Will your family turn you into a chatbot after you die? Plus, synthetic squid skin, and the sway of matriarchs in ancient Anatolia
45:57 Share

This is a science podcast for June 26, 2025. I'm Sarah Crespi. First up this week, contributing correspondent Andrew Curry's here, and we talk about how researchers used ancient DNA to show the importance of maternal family ties in Çatalhöyük. That's Neolithic Turkey 9,000 years ago.

Next on the show, researchers were able to make a synthetic material that changes color the same way that squids do. Researcher Georgi Bogdanov joins me to talk about how his lab was able to discover the subcellular arrangement of proteins in squid cells and then mimic that pattern synthetically.

Finally, we hear about the latest book in our series on science and death. Books host Angela Saini talks with Tamara Nese about her book, Death Glitch, How Techno-Solutionism Fails Us in This Life and Beyond.

Çatalhöyük is a Neolithic settlement in Turkey dating back 9,000 years. This ancient place was occupied for about 2,000 years and has been well studied by archaeologists. This week, contributing correspondent Andrew Curry wrote about what ancient DNA from this place brings to the picture. Hi, Andrew. Welcome back to the Science Podcast. Hi, thanks. Yeah. So have you been to this place before? Yeah, I was there in 2017. It's a pretty incredible place.

I think since I was there, they've built a big new museum and visitor center. It's a UNESCO World Heritage Site. But the excavation itself is massive. It's like a high school gymnasium maybe or bigger. They spent decades excavating individual houses and all the houses were just piled up next to each other. There were no streets. People would get in the houses from their roof. So cool. Yeah. Immediately makes you think how the thinking was different. What

what people thought was a good way to live was different all the way back then. So this is a tell, I think I read, which is like as the society ages, they build on top of old places and you get kind of a hill. Yeah, exactly. You know, as the houses got, they collapsed or there was a fire or something happened, they would just start over and build on top of the old stuff. So as archaeologists dug down, I think they have 18 different layers. That is so cool. A

over time with the oldest at the bottom. Besides being extremely, extremely old, what else marks Tattlehoek as a special place? It's one of the earliest settlements where people were also farming, which is what makes it Neolithic. Like it's still the Stone Age, but the point in the Stone Age where people are moving from hunting and gathering to

and increasingly planting crops and herding animals and stuff like that. And one thing I've learned interviewing a lot of people who study this is that that didn't happen at the same time all around the world. So the Neolithic, it's the thing that happens in a place and that timing might be a little bit different. So here in this part of Turkey, you know, it was at this time, but it might be a little bit different in other parts of Europe or in China. This part of Turkey is where it started. This is the beginning of the Neolithic.

It spreads from here or comes independently in some places. But this was really fascinating to archaeologists because it's where they could see the society that originated at the same time people were really embracing this new technology. When people went and looked at

The kinds of art they made, the kind of buildings they made, kind of drew some conclusions about the roles of men and women. And this has been very long debated. Was this egalitarian? Was this a matriarchy? Now here comes ancient DNA to perhaps shed some light, perhaps just throw some more confusion on the fire.

So what do they do? How do they sample and how far back did they go? So the other really characteristic thing about this site and other sites in the area from around the same time is that people buried the

the dead in the house, under the floor, right close to you. And these are not large houses either. So this is like a really intimate connection with presumably your loved ones or the people who are important to you. They would also sometimes remove the head later and use it as decoration, which is... Yeah, that hung out in the house with them. But unlike a cemetery...

where you can see what people did with the dead, but it's sort of removed from the living space. This is really connected with the living space. So this most recent research, they went and got DNA from about 150 people who were buried in the floors of these houses. Most of the ones where they could recover DNA successfully were children. And then they looked at how those people were related, which is something you can only do with DNA.

First of all, children mean that they're not parents. And so you kind of are just at this like terminal branch of the family tree, right? So it doesn't necessarily tell you as much as you want, but what do we know about the relationship between the children that were buried in the floor of these houses? So what they found was that the children tended to be related through their mothers, which isn't how we usually do family trees in Western Europe.

or North America, but the children were related through their mothers. Girls and women tended to be more likely to be related within one house, which suggests that men were coming in from other houses, whereas women were staying in the houses maybe where they were born or the family idea was constructed along the female line. Yeah.

