You can find sea shells and various sea deposits high on some 8000m peaks or 1000kms inland.
Everybody who cares at least slightly knows this, and I am pretty sure author knows this too, he could have spared us the initial hyperbole. Analysis itself is good but not everything needs to read like dramatic novel.
If all the ice melted it would raise the oceans by something like 230 feet, so modern Vienna would still be above water at 495 – 1778 ft elevation.
Although some estimates suggest Earth loses 20 - 30 cubic kilometres of water to space annually. Plus whatever water is bound up in mineralisation annually.
450 million cubic kilometres of water lost over a 15 million year period would lower ocean by something like … a bit?
The total volume of water on Earth is presently estimated to be around the 1.386 billion cubic kilometre mark.
The volume of a sphere increases to the cube of its radius … carry the 1 … nup, that’s to hard for me.
Herodotus wouldn’t have written it quite the same — no Jupyter notebook for example,
Jupiter being a Roman term; the nearest equivalent for the Ancient Greeks was Zeus.
I've got some 'shells' from the deserts near Siwa Oasis. Quite a cool feeling walking on top of what used to be the floor of the Tethys Ocean, especially when everything around you is just rock and sand without a drop of water in sight.
yea, finding shells all over was one of those things I was taught in elementry school. We even had a field trip to some place to find them along with geography lessons on the various layers of the area
Gemini says "As the crow flies (Straight-line distance): Approximately 900 to 920 kilometers (roughly 560 to 570 miles) directly north of the coast at Karachi"
Maybe a nitpick but Latamber is not directly north of Karachi and it's about 1000 kilometers away (the closest coast is 950 km but not in Karachi). It's easy to see and to measure on a map.
Maybe some geology buffs can correct me, but as I understand it there has been three periods with ocean on top of the crust we call Pakistan today. The Proto-Tethys, Paleo-Tethys, and Tethys Ocean. Many hundreds of millions of years of being ocean.
Why would you write a lot of software to find the closest match (which doesn't even seem that good) if you could also ask a subject expert? I guess you could even just post a photo to some subreddit with people who could tell you what it is...
Also: "it shouldn't be here; the nearest coastline is Dammam's, 500 km away." - are people really that ignorant about plate tectonics and sea fossils in mountains?
He is losing a lot of information in that normalization pipeline (whole shell reduced/feature engineered into nothing but an outline). A CNN or something similar would be better and he can maybe get a better depth map of the mouth shape.
The analysis is nice, flashy, and wrong. Several weak assumptions here leading to hallucinate an obviously wrong result.
Taxonomy IS a science. Just use the wide corpse of knowledge that has been built for the last 229 years, where the class Gastropoda was created.
First wrong assumption. This is a seashell.
This probably is a seashell, yes.
But fresh water snails have also shells; and savannas can have a lot of lagoons before eventually turning into deserts. If you train your model only using zebras, your model will happily conclude than an hippo is a sort of non stripped obese zebra.
1) The model use incomplete data. The data used to train the model is based in 7800 species alive. After wikipedia, Gastropoda have more than 75000 species alive, plus 15000 fossil species known. (We can assume safely that this is a snail, but remember that some cephalopods also have coiled shells).
2) The model use spurious data. All clams and Tusk shells must be removed (because we want to classify a snail). This means that the number of snails available to train the model is much lower than 7800. Including non-snails just gives us a false confidence in the strength of our model.
3) The model covers only one couple traits in this species, but this particular traits can vary within members of the same species. Taxonomy uses thousands of traits to classify a mollusc and some are particularly fastididious. Dozens of items only to describe the shell. Often the soft parts are needed (Is the penis shaped like a club? this genus, shaped like a whip? this other one; the penis in your sample is contracted because you didn't put to sleep the animal first with mint crystals, though luck, we'll never know).
4) The model is based in extant alive species, but we want to identify a fossil. Alive species have non-distorted shells. Fossils often lose their shape by the weight of sediments and compression. Only the thickest shells would keep its real height/wide proportions.
5) The model ignores important details. The species found in the desert has a very evident shell groove at the top of the spire, that the targeted species does not have. This alone, tells a newbie taxonomist that the result is wrong.
I found a seashell in the middle of the forest in (well inland) Mississippi. That was an interesting find, and lead me to learn that much of the continental US used to be covered in a sea, I believe https://en.wikipedia.org/wiki/Western_Interior_Seaway
This means the shell was dozens of millions of years old, and may be the oldest thing I’ve ever held, except maybe some rocks.
It's interesting that saying the Earth is more than 10,000 years old is not haram in Saudi Arabia. I thought it would be, since they are so religious, but it turns out the Koran doesn't make any claims about the age of the Earth, so you are free to say that the Earth is billions of years old and not be accused of blasphemy.
