Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Disclaimer: The Parliamentary Library does not warrant the accuracy of closed captions. These are derived automatically from the broadcaster's signal.
Big Ideas -

View in ParlView

(generated from captions) This Program is Captioned Live. # Theme music I'm Waleed Aly. Hi and welcome to Big Ideas, In this edition of Short Cuts from TEDx Sydney - we've got another great selection

environmental ecology robotics, euphemisms, Hugh Durrant-Whyte. First up, robotics engineer who builds robots for cargo, mining, He's a greatly enthused bloke We are, in Australia, apparently, in the world. the greatest users of robots

place in the world to do robotics. And, as Hugh says, it's also the best not a lot of people, We've got a whole lot of space, like a huge mine in the Pilbara and we can run something in Perth robotically from 2,000 miles away in air-conditioned comfort. Good. It's great to be here. thing you could ever want to work on Now, robots are about the coolest and take a look at them up here.

They are great looking guys, to their secret hideouts they carry LA women with red dresses and everything else. talking about robots for a long time Robots, you know, people have been

and that sort of thing. in stories, in film what I want to tell you about today But I think

is where robots are today

in the future. and where it'll be going in Australia in this area. And in particular, what's going on to a couple of my favourite robots, So, I want to introduce you to start with. will have seen her before. This is Maria, most of you And I think Maria's interesting visual incarnation because she's the real first a robot should look like. of what people thought comes from Czech word for slave And if you remember, robotics a worker who will do your deeds and the idea here is you really build of that type of thing. and Maria was really the best example This is Robbie. So, sorry, excuse me. of what happens in robotics. Robbie is the '50s version You can see he's a bit more clunky, Chevrolet, that type of thing. he's a bit more like a 1950s You've got the sensors in his head. the sort of thing that happens, But this is, again,

in robotics. what people like to think about but they will do your bidding Not just that they're slaves sense of fear that goes on in this. and also in some sense, there's a In the sense that, in the end, the uncertainty that people have Robbie is an important part of the Forbidden Planet. about living in about robots. So, that's what we think like what you might imagine. But it turns out things are not quite And in particular, Australia itself - a long way from anywhere - big, empty, the world to do robotics. turns out to be the best place in And the reason for that in Perth. but from 2,000 kilometres away not from up in the Pilbara, This mine is now run it is becoming completely automated. it is not yet there - it is becoming - And essentially, again, in the East Pilbara. in fact this is the West Angelas mine This is the Pilbara, Some of my earlier work in this area. Here's another very good example. that you can put robots to scale and the kinds of applications but it does show you the kinds of operation Now, it may be a particularly simple The robots run the entire operation. completely automated. So it is essentially with each other and so on. how they should coordinate how they should put them down, where they should take them, which boxes they should pick up, and they use that map to figure out of what's on the ship And these robots basically take a map the terminal in Brisbane at all. So there aren't people on is actually run from Sydney. And in fact the Port of Brisbane on the terminal. There's about 36 of them You can see a whole bunch of them. So, here is one here. entirely run by robots. The Port of Brisbane is in fact this is the Port of Brisbane. Here is one, a couple of examples of this. So let me show you And that sometimes surprises people. and users of robotics in the world. one of the largest developers Australia has actually become to put these kinds of robots, where you'd like to be able of big, empty spaces has all these characteristics is that because Australia one should realise from this and I think one thing in Australia in marine, in defence areas and so on in cargo handling, in agriculture, robotics going on in mining, So, in particular, there's lots of bit about some of those examples. And I'm gonna talk to you a little for robots. is we have lots of great applications

They have automated 300 tonne trucks and the processing plant, that move between the digger they have automated survey systems. they have automated drills,

automated trains They are now just talking about all the way up to the port. to ship the ore from the mine why this is a good idea, So you can imagine

very dangerous kind of job this is a very dirty, into this kind of application and being able to put robots

and use them for basically process itself more efficient for trying to make the mining is an obvious thing to do. And using that information up in the mine, so you don't have to have geologists and indeed, in principle, they can in fact live remotely while they're staying at home. they could be part of the operation

application. Here is another quite different This is underwater. then its exclusive economic zone, If Australia is big and large,

Australia is truly enormous. its marine environment around monitored for environmental purposes And very little of it is mapped or with the flora and fauna. or understanding what's going on This is one of our robots, going down to depths of about 1km it's a submersible capable of it has sensors on it, but more importantly, it is able to build accurate maps, in three dimensions. this is in fact Ningaloo Reef to control - It's able to use that information

to go on expeditions all by itself. it comes back eight hours later, You simply drop it in the water, but with an understanding not just with a map are different types of flora, of where in that map different types of habitat

that you need and other sorts of information that environment in that sort of way. to monitor and control of course, Now, this is a different environment, from things like mining and cargo handling. It's quite complex because there are no obvious structures out there

and we'll return to this when we try and understand how you actually make a robot understand what's going on in its environment.

So that's a nice example of what can be done. Here is another example. If you're on the way, you're driving down to Canberra, we actually operate a flight test facility between Sydney and Canberra. And there we fly robot aircraft. Not just one robot aircraft but up to four robot aircraft all at the same time.