Anthropologists have been able to find this through ethnography in certain parts of the world. And you can say, oh, well, when you get married, you go to your mother-in-law's house or your husband comes to your house or that kind of thing. But it's much harder to say that about something that happened 7,000 years ago. So this is kind of a clue that that might have been going on, but it's not 100%. It's not 100%. When they first excavated this site, I think they started in 1958 and there were excavations in the 50s and 60s.

It was a big deal because there were these prominently displayed female figurines, very feminine figures sitting in what people interpreted as thrones. People suggested that maybe they were mother goddesses and that was a suggestion that this was a matriarchy where women ruled. And then there was a break in the excavation. There was a lot of pushback against that idea. People said, well, you know, maybe this was overinterpreted.

All we can say from the evidence is that men and women seem to have been equal. So this also comes from the bones. And you say, oh, well, some places where there is inequality, a lot of times you see kind of an undernourished sex and an overnourished sex. Right. Or one sex has more evidence that they're doing the hard physical labor every day from their skeletons. And you don't see that. And so the skeletons here, they say...

Everybody's doing the same thing, eating the same thing. They got to egalitarian. And now the DNA evidence says, well, people were paying attention to whose mother was whose, maternal and matriarchal stuff. But we don't know the power dynamics from that. There was this pushback and people said, well, maybe gender didn't matter. And this evidence really shows, at least in terms of how they thought about family and burial practices, they weren't.

they were really paying attention to gender. And then there's some other like little clues. For example, girls tended to be buried with more stuff, beads and pottery or whatever, like five times as much stuff as boys. So again, something's going on, but more stuff doesn't necessarily mean you're more powerful or you're in charge or you're richer. It could just be one person I talked to suggested that

Maybe it was the male's role to give things. Who knows? But gender mattered and women clearly had an important role in structuring this society. What is really interesting for me is that all of the other genetics papers about the Neolithic have shown the opposite.

that it was patriarchal or patrilineal, that society was organized around the men. And this is like if you just hop over to like Europe? If you go all the way to France or Britain, in this place where the Neolithic started, they were doing it differently. This settlement was inhabited for thousands of years. Was there a change through these layers about these relationships according to the ancient DNA? Yeah. So this is another exciting element that just shows you how different the past must have been or

or could have been. In the earlier layers, there's strong evidence of genetic kinship. People are biologically related. Later, that evidence declines, like girls are still more likely to be related, but you start seeing houses where people are all buried together, but they don't necessarily have genetic relationships to each other or close genetic relationships to each other. It

including kids who are genetically unrelated or not closely related, but they're all eating the same things, which they might be all nursing together. Isotopically, you can say they had the same diet. Isotopically, yeah. So then that also raises this sort of interesting, you know, we think of family very much as biological mother, father, child, but maybe they're fostering or

or there's adoption, or maybe it's some other way of constructing it. Nurseries, yeah. I mean, it's not nurseries in like daycare sense, because you're also buried in the same house where... Yeah, yeah. You stay there forever. Yeah, exactly. That's so interesting. And that was how much later? What kind of timescale are we talking about? The genetics that they looked at is about a thousand years. So it's sort of...

continuum, but in the last 500 years of the period they looked at. I'm just going to point out how vast of a span of time. It is. It really is. This is a hot, dry part of the world and DNA doesn't preserve so well. I think they tested hundreds and hundreds of skeletons and they only got evidence from, I think, 131 and they could do sex estimation on 180. But

But they've done a lot of statistical work and this really cool combination of archaeology and isotopic evidence with the genetics that gives you this cool picture of what was happening. Are they going to keep looking for ancient DNA among remains there? Or are they going to look for this pattern in other places? What's next?

