More Science Talk
Science Talk December 27, 2006 -- How to Survive a Robot Uprising: Tips on Defending Yourself Against the Coming Rebellion and Wrangling Komodo Dragons
Welcome to Science Talk, the weekly podcast of Scientific American, for the seven days starting December 27th. I am Steve Mirsky. This week on the podcast, Kurt Auffenberg will talk about wrangling Komodo dragons. Plus we will test your knowledge about some recent science in the news. First up though, robots. Bill Gates is the author of the cover story of the January issue of Scientific American called "A Robot in Every Home." The article is also available free on our Web site, www.sciam.com. To accompany the Bill Gates piece, I spoke with PhD roboticist Daniel Wilson. He is the author of the book How to Survive a Robot Uprising: Tips on Defending Yourself Against the Coming Rebellion. You know, just in case the robots in every home try to take over the world. I called Wilson at his home in Portland, Oregon.
Steve: Hey Dan, thanks for talking to us today.
Dan: Thanks for having
you me Steve.
have got your PhD in robotics from Carnegie Mellon. So, what did you do your actual research on for your doctorate?
Dan: Ubiquitous computing. So, this is the idea – that the regular interface that people have with computers, which is pretty unnatural
ly – sitting in front of it, you type at a keyboard. Humans are not really designed to [for] any of that. The idea behind ubiquitous computing is that all of those computing resources are going to sort of disappear into the woodworks and we are going to begin to interact with computers the same way we interact with each other. And so our specific application was basically smart houses that take care of the elderly. So, the idea is that you spread all the sensors throughout the house and in my case, I used really simple sensors that didn't freak people out, you know, you put a camera in someone's bathroom, but it's not a good thing. So, I had only used sensors that you'll find, you know, sort of like at supermarkets, you know, pressure mats that you can step on, contact switches, motion detectors that detect movement. Spread these all throughout a house—and I am talking about a lot of them—and then collect all this information that goes through a learning algorithm that over time learns patterns of people that [who] live in the environment. And over time it can learn to predict what they are doing, where they are, and who they are, and it can also form patterns for specific individuals, so it can determine if, for an instance, you're losing functionality over time. So, a lot of times what happens is, people [are] getting older and they are living independently and there is no one really to look out for them, and over time, they lose functionality and eventually [it] becomes dangerous for them to be living alone.
Steve: Right, and the sensors can tell, for example, if your gait has changed a little bit and you might have had a mini stroke and don't even realize it. Now, you just brought up the smart house and that's one of the chapters in your book, How to Survive a Robot Uprising, and why is a smart house potentially, though in your, you know, admittedly kind of farfetched scenarios in the book, why is the smart house so dangerous?
Dan: Well, the scenarios are farfetched. I took all those from Hollywood that I will be actually advised as completely realistic and a smart house is particularly sort of uncanny because it learns your every move. I mean, it's constantly collecting data about you and everything you are doing and your friends, and depending on how many sensors they have, you know, it can be filling up hard drives with information all about the mundane activities, every mundane detail of your life, and the sad thing is that most people are surprisingly predictable. As it turns out, you probably take the same route throughout your house all the time and if the smart house wanted to use
an information [in] sort of an evil way, then, you know, you might find a pair of roller skates at the top of the stairs.
Steve: I see. One of the other scenarios that you talk about in the book [is] the threat of modular robots. Where are modular robots right now in terms of development and why would they be so dangerous?
