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 or accept liability for the accuracy or usefulness of the transcripts.These are copied directly from the broadcaster's website.
Science Show -

View in ParlViewView other Segments

Metal ions in proteins potential cause of disease

Metal ions in proteins potential cause of disease

Copper-containing proteins play important roles in organisms ranging from bacteria and yeast to
plants and animals. The objective is to understand the properties and biological functions of wild
type copper-zinc superoxide dismutases (CuZnSOD) and to understand why mutant human CuZnSOD
proteins cause familial amyotrophic lateral sclerosis.

Transcript

This transcript was typed from a recording of the program. The ABC cannot guarantee its complete
accuracy because of the possibility of mishearing and occasional difficulty in identifying
speakers.

Robyn Williams: Well Stephen Hawking, the Cambridge astrophysicist, is now half robot, half
paralysed, shrunken body as a result of motor neuron disease. No hope of cure in sight...until last
week, from two sources.

[excerpt from AM]

Peter Kay: Good morning, this is AM, I'm Peter Kay. Australian scientists have made an important
discovery about what causes the devastating illness motor neurone disease. They've found it's an
abnormal gene which kills the nerves from the brain to all the muscles in the body. National
medical reporter Sophie Scott is speaking to Professor Garth Nicholson from the Anzac Research
Institute.

Garth Nicholson: Some years ago it was found that a particular protein was present in large amounts
in the spinal cords of people with motor neurone disease and associated dementia, which is often
occurring with motor neurone disease. This protein's called TDP43. So we looked at TDP43 in
families with motor neurone disease because we didn't know whether the protein found in the spinal
cord was trying to help the body recover from the disease or poisoning the body. This was the smoke
that it had something to do with motor neurone disease and we've actually found the fire, because
we have proven in the families that have motor neurone disease with a mutation in this gene, this
causes motor neurone disease.

Sophie Scott: So how important is that, given that we haven't really known what causes this disease
up until now?

Garth Nicholson: Many people have thought for years that motor neurone disease might be some poison
of some sort in the environment. But here we've been able to show that it seems to be a poison in
the body itself. Something's going wrong, so the mechanism actually becomes dangerous and leads to
the death of neurones.

Sophie Scott: What are the implications of that for things like prevention or treatment?

Garth Nicholson: If we're right, this opens up a chance of trying to reduce this protein or get rid
of it or prevent it. And then you'd have either a prevention or a cure for the disease.

Sophie Scott: At the moment what's the situation with either treating patients with motor neurone
or offering them some hope?

Garth Nicholson: Motor neurone disease is probably many diseases and some varieties are extremely
lethal, killing people in three months, and some varieties go for years and years-doesn't really
kill people. So there's a big variation in motor neurone disease and this may reflect all different
varieties of motor neurone disease with different mechanisms. But if there's an underlying
mechanism that's common to all, then you've got a chance of a treatment. And because this protein
is found in all motor neurone disease, this does offer some hope that we might find a general
treatment for all.

Robyn Williams: And Garth Nicholson, of the Anzac Institute at the University of Sydney tells me
that's the clue; many common diseases have lots of different causes. And the challenge for the
scientist is to find a pattern.

Dr Nicholson has been looking at the genetic end. Joan Valentine in Los Angeles has been starting
with the chemicals, the poison itself. And she emphasises how unusual Stephen Hawking has been as a
victim of motor neurone disease.

Joan Valentine: Stephen Hawking is very unusual in living so long with the disease. Most people who
have the disease do not live very long after the diagnosis. And they're very interested in what
science may be doing, even if it's not going to benefit themselves, they're very interested in
hoping that there's going to be a cure some day. It's very affecting, and the graduate student
researchers have been very inspired to work even harder on their research.

Robyn Williams: That's Professor Joan Valentine. She was analysing the chemical structure of
proteins, just for basic interest, to see how the metals were tagged on, when it became known that
there was a link with the disease. Then it all took off.

Joan Valentine: I was an inorganic chemist (still an inorganic chemist) and interested in metal
complexes. Not interested, actually, in anything biological. But then this protein was described in
the literature and it bound both copper and zinc in a very interesting coordination environment.
And so I was curious; I started working on it. And it kept on going from there.

Robyn Williams: An awful lot of people who know science say that if you do excellent basic research
you find that it often-invariably in fact-leads somewhere. Now what happened when you were reading
The New York Times and saw a mention of your protein?