I think what they hope to do is now people's expectations about what the Neolithic should look like have changed a little bit. Maybe when they start looking for these kinship patterns at other sites, they'll be more open to different interpretations that there wasn't always this one patriarchal structure and societies took different paths. Thanks, Andrew. This has been really fascinating. Yeah, thanks for having me. Andrew?

Andrew Curry is a contributing correspondent for Science. You can find a link to the story we discussed and related research at science.org slash podcast. Stay tuned for a chat about color changing materials based on the principles of squid skin.

Squid are able to change color or even in some cases become transparent for camouflage reasons, for communication reasons. To do this, the squid has these cells called iridophores. They're iridescent cells that contain elaborate structures built of reflective protein plates. This week in science, Georgie Baganoff examined the details of these structures and created a color-changing synthetic material based on squid skin principles.

Hi, Georgie. Welcome to the Science Podcast. Hi. It's a real pleasure to be here. We're going to have to do a bit of a deep dive into what structural color is today.

And let's just get the basics out of the way. Pigments, which is kind of what most people are used to thinking about this is like dye or paint. It absorbs certain wavelengths, reflects other wavelengths. That's what you see. Structural color is more about kind of like the texture of the surface or, you know, tiny elaborations on the surface that interact with light and scatter it in these ways that you can kind of set up so you can show certain colors. You see this in butterfly wings or if you look at a bubble, a soap bubble, it has a difference in its reflectance

refractive index, and that will bend light, separate it out into different colors, and you can see different colors from different angles. And that's kind of more along the lines of what we're going to talk about today. So let's start with the squid's amazing ability. What happens when a squid is changing color? What does that look like? What they do is they utilize their multi-layer structure. So the pigments that you mentioned, yes, people usually think of them, but there's just only the top layer of skin. Those contain pigments.

Everything else is just a cell-filled tissue. And those cells contain mostly protein reflectin. Right. Reflectin is a very aptly named protein. The interesting part about that protein is that it's not absorbing almost anything. So it can be used to purely reflect, scatter light and interact with it. The animal can adjust their skin layers separately and interact.

modify their appearance in the visible light.

In order to achieve these complex effects on the skin of a squid, they have this chromatophore layer. This is where you get pigments and little pouches, and they can squeeze or expand those to change the color of their skin. Under that layer is this other tissue that's packed with iridophores, and these are the cells containing this reflectin protein, and that basically interacts with light using structural color principles. We're going to talk a lot more about that in depth.

So within these erudifers, you have this arrangement of the plates of reflectant

What do we know about this layer of tissue, this part of a color-changing process? We found two different clusters of cells. So first, these splotches, the colorful splotches, they actually actively change color. They can go from blue to red very fast, and they can become transparent as well. And then you have surroundings. Surroundings are not that colorful. You can see color only at certain angles, mostly, and they give just like a

small shear of some color. It is not something specific. And that's actually because of the different morphology of those cells. So those colorful splotches, they actually have, we estimate about 96% of their insides are filled with those specific cells of specific morphology. And those cells only contain the stack-like structures, which was described before, but we went...

in depth on those studies and we actually focused more on the subcellular level of those structures. The cells are really packed in there and they're full of these plates and those plates, those reflectin plates, are arranged in different patterns and those patterns

If you change, you squeeze on them, it changes the distance between the parts and that changes what kind of light they interact with. That changes that wavelength. And not just that, you can also do both mechanical and chemical actuation, meaning so you can, for example, hydrate the protein. And again, separate studies show that when you hydrate those proteins more, you get different refractive index. So it also changes the color.

If you can manipulate both what we call periodicity of those structures, meaning the thickness of the blades and the spacing between them, as well as refractive index of the platelets themselves, you can basically achieve any color. Breaking down these are...

These are two approaches. You have the periodicity, which can be changed by squeezing on the plates. That's changing their thickness and how often they're repeating. And then you can also change the refractive index by hydration, by adding water or a liquid. And that's changing the way the light is passing through these structures and these cells. So these are kind of the two different approaches of achieving color in their rotifers.