Dan: The reason is that they would be especially hard to stop is because just like in Terminator 3, you can see that losing pieces doesn't really affect the overall robot. It can constantly retain [its] figure and reshape itself because it solves all sorts of problems. And so right now, obviously, modular robots are not nearly as complex as the robots on TV. The major problems are exactly the strength of that robot, which are: how do all these modules communicate in order to solve new problems, and how they do it really quickly? So, right now what we have are modules that are maybe … there was some work at … there was some work at Jurassic Park, which is just a park now, and some work at Carnegie Mellon as well, and in both of those cases, the modules are about the sizes [of] children's block[s], and what they do is they dock with each other so they can sense when they get near another block, and then they can sort of bind into one bigger robot and then, depending on where they want to go, they can form into different shapes in order to get there. So, they could combine into … you have lots of blocks that combine into the shape of a spider and then it would walk, or the shape of an inchworm and then it would inch along, or a snake, which is very similar to an inchworm, except it would maybe sidewind along. They are pretty creepy. I mean, it's really interesting to watch a flat sheet of modular robot crawl along the ground and then, when they find an obstacle, just park down the middle of water.
Steve: And you have actually seen that?
Dan: Yeah, I got used
of [to] that. So, anybody can see them.
Steve: So, there is the possibility – once the modules get small enough and faster enough and there were enough of them – I could walk into a room, sit on what I think is a chair, and the chair reorganizes itself into some kind of an animal-like robot that then tears me into half?
Dan: Yeah, exactly, and that's just thinking about it conventionally. The other really interesting thing about modular robots is that if you were engulfed by one it could actually manipulate you from the outside. So, from the outside it would [look] like, let's say, an egg, but there you are in the inside. Or it engulfs a Rubik's Cube and then it spits it out, you know, fully solved. That's one of the scenarios.
Steve: So, Captain Kirk was always getting robots to commit suicide basically by pointing out logical inconsistencies or paradoxes that the robots endure or the machines or the computers had fallen into. So, will that work for us? Are we going to be able to convince the robots to just kill themselves because they are violating some command?
Dan: I do have a little section on evil robot logic, which appears constantly in popular culture, and so, you know, anybody that reads the book will be inoculated against that. But in my opinion as a roboticist, this is key: the new robots that are coming out are breaking the definition that we have for robotics. So, when you think robotic you think it's logical–black and white, deterministic. But actually, if you make a robot like that and should set
at least [it loose]in the real world, it never works. So, all the machine-learning algorithms and artificial intelligence that in words, it's all probabilisticSo, there is always a different hypothesis that maintains a lot of different possibilities at any given time. So, you are very unlikely to be able to make its head explode by saying something like …
Steve: They do it a number of times, you know.
Dan: … everything I say is [a] lie, I am a liar.
Steve: Norman goes crazy, right. I think there is a lot of evidence now that human character, human sanity, is not just a mental function. It comes from our whole bodies and the way that we interact with our environment and with each other. If you could create a robot that seemed to be, you know, smarter, smarter than humans and really faster with the environment, wouldn't they'd still probably be nuts?
Dan: Probably. It would be considered more than of a survivor. Sometimes you see humans
you have the [who are] mentally retarded in all areas except for one very specific area sometimes, you know, with counting on numbers or sometimes at language or at music, and the way robots are designed, they have everything – all their sensors, all their thinking is all aimed usually toward solving one goal really well.
Steve: So, why did you write the book then?
Dan: The reason I wrote the book actually was kind of [to] make fun of all the Hollywood scenarios. So, everyone is carrying around this cultural baggage. We all know what terminators are. Most people know about HAL 9, targon and "I am sorry I can't do that today," and you see robots scaring people on TV and in movies all the time, you know, and I was at Carnegie Mellon – I worked with robots and I have a lot of friends that also work with robots and I mean all kinds of creepy robots, robots that locomote across the surface of a beating heart, robots that run, robots that swarm, snake robots, you know, all kinds of crazy stuff and that all … we have always doing it for benevolent reasons and so I kind of wrote the book to strike back. And then, of course, I immediately auctioned the book for a movie and now they can make another robot movie. (laughs)
Steve: Right, based on your book. (laughs)
Dan: Which, you know, I am not complaining, but I do appreciate the irony.
Steve: Yeah, when is that movie coming out? [Do] you know?
Dan: I don't know. It's not being filmed yet. It's still
at [in] preproduction, but Mike Myers has signed on to star in it and it's listed on IMDB as coming out in 2008, and that's how cool as I am, and I have to look at my own IMDB page.