Joan Valentine: I will admit that over the years working on this protein I did get interested in
the biology. But certainly I had no idea that our research would be related to anything of any
health relevance. I was just interested in how the protein worked. But then in March of 1993 my
husband was reading The New York Times and he said to me, 'Hey, Joan, looks like superoxide
dismutase is involved in Lou Gehrig's disease.' I said, 'No, that can't be right.' So he showed it
to me, and so immediately...they said superoxide dismutase...well, it turns out that there's a
manganese superoxide dismutase as well as a copper zinc superoxide dismutase. And I work on the
copper-zinc superoxide dismutase, so that day I went in to work, I spent all day trying to track
down somebody who knew if it was our protein that actually was related to Lou Gehrig's disease, and
this involved calls to Harvard and MIT and finally I tracked down somebody who said yes, it was our
protein. So it was a very exciting day.

Robyn Williams: And what happened next, as a result?

Joan Valentine: Well, it turns out that in our research we'd started out from a very inorganic
chemical point of view and characterised the protein, the copper and the zinc. And as our interest
in this developed, also the tools of molecular biology had developed, enabling one to do protein
redesign. And so we had been taking the protein and redesigning it by mutagenesis to see what
effect that would have on the copper and the zinc. So in my laboratory we had the skills to make
mutant forms of the enzyme. And what The New York Times was saying, and then the article in the
scientific literature that followed it, was that mutations in this enzyme were causing the protein
to be toxic and cause Lou Gehrig's disease. So we knew how to make these mutants and we immediately
started making the disease-causing proteins in order to figure out what effect these mutations
would have.

Robyn Williams: Do it indicate at that stage that linked to a protein obviously there might be one
or more genes and therefore there's a kind of genetic syndrome going on there, hence the mutation
and hence the changed protein?

Joan Valentine: Yes. That's exactly right. It turns out like a lot of other neurodegenerative
diseases, Lou Gehrig's disease, or ALS (amyotrophic lateral sclerosis), has a familial form which
is genetic, it's linked to a gene, and a sporadic form. And back in 1993 nobody had any idea what
caused any ALS except it was known that some of it was genetic. So this was the first breakthrough
in actually finding a cause. And this would be a genetic form of ALS.

Robyn Williams: Any tests done on animals, to see whether your mutant forms caused problems?

Joan Valentine: Yes, actually we were not the ones who developed the transgenic mice, but the
discovery of the mutations in the gene caused the disease enabled other investigators to make
transgenic mice where they took the human gene that was found in the people with the disease and
they put it into mice and showed that it caused the same disease. This then enabled them to have an
animal model in which they could test out hypotheses and has really enormously advanced the field.

Robyn Williams: What is the protein actually doing, because the nervous reaction, obviously, the
breakdown in the nervous system, is something that's quite profound, to do with the conduction of
impulses and so on. So how much is known now about how much that protein is involved with
organically in the body?

Joan Valentine: This is the real irony about this enzyme and this disease, and that is that the
normal function of this enzyme has a very beneficial function, it's an antioxidant enzyme. It
reacts with the superoxide and protects against oxidative damage. So the first thing you'd think of
was, well, mutations must be inactivating it and that's why you get the disease. But it turns out
that's not true. It turns out what's actually happening is the mutations are changing the enzyme
and making it toxic in some way. Based on the transgenic mouse studies it actually looks as though
this disease is a protein aggregation disease, and so what is happening has nothing to do with
superoxide but actually to do with the protein forming toxic aggregates, like Alzheimer's disease
and Parkinson's disease, for example.

Robyn Williams: Does that give you any leads as to treatment? I know you're a chemist but are your
colleagues working on that sort of front?

Joan Valentine: Well, yes. And in fact it's interesting, because many of these neurodegenerative
diseases that are believed to be diseases of protein folding have similarities and people are
hopeful that one will be able to find drugs that can slow down this protein aggregation process.
And it's actually possible that drugs that are effective for some of the diseases may actually be
effective for the others. And there's a lot of attention now to drug screening to try and slow down
this process for all of these diseases.

Robyn Williams: Yes, I remember talking to a professor of chemistry in Cambridge (in England, not
in Harvard) who was looking at this global picture of the proteins, and their shape, of course,
because a whole family are involved in such diseases, suggesting that if you find the key to one,
who knows, it might lead you to many of the others. It's still a long way off though, isn't it?

Joan Valentine: I can imagine that was Chris Dobson, is that right?

Robyn Williams: Yes, it was Chris Dobson. You're absolutely right.