Also, we should mention that one of the colors is transparent that can be formed from these processes. It is transparent because we just don't see it. It shifts to the IR range. Oh, okay. We show that it shifts when you increase the spacing and the periodicity, it gets red shifted. And so we stop seeing them. The squid's not invisible, it's not see-through, but the patches in this layer of tissue, they're

they're not visible anymore because they're only reflecting in the infrared in the IR. We like to think that if we see something of certain color, that it has this color. But no, it's just because it reflects only that color. So the same is here. If we see something colored, it's not because it has that color, it's because it reflects to us and that's what we see. But if you had some other

structure of the eye, you would probably see a different thing. We should probably get into the subcellular structure here, the arrangement of the reflexive plates and how this protein and its interaction with light work out. The most important finding here is that there's this gradient of the refractive index within the plates, within the stacks of plates.

The refractive index varies through space and that variation can be described as a wave. Like I know it's complicated, but if you think about, you should see have a stack of plates, any point across that stack, you're going to have one refractive index and then at a different point, it'll have a different one. This gradient across the plates is something that you're really able to describe in detail here. And actually, you know, this

gradation, this distribution of the refractive index, it has a specific shape. The different types of distributions that we found and we compared to are sinusoidal triangle and square wave. And these waves are basically the functions, the mathematical functions, which describes the way the refractive index changes when you go across that stack of plates.

It goes to a higher value in the middle of the plate and then to the lowest between the plates within the spacing. When you reach next platelet, it starts going up and then you leave the plate, the boundaries of the plate, and it goes down again.

When you have a sinusoidal wave, so it goes up and down gently, no abrupt changes. When you have triangle, it goes up and then it suddenly goes down. With a square, it's a discrete function. So it's either minus A or plus A value that it jumps between those two. So the refractive index changes across the depth of the plate.

Yes. How were you able to figure out that there were these different gradients, these different patterns and what they corresponded to? So we use a relatively new technique called whole tomography. It allows us to image single cells extracted from the tissue in three dimensions and map refractive index distribution within that cell.

So it allows us to directly measure, not just estimate, but directly measure the refractive index of a specific place within a specific cell. And we also were able to use that three-dimensional data directly to translate into our computational models where we applied

the parameters found within the squid skin cells to estimate their optical properties. We performed a bunch of simulations where we compared the sinusoidal wave to triangle wave and square wave type. We noticed that

With those types of distributions, we would be able to achieve specific color with different sets of parameters. While having a sinusoidal gradient distribution gives us the advantage to achieve each color only with one set of parameters. We have this waveform that you investigated.

And you said, well, which one of these corresponds to kind of the continuous ability to change color rather than just switch red, blue, red, blue? You can kind of set up parameters so any color can be expressed with these plates. You found that computationally. You think that's what you see in the squid. Then you took it over to the bench and said, let's make a material that has this gradient and see if it can also accomplish this. It has the same properties as these cells in the squid. What did you make it out of?

Did you just happen to have that kind of material on hand? This is actually based on studies that our lab has done previously. We utilized the evaporation technique where you can basically evaporate metal and it would reside on the substrate in a certain way.

Basically, you can create structures with different refractive indices. And if you do it continuously, so you can grow these stacks where, again, that sinusoidal or triangle or square wave type of refractive distribution will actually result within each column. There will be a lot of these columns next to each other.

basically emulating what we see in the squid. So you're able to deposit shapes of the appropriate refractive index as well as spacing to try this. But then did you also stretch them, compress them or apply a chemical to see if it could change? The dynamic aspect of it was one of the key things we were after.

So of course, when you're depositing inorganic material, in our case it was titanium dioxide, when you deposit it, of course, it's not going to be acting as a smart material. You need to apply something external to it. First, we embedded it in polymer matrix, which allowed us to later stretch it. And that emulated basically the mechanical actuation within the squid skin.

Also, what we wanted is the chemical actuation. The combination of two can give us both transparency shift and color shift. So the stretching makes it more transparent and the chemical actuation would allow us to shift colors.

This work also explores the ability of the synthetic material to interact with the infrared portion of the electromagnetic spectrum, basically with heat, which I would think squids would care a lot less about. Squid do not need infrared capabilities because they live in water. They don't need it in the ocean, but we might need it here in our above-the-water world.