Again, there are many applications for these types of robotics, particularly in areas like agriculture and defence and so on and they have the same sorts of issues. If you look at these robots here, they've got funny little noses on them because they have lots of different types of sensors in, they've got to use that sensor information to figure out where they are in that environment, what it is that's on the ground, and in fact to cooperate, to work together, to figure out how they should be mapping features or doing other problems which otherwise would be quite hard to do by flying a manned aircraft around there. So while they're in the air, they decide which direction they're going in, which features they need to look at on the ground, how they should coordinate with other aircraft and so on. So, you can see the skill base that you need in these types of robotics. And again, Australia is an ideal place to run these sorts of things. You wouldn't try to run this in a densely populated country in Europe, that's for sure. So here's another interesting example. This is a start-up company that came out of my lab,

these are robots that drive around a shooting range for the SAS And what happens is the SAS are allowed to practice shooting them. And they travel around at a fair clip but they also cooperate with each other

so you can have one person that you'd like to be able to shoot and other ones, perhaps, you don't want to shoot - and you can see where they've been shooting at the walls there and that sort of thing. So, basically, if you like, it again brings in these things, there's a bit of sensing, there's reaction to what's going on

there's the way they cooperate and so on. And it's a successful company - it won last year a $50 million project to supply these robotic systems to the US Marines. So this is the kind of technology that's involved in these areas. And just to show you that there are also some different things - this is - and you'll see an exhibit outside of a slightly newer system - but this is a media art... robotic system. Here we have two robots, Fish and Bird. Again, there are sensors in the environment, they detect how they work with each other and how they interact with the audience, OK?

And then they write different love messages to each other, depending on your behaviour.

So, what can we say from all of that? What is a robot, in the end? Well, what it isn't is what I showed you at the very beginning. It's not a humanoid, OK? It's not something that looks like a human.

It's basically a system that connects a computer to the real world. And it connects those two things together in two ways. The first way is through perception - that is the sensing, OK? Looking out there, trying to get some understanding of what the real world looks like

using a variety of different sensors. The second part of a robot is action. That is, when I have that internal model as to what's going on, what do I do with it? Do I pick something up, do I drive in a particular direction, do I fly, and so on. And the third very critical part, which is the learning. Which is, as I do this, I look and I act and so on,

I must improve the model I have of the world and my ability to work in that world.

Alright? So to really change the way I behave as a consequence.

And we roboticists think of robotics not as a machine but as that intelligent connection between perception and action. And it could be in any form you like - an aircraft, a humanoid, whatever. But it's not about the machine,

it's about that perception, that action and that learning. So I'd like to say a little bit about learning and just give you really what the state of the art's thinking is

in how you get a computer to think, OK? So let me start by showing you a picture. This is the sort of problem that we pose ourselves - how do you build a model of a tree? OK, here are four trees. And what people did, I guess, for a very long time is really to build a geometric model of a tree. And you can see, that's a pretty hard thing to do but what we've learnt now is that you do not build models of trees, you simply look at a lot of trees, OK? This is basically how you learn what a tree looks like. So, for example, if I had photographs of trees and those photographs were, say, 100 by 100 pixels, then I could think of this

as a big space of 10,000 dimensions - 100 by 100. And then every possible picture I could ever take of anything would be a point in that space, OK? And then a part of that space is the space of all pictures of trees. OK? That is a tree. That's the definition of a tree. It's all my experiences of all trees I've ever seen. And in the world today we have so much data capacity that that is exactly how we represent trees. Just every experience of a tree that we have ever seen. Let me show you a second example. We often talk about lemons in my lab. And the reason we talk about lemons

is that I can say the word lemon to you and you can envisage it in your mind. You envisage the smell, the taste, all the perceptual properties that you have about a lemon. But you don't have a model of a lemon, you simply have the experience of all the lemons that you have ever tasted. And when we've summarised all that, whether I'm a smell sensor and you're a taste sensor, we can say the word lemon and we can actually imagine what a lemon looks like and all its sensory experiences. So our representation of a lemon is simply all we know and all we have experienced, perceptually, about a lemon. And that's the way we now represent these things. Let me show you some examples. This is a robot that flies along and finds particular types of weed - prickly pear or whatever you might want. Of course, every weed you ever see is different but it finds them reliably no matter what they look like because it's always had lots of experience about different types of weeds. And so this robot can fly along, find particular types of weeds and spray them one by one, OK? Instead of spraying the entire field. So this is one application of this type of learning algorithm.

Here is another. We're actually, in this case, trying to detect whether we've got high-rich iron ore or low-rich iron ore and in this case we take a number of different types of sensors - vision, laser, and also what's called hyperspectral sensing -

and we automatically detect where we should be digging and what we should not be digging. And yet every time we look at these different ores, they will be different. They have simply learnt that by looking at lots and lots of them. OK, it's a data representation. So, our concept of learning now is no longer about building models and computing things. Our concept of learning is simply memory.

And that's the way robots are learning to learn. There's an example here I want to talk about. You will have seen it out here. This is actually a research tool about this big problem of perception,

it's a robot with lots of different sensors on, it builds up big pictures for what's going on, you can see one over here and then one slightly bigger representation of a tree down here and as a consequence - sorry, we go back here - you can use this information to label things like trees and cars simply because it's seen that before.

Maybe if we could cue what's going on outside at the moment, we'd see whether this technology actually works. Um, I have it on this screen but you don't have it on the main screen. Can we try it again? There we go. This is happening in real time outside. Maybe the people outside could turn around and have a look at me. Alright? Or move around in some kind of way? And here you can see all the different types of perceptual sensing

that that robot has.