Steve: That's pretty good. Well, again, it's How to Survive a Robot Uprising, but really what's in there is a lot of good information about where robotics is right now and where it's going to be in the next few years. Dan, thanks very much.
Dan: Yeah, thanks for having me.
Steve: Dan Wilson's next book comes out in a few months. It's called, Where's My Jetpack? A Guide to the Amazing Science Fiction Future That Never Arrived. For more, check out his Web sites, www.danielhwilson.com and www.robotuprising.com.
Now it's time to play TOTALL.......Y BOGUS. Here are four science stories. Only three are true. See if you know which story is TOTALL....Y BOGUS.
Story number 1: The same gene involved in extreme longevity seems to keep the mind sharp too.
Story number 2: A study comparing physicians and surgeons found that surgeons are taller.
Story number 3: You can make $40 dollars an hour scubaing through giant sewer lines in Mexico City removing clogs with your hands.
Story number 4: A zoo in England is awaiting virgin births or hatchlings of Komodo dragons.
We will be back with the answer, but first, since Komodos were possibly in the news this week, I got in touch with Kurt Auffenberg. He is officially the operations coordinator in exhibits and public programs at the Florida Museum of Natural History at the University of Florida. He is also the son of the late Walter Auffenberg, who was the world's foremost Komodo expert and the author of one of the fundamental works on Komodos called, The Behavioral Ecology of the Komodo Monitor. Komodos are [a] species of monitor lizard. I read that book while editing our big 1999 Komodo article by Claudio Ciofi. I knew that [his] son Kurt had spent time assisting his father. To find out more about wrangling Komodos, the world's largest lizards, I called Kurt at his office in Gainesville, Florida.
Steve: Hi Mr. Auffenberg, good to talk to you today.
Auffenberg: Yeah, it's nice to talk to you, Steve.
Steve: You had some interesting experiences as a teenager. How long did you actually spend with your dad studying Komodo dragons?
Auffenberg: It was right about 11 months. My entire family—with the exception of my older brother, who was attending the University of Florida at that time—spent from August 1969 to July of 1970 on the island of Komodo.
Steve: And how old were you?
Auffenberg: I was 15 and I had my 16th birthday on Komodo.
Steve: And what kind of a party did you have?
Auffenberg: There was a cupcake-type bread thing with a single candle on it, and I believe I was able to listen to the top 10 pop songs out of a radio station in Australia, and that was the extent of my birthday party.
Steve: Really? Field research on Komodo in the late '60s or early '70s must have been pretty Spartan.
Auffenberg: It was extremely Spartan. We were the only English-speaking people on the island, so at that time there were about 400 or so native Komodoans that lived in a village. About a mile or so away, there were several, couple of smaller villages, one very close within a half a mile, and we had the only motorized vehicle, which was a little motorcycle. We lived in a little A-frame type structure just a few yards from the beach.
Steve: Were there more Komodo dragons than the Komodoans?
Auffenberg: Oh yes. Yeah, there were, where my father estimated four or five thousand on the island, and so there were many more Komodo dragons than people.
Steve: What kind of things did you actually do?
Auffenberg: Things like their home range, their feeding strategies, their foraging patterns, their feeding behavior, reproductive behavior.
Steve: So you actually acted as a research assistant to your father?
Steve: What were some of your particular responsibilities?
Auffenberg: I tried to stay out of the way most of the time.
Steve: That's pretty wise.
Auffenberg: (laughs) Well, we had those heart-shaped traps that we would bait with dead animals of one sort or another and then my father and Putra Sastrawan, who is now the dean of science at Udayana University in Denpasar, Bali, would catch them, often with local assistance – unfortunately the people that happened to be at their house at that time.
Steve: Anyone who was visiting got recruited into being a field researcher?
Auffenberg: That's correct! That's correct and they would 9.9 times turn a local person, and then they were all trussed up and duct-taped up and we would bring them into the …
Steve: The dragons, not the visitors.