What we wanted was to achieve the camouflage properties, not just within the visible spectrum, but infrared. So by stretching, our materials can let through more infrared light or heat. We can detect more heat through. And if we stop stretching, it just completely seals it and you don't see anything. Why do you want it?

a material that can change color, become transparent, let heat pass through or lock heat from passing through, all built into the same kind of structure. And it's also stretchy. What would this be for? Well, you can find a lot of different applications. It can be used for something decorative or, I don't know, for some camouflage properties. Right. So you can make something match the environment visually, and then you can also match heat or at least control the thermal output.

to match the environment or to tamp it down? You can, for example, use it as a space blanket alternative, which we have made before in our lab. But what I think is the most valuable part of these studies is that the gradients of refractive indices that we found

It actually allows you to overcome a lot of issues within the materials when you use them for optical properties or other properties as well. So if you're interested in controlling or interacting with other objects,

other parts of the electromagnetic spectrum, not just visible light, not just infrared light, but other parts will also be susceptible to these structures and this arrangement of refractive index. So if it's within the visible light range, you can affect the appearance of those materials. If it shifts into different wavelength range, then you can manipulate different type of radiation. Based on that, you can apply

our findings to many things and many materials where basically sky is the limit. This could be used for shielding, letting certain types of rays pass at certain times and not others.

Going back to kind of where we started, though, what made you decide to study this in the squid? Why did you look to their skin to work on manipulating the electromagnetic spectrum? So our lab mostly works with squid-inspired things, if you will. And we mostly work with that reflecting protein, at least our biocide. What we're interested in is actually going back

not just work with reflectin protein and what it does, but also where it's all coming from. The physics of it. Yeah, there have been some preliminary studies on the properties of those skin cells. They were not quantitative enough on the level of single cells. And while we imaged and analyzed thousands of cells across several years, we were able actually to extrapolate those findings and map

those refractive index distributions that actually allowed us to find that key aspect that was missed before. Thank you, Georgie. I really appreciate your time. Thank you. It was a pleasure sharing our findings with everyone. Georgie Bagdanoff is a postdoctoral researcher in the Department of Chemical and Biomolecular Engineering at the University of California, Irvine. You can find a link to the paper we discussed at science.org slash podcast.

Stay tuned for the second in our six-part series on books exploring the science of death. This month, host Angela Saini talks with Tamara Neese about what happens to our online lives when we die.

Hello, I'm science journalist Angela Saini, and this is the second edition of our book series in which I interview authors on a theme. In previous years, we've explored food and agriculture and the science of sex and gender, but this year we're taking a far more existential turn by looking at death.

So this month, I'm speaking to Tamara Neese, who directs the Climate, Technology and Justice Program at the Data and Society Research Institute in the US. Her latest book, Death Glitch, How Technosolutionism Fails Us in This Life and Beyond, is a book

asks what happens to our digital remains and online belongings when we die. Tamara, thank you so much for being here. This is becoming a much bigger issue in tech circles, but first I want to ask you, what brought you to this subject in the first place? When I was in college, and this is back in the early days of Facebook, a college student at my liberal arts school died suddenly.

And everyone immediately checked his Facebook profile to, I guess, feel connected to him or to verify that they had seen the person in real life at some point, just as a kind of way of making his death feel more material in a way.

And I was struck by the fact that his Facebook profile photo at that time was something that his family members would probably not approve of. It was a image of him smoking a joint with a cloud of smoke around him.

And I thought about the problem of what Dana Boyd and Alice Marwick have referred to as context collapse. So this problem of creating a profile when you're a college student and having a good time in life

And then inadvertently your profile becoming a memorial to you after the fact in an ad hoc way, in a way that you did not anticipate. And that may, in fact, really not be the way that your family members and loved ones would want to have you memorialized.

That's such a vivid early example. You write in your book that death is a kind of glitch because it's an event that designers never originally considered when they were developing the internet. What repercussions does that glitch have for us as users? Yeah, so part of the problem is that individuals, when they create accounts, they're really intended for use by one person.