And it puts those all together to figure out what's going on, whether that's a person, whether that's a robot, whether that's a structure and so on. So it uses basically a memory of everything that it's seen to do that level of classification. If we could switch back, that would be good. So I encourage you to look at that when you're outside. This just shows you basically what's going on. Here you can see the laser, the cameras, IR,

different types of sensors that are all integrated together. OK, what about the future? This is one of my favourite systems that's out there at the moment. It kind of shows you what happens when you really put this perception-action loop together. I would argue that this particular example from DLR in Germany is probably the biggest integration of this perception-action loop

because it actually does use its cameras to figure out what's in front of it,

to figure out how to manipulate that in a fairly dexterous manner and it has sensors in its fingers to measure the force and tactile sensations and things like that. It has enough dexterity then to look, to find the water, OK? (Audience laughs) OK?

You can see, some of this is made up as you go along but nevertheless it combines that element of different types of sensors sorts - how you touch, how you feel, how you see - to some very, very dexterous actions that are going on. And this is kind of the way things are going, this whole perception-action component. People are also heading in the direction

of things perhaps looking more humanoid

or acting in some sort of anthropomorphic manner.

The picture at the top is Professor Hiroshi's android.

It's actually an android of his wife, believe it or not. That's slightly scary, I have to say. But nevertheless, it tries to bring a lot of those action components together. I think, personally though, some of these other ones are a little more interesting. There's a robot fish you can see up there. This is probably one of the smallest robots out there at the moment, it's a robot dragonfly - and it actually does have sensors and actuators and everything on it. My very favourite of all over there is the robot gekko, OK?

So it climbs up walls and things like that using, again, all these different types of sensing capabilities. So people are indeed thinking now, now that we have a better understanding about machine learning and about the applications of these sorts of things, what can we actually make them into, what can we produce that people

will actually see in their day-to-day lives. I think one thing though that also concerns robotic scientists is what we might end up with. I think that was well put by Julian earlier on. It's not my intention, of course, as a robot scientist to build these sorts of things but there's no question at the end, we have to be conscious of what we're doing. How we use the technology, how we put it out there in the field, how we engage it as humans

and how we always, as humans, engage technology.

Thank you. Hugh Durrant-Whyte, robotics engineer. Next up, linguist Kate Burridge on euphemisms - those smoke screens, those invisible words we use when we want to obscure what we really want to say. We use them out of extreme politeness or fear, or prissiness or when we're just being downright slimey.

or to borrow from Orwell, 'Words to make lies sound truthful and murder respectable.' This talk is all about that nasty business of how we actually talk about those things we really don't want to talk about those things we don't want to evoke too vividly. So we have at our disposal a range of linguistic deoderisers,

smoke screens, fig leaves - these are euphemisms. They are all about taboos, of course. Those things that go bump in the night. They're about politeness and sometimes they're about skeletons in cupboards. But no matter what population group you look at, past or present, you'll find them. In classical times they were those nasty little verbs like 'die' and 'kill' and these were avoided. Instead, people just 'curled up' or they 'went to sleep' or 'on a journey' or they were described as simply 'having lived.'

In the 19th century, Victorian moral code prevented those in polite society from uttering those dredful words like 'legs,' and 'trousers,' 'underclothing.' Novels at the time were full of invisible words like 'mustn't mention 'ems,' 'indescribables,' 'unwhisperables.' But you know, we shouldn't grow too smug it wasn't that long ago when we had 'smalls' and 'foundation garments.' And don't forget those curiously named 'abdominal protectors' on the cricket field. Another invisible word of the Victorian era was the monosyllable. These days it's the 'c-word.' Now, I know for many of you euphemism will be associated with deliberately befuddling language. Weasel words and puffery -

the sort of language that turns the loss of human life into collatoral damage.

Or lies into terminological inexactitudes or as George Orwell once famously put it 'Language designed to make lies sound truthful, murder respectable, and give an appearance of solidity to pure wind.' But you know the underlying motives are not always malign and besides there will be those occasions where you just dont want up-front language. Say you want to swear but it's not the right moment for a full-blown obscentity. Well euphemism provides the out with remodelings like 'sugar,' 'shoot,' 'fiddle-faddle.' So euphemisms. What they do is shield us against what's embarrassing, what's feared, what's disliked, what's unwelcome. And sometimes they're used to upgrade and inflate. Now prairie oysters and fries -

they sound a damn site more appetizing than calves testicles. And at this point,

I'm really glad you've got the slide to look at and I don't. for any culinary unmentionable. It nicely dodges references to which bits actually get cooked

and in the same way we've got Welsh rabbit. Well, that's already a diguise for cheese on toast but welsh rabbit becomes tastier when it's turned into Welsh rarebit. But you know even potholes are transformed via euphemistic magic to 'pavement deficiencies.' And life insurance - well that's insurance for when you're dead. Euphemism tells it how it isn't

and we have plenty of examples of visual euphemisms too.

Of course fig leaves used to hide the genitals of statues and in the Victorian era again occasionally these frilled pantalettes would modestly hide the limbs - remember legs could not be properly mentioned - the limbs of pianos and tables.

Public hair was airbrushed out of soft porn photographs until the 1960s. Our modern supermarkets, they're dripping with modern euphemisms -

the waxing of fruit, colouring that turns the meat even redder,

beautiful packaging, it's all cosmetic.