Auffenberg: Yeah! The dragons, yeah! Aand we would bring them into the shade and, because at that time it was very, very hot and the dragons could not stand too much of the heat. So, we would bring them into the shade and do all the measurements that you would do on any lizard – length and weight – and one of my jobs was to get a clavicle temperature with a temperature probe and also paint a large white number on their side and clip on like a cattle tag onto their lateral skin fold so we could see if we caught them later on again.
Steve: Some people might not know what a clavicle temperature reading would involve, but let me put it this way – you were sort of doing the equivalent of
a dealing with a rectal thermometer.
Auffenberg: It would be rectal thermometer for a reptile. That's exactly right.
Steve: And this is a big, mean reptile.
Auffenberg: Yes, they weren't at all happy at this point.
Steve: I would so, if it did not. (laughs)
Auffenberg: (laughs) Particularly with the rectal thermometer, yeah. And then we would take all these measurements and weigh them and then we would very carefully untie their legs and undo the duct tape or rope around their mouths and we would scatter in different directions, and most of the time, the lizards would run off in the opposite direction.
Steve: But every once in a while they would turn around and look you guys over.
Auffenberg: Yeah, they would say that's the last time you are going to get my rectal temperature, yeah.
Steve: Alright. They are incredibly charismatic animals from my distant vantage point. I did get in a cage at the National Zoo in D.C. with one Komodo dragon, but she was
an incredibly docile, obviously female and, you know, did not smell bad. I was led to believe that they might all smell bad, but I guess it's just the wild ones that really smell terrible according to your dad's writings.
Auffenberg: They do smell bad and that's in the wild, but mostly because they, for the most part, they feed on carrion and a lot of meat and that gets thrown around. They are very messy eaters, and so they don't smell very good if not very pleasant, but they are extremely charismatic. They are extremely intelligent and they are fascinating to observe, both in the laboratory as well as in a wild setting. We had one live very near us – number 19, who attained a length of almost 10 feet by the time we left. We caught him numerous times, but he would go into our trap, eat our food, stand on his back legs—rear legs—and eat the bait and then go down up to the ramp and get back on his back legs, put his front legs onto the top of the ramp, and just race himself out, and to watch that awesome power was amazing.
Steve: Wow, I remember in David Quammen's book Song of the Dodo, he talks about Komodos
are basically scampering up almost a vertical face and it is pretty amazing.
Auffenberg: Yeah, they are amazingly strong. A person
that [who] knows what he is doing can handle a good sized alligator fairly competently, but that's not the truth with those Komodo dragons. They are extremely strong for their size.
Steve: Yes, kids, do not try this at home.
Steve: Especially if your home is the island of Komodo.
Auffenberg: No, and I would do … when I say that after spending a year studying Komodo dragons with my father, when it came time to choose my field I chose land snails.
Steve: Land snails – slow, small creatures.
Auffenberg: And they don't bite, and so although the experience in Asia was wonderful and introduced me to Asia and most of my works have been on Asia since then,
but I knew I didn't want to go into large carnivore[s] again.
Steve: Right! Right! (laughs) And you mean that literally.
Auffenberg: I mean that literally.
Steve: You really didn't want to go into a large carnivore from either end.
Auffenberg: (laughs) From either end, no.
Steve: So what is your land snail study? What have they been like?
Auffenberg: Well, I monitor systematics and distribution and [there are] not
there many people in the world who care about land snails, although there are thousands and thousands of species. My research has been, as I said, in Asia, mostly in the Philippines, Pakistan, Thailand, etcetera, and describing these species, documenting distribution, things like that.
Steve: You can get some interesting insights into evolution through the study of land snails over a relatively small area is my assumption there.
Auffenberg: Yeah, you can and they attempt to reflex all their distribution patterns than particularly mammals and birds or much of their biological research has been. So, when you are looking at land snail distributions, you are usually seeing a much older snapshot and a closer relation to ancient geological events and it's a fascinating topic that has kept me going through years and years.