I mean, even in countries with extremely high levels of Christianity, such as Romania, you won't be accused of blasphemy for just... agreeing with science. Both Christianity and Islam are pretty similar after all, so hopefully that is not too surprising.
I can't think of a country where you even might be accused of blasphemy, though I'll admit I am not very familiar with the topic.
Even with AI, to try to replicate this on my own would take me a really long time, maybe impossible. Despite the use of AI,it would be a huge undertaking , such as having to come up with the blueprint and procedure for classifying the shells, setting up all of the environments, setting up repository, understanding the math, writing it up, coding the tool, etc.
This should allay fears that AI will render people jobless or automate everything.
Among a strong field, this is the single most depressing comment I’ve ever read on Hacker News. Several grim components but it’s the “I don’t understand why” which seals the deal.
It's a good starting place. As an analogy imagine someone wanted to look up the definition of a word. If someone wanted to know the definition and they went out and crawled the entire internet, built an LLM, and then asked it for the definition you would wonder why they didn't check an existing dictionary first. I wouldn't consider it silly at all to always check a dictionary or existing LLM first when you want to know the definition of a word.
Wanting to know the definition of a word is not an original problem. Similarly wanting to know what's in an image is not a new problem either.
I understand this is more about the process than the result, but note that: a) his result is completely wrong, he identifies a living land snail as the 100+MYa fossil; b) a conversation with Claude helps with some decent knowledge and provides a few possible candidates, that can be then double checked. c) Claude could have talked the author out of trying to identify a Jurassic fossil against a database of living species.
I trust a proper solution (even though I can be certain how accurate it is), which compares to a known dataset much more than just giving it an AI. For identifying current living species it is probably fine but this is something to nice for an AI to be trustable. Also this path is much more fun and you learn sonething along the way!
but, from my understanding what the author was really wanting was an adventure and to learn new things. he gained so much more than just learning what type of shell it is
No one. I'm pointing out there are existing AI models that can do this that the author could have tried before investing all the work to build his own.
I don't think you understand why the author did this on a fundamental level. Sometimes it isn't explicitly about getting the outcome directly, it's about putting in the work to understand how you get there.
The article says that the author wanted to have an analysis done, but ruled out the option of a human doing it. He did not rule out using a multimodal LLM. It's possible that he just is unaware that multimodal LLMs are capable of doing an analysis. Considering the approach wasn't mentioned in the article one can guess that the author just didn't know it was possible. But there is no way to know for sure.
The AI would confidently give him the wrong answer, since it has no way to provide the correct answer, and doesn't know its own limitations. (Or however you wish to describe "hallucinations", which is about as accurate as my description ;))
And he would think he has the right answer, perhaps write up an essay about his findings, which later AI bots will read and learn from, propgating the mistake...
Wait, the author identified the shell as "Sphincterochila candidissima". Which is a living species of air-breathing land snail, a terrestrial pulmonate gastropod mollusk. Completely off.
Non-tech-related stories frequently make the front page here too. It just so happens that this site has a lot of coders because, err, it's literally Hacker News.
Maybe. But I don’t see anything in this piece that says that it’s a fossil, rather than something that resembles this person’s idea of a fossil. It doesn’t look like a fossil to me. It looks like a piece of rock that’s been bashed about a bit.
And given the whole premise of the piece is “this should not be here!” I don’t really understand the point you’re making. The author says it’s a strange find in that area - so either they have a valid point or they don’t.
I don’t know if it’s a fossil. It doesn’t look like a fossil to me. I’m not a fossil expert. The only way to tell if it is a fossil is to do some analysis on the actual specimen before writing screeds about what it might or might not be based on visual similarity.
Well, it's your words against his. You're not much of an expert by your own account in other comments. It's irrelevant if it's from a correct geological period, it's a rock hard seashell. Go and read up the definition of "fossil".
Author points out themselves, in the second paragraph, that its not a strange find. The strangeness of the find is his personal experience. Not that its a strange find geologically.
And then the author takes a massive leap from “I found a fully solid rock that eerily resembles a seashell” to doing an analysis that treats it as though it actually is a fossil.
And that analysis finds out that the shell the assumed fossil most resembles is completely out of period.
If we're going to rate annoying takes on the internet, "some guy who knows nothing about a topic being snarky because AI was involved" is far worse than somebody doing something with AI.
I’m really not sure what you mean. Did you actually read the article? There is nothing in there that confirms this is a fossil. One moment the author says “I found a fully solid rock that eerily resembles a seashell” and the next minute they treat it as though it is a fossil but their analysis shows that the shell the piece of rock most resembles is from a completely different geological period.
A number of people in this thread arguing with me in a remarkably personal way apparently care quite a lot. And I guess you must care at least a little bit given you’ve made the effort to comment.