It says I'm tastier, I'm bigger than I really am. And... in the case of the visual euphemism I suppose... it's all dishonest but you've got still photography, television, film -

They're superb media for the visual euphemism. Where there's romance in the pasta sauce.

Where there's sedution and intrigue in the cup of instant coffee. And romance, sometimes a bit of nostalgia, poetry in that hand-crafted potato chip. As I said it's all dishonest. No euphemism, even the well-intentioned ones say it as it is. It always involves a kind of double think.

You know something that's tabooed can be acceptably spoken of by using a euphemism and not a tabooed word. 'I'm going to the bathroom'. That's fine. 'I'm having a piss.' It's not. (Laughter) It's as if the obscenity lies in the actual words themselves and not in what they refer to. We even call them dirty words and some of you might have had that appauling experience of having your mouth washed out by soap.

And such is the potency, the pungency of these dirty words that they can contaminate others

often bringing down innocent bystanders too The old word for rabbit, 'cunny', well, it 'bit the dust'

precisely because it sounded too much like the female body part. So people stopped using it or they remodeled it to 'bunny',

or coney as in Coney Island.

Weather cocks were transformed into weather vanes, and haycocks into haystacks And 'feck' that old word for efficiency -

well, it had nothing to do with the dreaded F-word but that didn't save it. They are such attention grabbers, these words. Even when we jumble up the letters they scream out at us. So how do you go about creating or disguising disagreeable reality?

Or at least putting a bit of a good spin on it? Well, one good strategy is nebulessness. Say you want to make reference to one of these perculiarly named private parts. Or something like the groin provides the fig leaf. It's imprecise location makes it a perfect euphemism for anything

in the general vicinity. The problem is, it will, over time turn into the direct term itself. and it can happen that these vague below-the-belt words turn into different body parts in different dialects hence the ambiguity of slang terms like 'fanny' and 'pratt.' No-one's quite sure what these terms refer to and we're all a bit too polite to ask.

It's actually a linguistic phenomnenon known as 'gential flip-flop.' That's its official title. And its a source of considerable confusion that any American amongst you will know if you've tried to buy a fanny bag say in Britain or in Australia. Now another really good strategy is long-windedness. So traffic jams - they're so much more bearable when they are 'localised capacity deficiencies.' And I recall too, reading the hospital records of one patient. This patient was described as having 'failed to fulfill his wellness potential.' It was 'a negative patient care outcome.' So under the cover of words we can tip-toe around any sensitive topic and the more words the better.

But, there are plenty of grammatical strategies too. You might remember that 1999 was 'The Year of the Older Person.' How old is older? Without any comparison given the edges are blurred a little. And suddenly 'older' is not as old as 'old.' Of course advertisers make great use of this construction. 'Our oranges are sweeter.' what? Sweeter than they were before? Sweeter than somebody else's oranges? Sweeter than lemons, perhaps? You can always provide the comparison. And that's what gives the advertiser the escape hatch.

It's a cunning devise, but it's... well... with the 'fuller figure' in the fashion industry, it's used a lot. But anyway - our thoughts I suppose are not as, well, they're not as manipulated or as easy to manipulate as George Orwell's Newspeak suggested. None-the-less, there is plenty of evidence to suggest that language can work to influence our memory and perceptions. So just the right assemblage of consonants and vowels these can conjure up a certain image, they can send a certain siugnal. OK. What's smaller and cuter - a freep or a frope? It's obviosuly the freep, isn't it? The cute little freep. The dear little freep. But back to euphemism. The tragedy of all this is that euphemisms are doomed, because the taboo senses continue to bubble away and reassert themselves and the ephemistic qualities will diminish. The next generation will grow up learning the euphemism as the direct term or worse. Take smells. English has a bucketload of words for bad smells, few words for nice smells

and few, if any, genuinely neutral ones. Even the word smell has a bit of a whiff about it. Helen Keller once described smell as the fallen sense. It's an area of social taboo and this drives the shift from sweet smell or agreeable smell to disagreeable smell. And what you'll find is the longer the word has been in the language so the more disagreeable the smell. So stink and stench are the oldest and they're the smelliest. The most fragrant are the French words. French is long a source of deoderising language for English. But just how fragrant these words are depends how long they've been recruited. So something like odour is the oldest French word and it's already started to fester. Might have noticed. So taboo areas will perpetually generate this sort of narrowing of meaning, but also deterioration. It's hard to imagine that the word toilet was ever a sweet smelling euphemism.

But it was and it was from French, from the dimunitive of the word for cloth, 'toile'.

So, what this chronic contagion does, is cause a flourishing of vocabulary items, as speakers seek the less offensive words to speak the unspeakable.

So, latrine turns into water closet, WC, toilet, bathroom, washroom.

And you will find that the more severe the taboo, so the longer the chain of replacements. English has amassed an astonishing over 2,500 words for human sex organs. Now that is lexical richness. Don't worry about words for snow or camel in other languages. But I should say too, that occasionally, very occasionally, a euphemism will endure. And intact. So, we've been 'passing away' or 'passing' since the 1300s. And we've been 'sleeping with' each other since the 10th century. Now these are long lived euphemisms.

Just how they managed to sneak through unscathed, I'm not entirely sure. But clearly, linguistic familiarity doesn't always breed contempt. Anyway, what would life be like without euphemism?