And yet after people die, their families often do need to gain access to their accounts. It may be for practical purposes or perhaps they want to stop spam from manifesting on, say, a Facebook group page or a Twitter account, a NowNets account.

And, you know, part of the problem is that legally companies do not have to give access to these accounts or allow people to actually get into the accounts of their dead loved ones. Although there have been many legal cases, particularly in Europe, where privacy laws do protect the rights of people whose loved ones have died in terms of getting access to their digital accounts. And so part of the problem, too, is that each individual

company, each platform has a different protocol for how to deal with death if they have any kind of memorialization policy at all, which many companies still do not. So TikTok, for example, is a company that does not currently have an active memorialization policy. Facebook has changed theirs various times over the years. One of the problems is that with Facebook, they have relied on individual users appointing

a legacy contact or appointing a kind of guardian for their account after they die. But that relies on individual users making that decision and doing that work upfront. And the vast majority of people are not going to do that.

It's interesting that you mention the difference between TikTok and Facebook here, because Facebook's demographic, of course, has become much older over the years. Younger users have migrated to other platforms like TikTok. So given that skew, what does memorialization look like at Facebook now? It must host a lot of memorial pages, I would imagine. Is there even a possibility of it becoming a kind of

digital graveyard in the future. Yeah. And there are other researchers, including Carl Oman, who recently published a book about the data of the dead, who have tried to speculate about exactly the number of dead profiles that are on the site. And that may continue, especially given, as you mentioned, the demographic changes that

But I think what's really troubling to me right now is that we're reaching a point where, of course, large platforms like Meta have now announced that they're not going to invest in content moderation at all. And it is actually content moderation that provides the mechanism for

memorialization. You do need human workers to do the work of sorting out all of the different problems that can occur when you're trying to memorialize an account. And

What is very interesting right now is that one, you have a deprioritization of content moderation and investing in content moderation because it is expensive for companies to pay workers to actually do this work, particularly in a global context for a large platform like Facebook.

And we also have the rise of generative AI, of course, and the decision on the part of many tech companies to create synthetic users. If you look at a lot of the feeds on a site like Facebook at this point, it's completely overrun by AI-generated content. There are bots replying to AI-generated images online.

If these large platforms become essentially digital graveyards and are also populated not just by real users, but by AI, it really changes that relationship to grief, to mourning. And we also see, on the other hand, a push for AI-generated or AI versions of dead people as well as a way to facilitate communication with the dead

I think we're actually at a very interesting point right now where companies do still see a value in terms of the value of dead users as a mechanism for providing data. But I don't know if mourning is a use case and if attachment on

on the part of an individual user to a platform because of the dead relatives or dead loved ones that they have on the site, I actually don't know if that is going to continue to be a point of interest for companies. Maybe we've kind of reached the end of that particular era when it comes to platform memorialization and we're moving into new territory.

where the dead are useful in a very different way. This is another aspect of your book that you explore is that we may need to worry about losing control of our digital selves after we die. In fact, it's already happening while we're alive that it's very possible to create very realistic AI versions of people using their photographs and videos. We've already seen famous singers brought back to life on stage through holograms and digital avatars.

Is there a risk then after death of our data being used to recreate us without our consent? How can we have any control over that? There is a way to sue, you know, companies that use your likeness. Your estate can sue them if you're a celebrity. But for ordinary people,

who don't have a formal estate, who don't have a legal team, who don't have people who are kind of trolling around looking for uses of their image for commercial purposes. I think the scary part is that many people will just have no idea. And so there is nothing right now, also from the perspective of what your loved ones choose to do with your data after you die. If

your loved ones choose to create a chatbot version of you to interact with, there is not a lot that you can do to stop that from happening. And so I think the unsettling part of that for many people would just be not knowing and also having a lack of control and a lack of consent. Yeah.