Steve: And that's just because they are so much less mobile than mammals or birds.
Auffenberg: Right! Right! You know, you get a population of land snails and a mountain range pops up in between them, you know, they don't move very much and the speciation occurs. And many land snails only have [a] range of
that are less than a mile long and that's an amazing thing to try to understand.
Steve: And that may have been the case for thousands upon thousands of years.
Auffenberg: Absolutely. aAnd then, of course, if humans come along and slash and burn their agricultural techniques you could potentially be wiping out hundreds of species that we don't even know exist, yeah.
Steve: Right, right. Mr. Auffenberg, great to talk to you. Thanks very much for reminiscing about those experiences.
Auffenberg: Well, thank you for having me.
Steve: Kurt Auffenberg and his late father co-authored the introduction to the 2002 book Komodo Dragons: Biology and Conservation. One of the authors of the book is Claudio Ciofi, who did the 1999 Scientific American article. In the intro, the Auffenbergs talk about the history of Komodo research. They also discuss one Komodo that they thought they were tracking until they realized that the situation might have been the other way around. That's Komodo Dragons: Biology and Conservation, and Ciofi's article on Komodos is available in our digital archive, www.sciamdigital.com.
Now it's time to see which story was TOTALL.......Y BOGUS. Let's review the four stories.
Story number 1: Longevity gene also keeps the brain sharp.
Story number 2: Surgeons are taller than physicians.
Story number 3: Mexico City sewer scuba divers unclog giant pipes manually for $40 dollars an hour.
Story number 4: Komodos can have virgin births.
Story number 1 is true. A new study out of the Albert Einstein College of Medicine found that a gene associated with extreme longevity also appears to keep the brain sharp. The gene codes for a protein that increases levels of good cholesterol and keeps cholesterol particle size small. So, the common threat in longevity and mental function may be unimpeded blood flow.
Story number 4 is true. Komodo eggs at a British zoo will hatch any time now as a result of parthenogenesis. The new Komodos will carry only genetic material from the mother. The funny thing about the story is that the wild factor is what the media concentrated on for the most part. You probably heard about this story, but the original paper in the current issue of Nature discusses the fact that the situation may be detrimental to the future populations of Komodos in captivity because of the loss of genetic diversity. The author suggests that female Komodos be exposed to males to keep parthenogenesis from occurring. One of the authors of the new Nature paper is the previously mentioned Claudio Ciofi.
Story number 2 is true. A study in the current issue of the British Medical Journal compared physicians and surgeons and found that surgeons are on average taller. They were also better looking; however, actors who play doctors were better looking than real surgeons or real physicians.
All of which means that story number 3 about making $40 bucks an hour scubaing in Mexican sewers is TOTALL.....Y BOGUS. Because scuba sewer divers get $400 dollars a month to wander around unclogging debris lodged in 20-feet wide pipes in Mexico City. You can read more at our Web site and a new story titled "Mexican Sewage Divers Submerge in Murky World." Coincidentally, I know some people in the sewer line inspection business in demand, where the divers get $400 dollars an hour, and this one guy was about to dive in full scuba gear when he tripped over himself, broke a tooth, and gave himself two black eyes where his goggle smashed into his face. You'd think it would be less dangerous than the actual sewer.
Well that's it for this edition of the weekly Scientific American podcast. You can write to us at email@example.com. Check out news articles at our Web site, www.sciam.com, and the daily Scientific American podcast 60-Second Science is at the Web site and at iTunes. For Science Talk, the weekly podcast of Scientific American, I am Steve Mirsky. Happy New Year. Thanks for clicking on us.
Web sites mentioned on this episode include http://www.sciam.com/podcast;www.sciam.com/news; www.danielhwilson.com; www.robotuprising.com; www.sciamdigital.com; http://www.sciam.com/article.cfm?chanID=sa006&colID=1&&articleID=9312A198-E7F2-99DF-31DA639D6C4BA567.