What would life be like if we all said what was exactly on our minds and in the plainest and most explicit of terms? Hell on Earth.

I mean, it might be initially attractive to think of a kind of no frills, say it as it is, value free language. But the reality is in fact no language ever dittoes reality. It always gets in the way,

by their very nature words and constructions will hint, they'll insinuate, they'll suggest. It's not just politicians, it's not just advertisers,

it's not just linguists trying to influence an audience.

We are all guilty of embellishing information just by the constructions and the words we choose. So, yes, be very wary of language,

especially when it tries to lead you by the nose. But remember that most euphemisms are there to make life easier for us, and they're central to what makes us tick. Thank you. That was linguist Kate Burridge at this year's TEDx in Sydney. Next up, ecologist Angela Moles on weeds. She's talking about the introduced weeds that turned out to be not such great ideas. The biggest problems being what they did to native biodiversity. The control and eradication of such weeds costs the economy about $4 billion a year. But Angela puts up the proposition that these introduced weeds may not be all bad. And we shouldn't, as ecologists and farmers, be bent on eradication, because these weeds can evolve into uniquely Australian species.

So what I'd like to do today is tell you something that I think might change the way you look at the plants in your garden. Sorry, the weeds in your garden, the weeds on our roadsides, and even the weeds that we see in our National Parks. Now my story starts over 100 years ago,

when people generally thought that introducing plants and animals from different parts of the world to Australia was a really good idea. In fact they thought it was such a great idea that they'd formed acclimatisation societies. And these are, as far as I can tell from the pictures, mostly groups of men with the most amazing whiskers. And they worked tirelessly to enrich our native fauna and flora. Perhaps my favourite example of the work of acclimatisation societies actually comes from New York, where the American acclimatisation society decided it would be just lovely if they could get all of the different varieties of bird species mentioned in Shakespeare's works established in New York's Central Park. That's how whimsical they were. So they did this, and one of them was a roaring success.

The European Starling became a really serious invader. Now, as a result of the actions of the acclimatisation societies, but also from introductions for agriculture, for our gardens, and also all those accidental introductions on peoples socks and boots over the years, we now have over 3,000 species of introduced plants

growing and reproducing happily in Australia. And lots of these have become a really noticeable part of our ecosystem. I'm sure you're all familiar with them, and I'm sure you're also familiar with the fact that introducing them wasn't actually such a good idea on the whole. They in fact cost Australia over $4 billion a year, partly in lost revenues, but partly in the cost of trying to control them. And perhaps worse than that hit to the hip pocket, is what they do to our native biodiversity. World wide, introduced species are the second biggest threat to native species, and that's second to just taking a bulldozer and just driving it through your native species. So, they're pretty bad. And if you talk to most ecologists or conservationists about weeds, you'd get an answer that wouldn't sound that out of place from these guys.

DALEKS: Exterminate. Now what I'm going to do today is move a little bit beyond that, because talking about all the downsides of weeds was just not not very new, and this is TED. So, what I'm going to do instead, is point out that these guys with the whiskers, actually set up an amazing replicated experiment for us. That can really help us understand evolution. And in particular what they did was without realising it, they followed a near perfect recipe for creating a new species. And what I'm going to argue today is that it is now inevitable the introduced species will, over time, evolve to become new uniquely Australian species. So before we get to the part where the Lantana's in utes, driving around with those Australian flag things on it's windows, I have to explain how I get to this conclusion. And to do that I'll start by telling you how new species are usually formed in nature. And I'll do that with this group of bird species, they're Finches from the Galapagos Islands. And they are, of course, famous for shaping Charles Darwin's thinking about evolution. So, to make a new species you start out with a population of your organism, be it birds or snails or plants or whatever. And you end up with that population in two or more geographically isolated patches. That could happen from having one original range that gets divided, through things like sea level rise, or mountain building episodes. Or if your taxon doesn't disperse very well, like these birds, you could just end up with them isolated on different islands, that they can't move between very easily. However you do it, once you've got them growing in different places, the door is open for them to adapt to the local conditions. So in this case, the important difference for the birds

between the islands, was about the types of food that were available there. So, over the generations selection acted on the beak form of these birds, to make them more efficient at eating the foods

that were available to them. Now at this point we have populations that look really quite different to one another. But most ecologists won't accept them as new species, until you've gone one last step and shown that they actually don't breed well together. So you take the individuals from those different populations, put them back together and you find that they either can't or won't breed to make successful babies. OK, so that's how it usually happens in nature for making species, and what I'm going to show you is this is exactly what we've done with our weeds. So, first step is to establish geographic isolation. And those guys with the whiskers did that in spades, because they've moved these things hundreds or thousands of kilometres around the world. And now that those plants are growing here, they're experiencing different conditions to what they got at home. Partly because they're interacting with a whole suite of different plants and animals, so different things pollinating their flowers, different microbes, different things dispersing their seeds or eating their leaves. And a whole flora of different plants for them to compete and co-exist with. And that's just the biological environment, there's also the physical environment to consider, and just to take an extreme example, imagine how shocking it would be for a little herb introduced from Scotland, arriving in NSW. So, lots of reasons for these to be undergoing selective change.