You also have a chapter in your book that looks at what you call haunted objects. And I've read this with a degree of horror because, again, a lot of what you're talking about evokes the idea of a kind of ghostly presence online, that when we die, we don't just get sucked out of online spaces. There is part of us that remains. And it made me reflect on how many smart devices I have in my home. I have a lot.

and what that might mean in terms of the information I leave about myself through these devices, how it might be used. Yeah, so that chapter is really focused on the material repercussions of inheritance. And even a lot of the people who are the subjects of that chapter, who are futurists or transhumanists, so people who really believe in the power of technology, who think that there's a way to transcend...

the human life form, essentially, through technology. They put a lot of careful planning into creating smart devices that map onto their preferences, their habits. And there's an idea that you can have those patterns and preferences outlive your physical embodiment, that you can kind of make your smart devices, your

metrics of, you know, comfort around your home or within, you know, your daily habits, you can make it into a form of immortality. So the temperature settings on the thermostat and things like that. In one case, I talked to somebody who created a transhumanist smart home and

and had very carefully planned out this entire house with all of these gadgets. But the people living in the house had no idea how to work them. And so he was this spectral presence kind of hovering in from his home in Seattle, watching this condo in Utah and seeing how, you know, awkwardness would kind of unfold as people didn't really know how the smart blinds worked or something. Okay.

But then there's a more literal version of haunting that happens. In the case of one person I interviewed, Jessamyn West, whose father is actually a very well-known figure in the technology world. Tom West had created a very elaborate smart home that she and her sister inherited and

And trying to work out how the house functioned became very difficult as devices began to break down. The fact that software has to be updated, the fact that eventually a laptop will stop working. These things all contribute to a further sense of haunting as, you know, the systems themselves begin to break down and fail. There is something interesting

exciting at the same time, I mean, depending on your persuasion, about the thought of being kept alive forever digitally. It's a very big thing in Silicon Valley at the moment, this kind of idea that we might live forever, immortality. But as much as we might like our digital information to survive us, there are also technical limitations here, right? The hardware degrades, there's digital formats that change.

subscriptions, lapse, all kinds of things can happen, what you call digital decay in the book. I think part of the problem is that it's very uneven. And so the things that you think of as your legacy online may not really be very well archived and may not be preserved.

If, say, the companies that you're depending on all go away. And so the record that you think you're keeping may not be the record that ends up existing in the end, where there may be traces that you're unaware of that, say, get used as part of an AI model and end up showing up in other places that you would not have expected. And the other problem is,

Aside from the kind of digital decay or planned obsolescence, the very materiality of the infrastructures that data preservation relies upon is the environmental impacts as well.

So is there a way for people to keep small archives, to have a form of digital scrapbook that they put together for themselves? Should we move past the idea of preserving every scrap of data that we produce and just

Do we also want to ensure that there are public institutions that are able to preserve the forms of data that will be of use to future historians, for example? Finally, for people listening, they might be wondering how to prepare for the end of their lives online. What are your plans for your digital assets once you pass away? Have you got...

a digital will out there somewhere? No, I don't. I think for me, I'm more worried about the very basic issues of, you know, life insurance policies and who has guardianship over my child if I die, more so than anything to do with my digital footprint. But I would say that people should definitely talk to lawyers about

and not focus on the various digital startups that promise to organize your digital belongings and pass them on to your next of kin. Because as my research has demonstrated, a lot of those companies are not very long lived. If you are really thinking about it seriously, talking to an estate planning consultant

lawyer who is trained in digital matters would probably be a better bet than going to a random startup company. That's very good advice. Dr. Tamara Neese, thank you so much. Thank you so much for having me. And thanks also to you for listening. Next month, I'll be interviewing sociologist Ravi Nandan Singh about his book, Dead in Benares, an ethnography of funeral traveling. See you then.

And that concludes this edition of the Science Podcast. If you have any comments or suggestions, write to us at [email protected].

To find us on podcasting apps, search for Science Magazine. Or you can listen on our website, science.org slash podcast. This show was edited by me, Sarah Crespi, and Megan Cantwell. We had production help from Podigy. Our music is by Jeffrey Cook and Wen Kui Wen. On behalf of Science and its publisher, AAAS, thanks for joining us.