But, evolution's a slow thing, right? These things have only been here for 100-150 years, surely we won't have seen anything yet, right? Well, actually the evolution ecologists have been showing more and more lately, that evolution can happen much more quickly than Charles Darwin ever imagined. It's not uncommon for it to happen in just a few generations, or within a decade or two. So we thought, that given the strength of the selective pressures

that are on these introduced species, we might actually have already had a chance for these things to adapt to Australian conditions. So we went out to test this, and one of my Master's students, Joanna Buzzwell, went to the nation's herbaria. Which are basically libraries full of pressed plants, and as you can see from this Plantago, which is 130 years old, they keep very nicely, so you can just go up to these specimens, and measure what the plants looked like at different times through history. So we did that on a whole bunch of different species, and lo and behold, 70% of the plant species that we measured, had undergone significant change since they've got to Australia. And those just weren't quibbling little statistically significant things that you had to look hard for, these were substantial changes that really affected the way the plants were growing and interacting with their environment. So for instance, this clover that we've got on the screen here

had become 60% shorter every hundred years since it had got here. And just in case you're thinking, oh, well of course it's shorter,

it's drier here than where it came from, these plants were all growing in NSW, so it's not that they're in a different environment through time, that's just them changing since they got here. OK, it wasn't just plant height that changed,

we also had species that dramatically changed their leaf size or their leaf shape.

So, no doubt about it, our introduced species, our weeds, are adapting to life in Australia. Now that's got a good news and a bad news aspect to it. The bad news is, holy cow, our introduced species are actually adapting to life in Australia, so they're getting better and better at living here. That means they're getting more and more invasive. So, that's hardly good news. There is good news though, and that is we measured all these different plants, and they were from a whole bunch of different species, different families,

and on the whole, they were remarkably good at adapting to this changed environment that they'd encountered. And I'd like to think that there's this little sliver of hope that our native plants might actually be better able to adapt to the coming climate change than we feared. OK, so back to evolution, we have established geographic isolation,

and these plants are already adapting to the conditions in Australia. So, we're a long way down this train towards these things already becoming seperate species. The only thing we still have to show to prove that they were already seperate species, is that they can't breed successfully with their source populations. Now we're actually working on that at the University of New South Wales at the moment, but it's going to take me a few years to get an answer. But my thesis today isn't that as of the 26th of May, 2012, they're already new species - although I suspect some of them are already -

but my big point is that if things go on like this for enough time, it's inevitable that they will become new species at some point So the only question really is, are we going to give them that time? And a lot of us would like to think, 'No, we're going to eradicate them, right?' Well, people are working really hard on controlling introduced species in Australia and they're doing a really great job. This stuff's back-breaking work and a lot of the control programs are really, really benefiting our native species. But the resources that we have for weed control are not infinite. There are whole areas of Australia

where there's no control of introduced species going on and whole lists of species that we're not even trying to control so they'll be here into the future. And even the big bad weeds that we'd love to see eradicated, the bad news is actually eradicating species from a country as big as Australia is next to impossible. So whether we like it not, and, really, we don't, if you come back to Australia in 100 years, 1,000 years, 10,000 years there will be introduced species here. And if they keep on adapting like they are,

to our environment, at some point they will have diverged enough

from their source population that they will become unique Australian taxa. And that's got some really interesting implications for our management of these weeds. Perhaps my favourite implication is the fact that if they do become unique Australian species, things that only live here in the planet, then at some point we might want to actually turn around in our management of these things and stop actually trying to exterminate them and actually start trying to protect them instead. Now I know that's not going to be an easy thing for a lot of people to accept -

I imagine there's a lot of people sitting there with their arms crossed going, 'Yeah, right, you can say what you like, I am not accepting lantana as an Australian.' And fair enough. I've got three pictures for you though. This is my son, Sam. He's a second-generation Australian.

He was born here, speaks with an Australian accent. Most people would have no trouble accepting him as Australian.

For some reason we don't extend the same favour to the clover. (Laughter) It's been here for 130 generations. This is the one that I showed you is getting shorter through time so it's growing with an Australian accent and it's definitely part of the Australian environment. Lots of Australian animals eat this stuff for breakfast.

But for some reason we're not accepting this as Australian yet. Is that because it was introduced here by humans? Well, I don't think so because our third species is the dingo. That was introduced here by humans about 4,000-5,000 years ago. And as long as you're talking about a pure-bred dingo, most people are pretty happy to accept that as an Australian taxon, even an iconic Australian species. And National Parks and Wildlife Service do spend their time trying to protect those pure-bred populations. So will we ever accept introduced species as part of Australia? Yeah, I think, like the speciation process itself, it's just a matter of time. And, you know what? It might be a bit sooner than you think. Thank you. That was environmental ecologist Angela Moles. And last in this edition from TEDx Sydney, a two-handed presentation from two architects, Gerard Reinmuth and Anthony Burke. They're the co-creative directors of the Australian pavilion at this years's Venice Biennale of Architecture. This is a punchy and moderately self-deprecating presentation, pretty much what you might expect from two architects. This is the architect that we're all supposed to be, of course, and, you know, this is Howard Roark, mid-century, standing there - how do you match up to that? I'm only 5.5 feet. I can't get anywhere near that. So Howard Roark, with his autonomous building, sitting there in the city as a sort of - it is the product of modernism, the product of individuality, the product of rational order in the world. The thing about that is that that's not how the world works and we were sort of convinced that that was how the world worked for a long time, and I particularly blame the Bauhaus for this and Plato, who I think actually is a bit of a jerk for doing this to us. So where we got to, though, this is sort of the way the world actually works. This is a quick slide to show you that the world we've discovered through network theory and these types of things in the last 30 years really the sweet spot for us is right in the middle.

There's sort of highly regular networks on the one hand and chaotic networks on the other hand but what network theorists have found out, actually a little bit of dirt, a little bit of chaos, a little bit of disorder actually makes the world go better.

We need that disorder, we need that dynamism to make the world generate. So to explain how it all went a bit wrong with architects because Le Corbusier, one of the most famous architects of the last century, said 'Architecture is the masterly, correct, magnificent play of masses brought together in light.'

Very platonic idea - platonic solids, cylinders and cones, and spheres and things like that, rendered beautifully in light.

And he did some lovely projects. He did this house - there he is, looking very serious on the left. He did this house on the right which is a classic plantonic project. It's like the whole universe crammed in that building. It's a building for measuring the universe. It's so important, the garden, of course, sits around it so the building can be understood as the centre of the universe.

Now this is all fine until the upscale. So what happened then is you may or may not guess that city but the island in the bottom right's a bit of a hint for the Francophiles in the audience. And so of course the idea was that as you can control a single house you could also control a whole city. And it was actually very well intended. It was all about health and social equity and being outside and fresh air, and all that sort of thing.

But of course the problem was this single idea had been upscaled. Now the problem we've got as architects, we've been dealing with that ever since and you haven't forgiven us for it. So the thing is, the doctor he did that house for, if you think, that doctor, at that time, was probably using leeches, all sorts of pills and potions you certainly won't get on your Medicare rebate nowadays. But the thing is doctors somehow advanced. You let them have ER and Grey's Anatomy. (Laughter) We're still struggling with that image of that city. So this is a quick example about how you approach a city, I think, if you're not coming from the top down. If you're not that rationalist sort of image. There's something I did with some students at Berkeley when I was there before coming back to Sydney, called 'way fi' and the task really was just to understand the city is already there. It's a pretty radical idea. You know, the city's already there.

If you actually go looking what you find is that in San Francisco, for example,

there's 400 manhole covers in any one block in down town. So rather than actually having to put more stuff in the city we just said 'Why don't we repurpose those manhole covers?' They can still be manhole covers but as a way-finding device to find your way elegantly around the city, perhaps we should just combine them with a mobile phone and an RFID chip and that will be a nicer way

to tell you if you're getting hotter or colder as you move towards your destination. So, for us, architecture is not actually the masterly correct and magnificient play of masses brought together in light,

rather architecture is a mongrel, not a thoroughbred. And that's kind of the big message I want to leave you with today. Gerard's probably going to disagree with me in five minutes. So, architecture is a mongrel. And it's this idea of complexity that we actually want to embrace, we think designers have been trained - creatives all over the world in all sorts of domains and disciplines have been told that creativity is something that you kind of buy

at the MCA gift store. And we don't believe that. It's too messy. And it's too rich. And that diversity's what we have to strive for. So we're going to talk about a communal creativity, and we're going to start with a guy who's certainly influenced us, Leon van Schaik who wrote this book called Spatial Intelligence. And this came out of Gardner's seven types of intelligence that, of course, has been contested. But the important thing about that list is linguistic and mathematical sit at the top,

the stuff we all do to get through school

and high school and into our jobs but of course there are other types of intelligence and spatial intelligence he thinks is very important. Because the mistake we all made is when everyone was manning up in the 19th Century to make professions and architects want to sit along doctors and lawyers. It's like, what do we call our world? What's our discipline?

And the idea was at that time they encapsulated building, said, 'OK, all the stuff around building,

we'll make that the stuff that means you can call yourself an architect.' And Leon's contention is that was a complete stuff-up because actually, what we did in that moment is gave away the one thing we do really well, which is spatial thinking. So all of a sudden, architects are trapped into a paradigm of building rather than thinking about the complex spatial problems we have in our cities.

And this is linked to a message from Ken Robinson - Sir Ken Robinson, best TED talk ever. Am I right? Bastard. And he gave his first talk in 2006 when it all went online, so we're only all destined to be second after this. But he made the comment that we're educated out of creativity.

And our talk is in a way, raising a more specific version of that idea, is that we are all educated out of being free to use our spacial intelligence.

So, here's a quick example. This is actually my son, Alec,

playing with one of those corporate toys you get after a conference. At the kitchen table last Saturday, as I was sitting there thinking about this talk. Actually the miracle there was that it was breakfast, and I was actually thinking. So that was a whole other thing. But the thing that I was noticing as Alec was moving this toy around his hands was that there is absoloute sort of spacial fluidity with the way that he was moving. This toy in rapid succession went from a car to a superstructure to a rocket launcher and a giraffee and all the rest. And as Alec moved it around he didn't stop to think about the consequences of these things, but the point was he was just sort of excersising this spatial thinking without thinking about it. That's, I think, the sort of thing we're talking about. We as adults, we get trained out of that. It's a really hard thing to do, actually. When was the last time you really found yourself thinking about the world around you in those sorts of plastic terms.

When did you go outside - out the front here, for example,

and think to yourself, 'Oh yeah, I think maybe I should move the door. This is a great space that I could do something and like have a trapeeze artist. And this is what kids do very, very naturally. The thing was, he was just doing it under the table at the time,

which sort of gave it a whole degree of difficulty, which was about an 8.5, I reckon. But it's a reminder there, which is really that, um, kids have this natural capcity to see the world in this completely fluid and spatial way

and we have come habitualised as adults

into thinking the wrong way about our environment. Now the best architects, we argue, therfore are big kids and one of the biggest kids of all was Jorn Utzon and his son Jan once told me when Jorn got back to Denmark he got second all the time in competitions and that was very annoying, of course, and it was particularly annoying, because the winners of the competitions would do so called standard modernist boxes and of course Jorn Utzon was always looking for something more. Apparently in exasperation one day with a matchbox he said, you know, it's easy for them, isn't it?

That's a public building. That's an apartment block. That's a tower. And the thing about that is -

(Laughter) Damn those modernists. You can see his mind working in something like the Opera House because it is many things at once. It's The Scrum of Nun's, amazing turtles, it's the icon for a city, but it's also the platform upon which we carry out the most important rituals we have in our city here. So coming back to children again. We did a project a couple of years ago - I'm looking slightly delirious at the amount of sugar we had to feed that child. And what happened was we did a project in Stockholm for a gallery and your client was a six-year-old, a six-year-old girl. And it was a very serious project about how children think about space and so run very formally as a proper comissioned project. And so we asked this girl what would she like? 'Well, I'd like a pink princess castle.' And we thought, OK, we can deal with that. And so we made a pink princess castle. But that was also a bit disappointing, because already six years old she's trained to this certain paradigm of thinking, but then it got interesting. Then she said, 'Yeah, and I live in this thing

and my sister will be in it as well and a friend will come over and they'll live in it and we'll all be in our own rooms and we'll go down these chutes and slides and meet in the bathroom. (Laughter) Now... That is awesome. The most disturbing thing about that is a six-year-old has such an obsession with plumbing, but the second thing is that of course what it shows

is this amazing spatial mind. And so for us it was the best project we did that year. I mean, imagine - you're in Sydney, talking to your residential clients. More bedrooms, more bathrooms - oh, I'm over-capitalising, can you meet me in Surry Hills on Saturday to choose bathroom tiles? Or you can hang out with kids and design the most amazing buildings.

Option B. So this is actually the city.

This is probably the city that you probably recognise.

It's the one that most of us live in. But it's not that idea of a rationality that we've been trained to try and bring to design. Rather, it's a kind of complex and messy environtment and requires a whole sort of spatial intelligence around it that we've sort of almost forgotten about, So of you actually think about it all the issues we are facing as a society today around our cities are things like transport, density, housing. cost of living, those types of things. Those are all spatial in most of their compoenents. This brings us back to Venice. Because this image was taken by John Gollings was a part of the last Venice Biennale of Architecture, Australian Pavillion 2010. And Peter Werner

gave a talk head of the property council you know, lead agent in the $670 Billion property industry. Gave a talk at the press club a couple of months ago. and he has made perhaps the most obvious point that 80% of us live in cities. 80% of our productivity is from cities but then he went on to say the problem is what good is fair work Australia if you can't even get to work in less than two hours? What good is a new hospital if it is built in the wrong place and then he started making this very interesting discussion because of course he's not an architect. but he said in Canberra they talk about, of course, our properity. We need to advance and increase our properity and the jargon apparantly for that is the three Ps. Population - getting more of us. Productivity doing a lot of stuff And Participation, getting together so we can do all that stuff. And he said all those three Ps are spatial issues. So the lesson there is why is it to make our cities we drill through 230,000 pages of planning law? Why is it that us planners all the people making our cities rely again on all this stuff written down. And often decided in seperate places at separate silos without really any sophisiticated spatial thinking tying it all together? So I guess the thing we want to leave you with today is that we want you to embrace complexity. By doing this it makes you bring design from the hand of the master architect from the grand design - if I could say that - down to something which involves us all. And we also want you to embrace your spatial thinking.

We want you to go out there and try and exercise that muscle in your brain, that intellitgence you have probably forgotten about a little bit. Try and get that going again. And I think I can all meet you down at Callmore Park next Sunday for a bit of spatial callathetics. So that's where we want to sort of leave it, I guess. So, we'll see you in Venice. Every Australian who comes to Venice will get this little red light. And they make these very interesting formations as they walk around the city. And what we're doing in Venice is foregrounding not architects and their buildings but architects who work with other people - with doctors, with enironmental planners with welders, with a range of other people to do things that actually effect the city in very, very large ways beyond the individual building so if you can handle two days with a bunch of sweaty architects, see you soon. Thank you. That was architects Gerard Reinmuth and Anthony Burke at this year's TEDx in Sydney. And you can find a whole range of great talks from that event

on our website. Remember you can follow us also on Facebook and Twitter. That's all for this edition of Big ideas, I'm Waleed Aly See you next time.

Closed Captions by CSI

This Program Is Captioned

Live.

Military defections and Military defections and

cross-border flak as Syria's

rebels get more organised.

We've hit armoured vehicles

the commander says. Gina

Reinhart boosts her stake in

the TEN Network and gives

Fairfax Media an ultimatum. I

don't think any board at any

public company can be pushed

around. Doing the numbers on

the carbon tax. What's going up and

and what's not. In these sorts

of changes you see the best and

worst of people. And - hard

pressed retailers say they've

been shocked by a determined Buy Nothing New Month. It's

simply encouraging people to

also think about a sustainable