Joan Valentine: He's done some beautiful work, real fundamental work, and I think he's right, this
is a general property of proteins and when it starts breaking, when we really understand how to
control this, it's going to have very widespread implications in disease.

Robyn Williams: Joan Valentine, professor of chemistry at the University of California, Los
Angeles. And Lou Gehrig was the famous baseball player who, like Stephen Hawking, succumbed to
motor neurone disease. But now at Cambridge, Los Angeles and Sydney secrets are being revealed, and
maybe the disease in all its varieties will come under control.

Gut microbes

Gut microbes

The average person has 1.5Kg of gut microbes. They affect biology and health. Abnormalities in gut
microbes are linked to the development of disease. Over the coming years, molecular scientists will
discover more of the secrets of each individual gut bug and the links between them and potential
diseases.

Transcript

This transcript was typed from a recording of the program. The ABC cannot guarantee its complete
accuracy because of the possibility of mishearing and occasional difficulty in identifying
speakers.

Naomi Fowler: The first surprise about gut microbes, or gut bugs, is the sheer amount of them, as
Professor Jeremy Nicholson explains.

Jeremy Nicholson: The average person has about one and a half kilograms of gut microbes, so that
has a big influence on your body, and if you totalled up the number it's something like a hundred
trillion microbes inside you. And you only have about ten trillion cells in the whole of your body,
so they sort of outnumber us ten to one in terms of the number of cells. So they have a big effect
on our biology, and also our health. They're extremely difficult to culture by conventional
techniques, so it's only in the last maybe five to ten years that we've really known about their
diversity and how to study them.

Elaine Holmes: There are a million and one different influences that have been associated with the
presence or activity of certain types of gut microflora. For example, people who are born by
caesarean section have a different microflora from natural births. Professor Elaine Holmes.

Elaine Holmes: There's been proved to be a difference between babies that are breast-fed and babies
that are bottle-fed. Other things that are known to affect them are where you live, the type of
food you eat. Even a change of water can affect the equilibrium of the microflora and it can take a
few days to settle down until it reaches a stable level again. Almost any environmental influence
can change your gut microflora.

Naomi Fowler: Until now most drug discovery has targeted the mechanisms in the human body,
re-engineering human cells and signalling pathways inside the body often after disease has been
detected. But scientists now know abnormalities in gut microbes are linked to the development of
disease. So diagnosis becomes predictive and prevention becomes key.

Jeremy Nicholson: There are a lot of diseases that have occurred that have increased in the last
30, 40, 50 years; diabetes, obesity, some sorts of cancers, even neuropsychiatric disorders. What
we know now is the causes of those are not just in our own genome, although some of our own
genetics is involved, but it's the interaction with the environment. And it turns out that in
almost all of those cases there are unusual interactions with the gut microbes. So with our new
technologies we can start to probe those interactions and thus understand how people with certain
genetic predispositions based on dietary and gut microbial interactions may or may not have greater
probabilities of having particular sorts of diseases. So we can look at it from the point of view
of prospective risk, and we can also look at the prospective targets for intervention, whether
those are drugs or indeed dietary.

Naomi Fowler: Apparently before this study only about five people in the whole world had ever had
their gut microbes analysed, is that right?

Jeremy Nicholson: Comprehensively analysed, that's absolutely true. The reason for that is that the
microbiological techniques and the genetic techniques they use are so complicated and expensive,
that you're very restricted on the number of people that can be studied. At the moment...it's going
to get cheaper in the future, so in the future we'll be able to study a far greater number of
people.

Naomi Fowler: So you actually analyse their poo?

Elaine Holmes: Yes. What we actually do, our part of it here is to take the urine or the faecal
extract and tie it in with the microbial profiling. We collaborate with a group in Shanghai who
have a very good microbiological department, Professor Li Ping Zhoa. They had already set up a
study on a seven-member family. They had samples, then we offered to do the metabolic profiling for
them and to mathematically link the two, which is the expertise that we have within this group.

Naomi Fowler: And can you tell me what is the significance of all this for drug development?

Jeremy Nicholson: Conventional drug companies are running out of targets. They've found all the
easy drugs to get and they're getting worse and worse at discovering new drugs and in the future
they're going to have to move their business model a bit to try and take into account prevention.
So there'll be a convergence over the next five, ten, fifteen years between what we might normally
consider to be a drug company and what you think of maybe as a nutrition or a lifestyle company.
That may be the business future in this area.

Naomi Fowler: So could we see some time in the future perhaps gut microbe analysis clinics where
people would go in to have a personal analysis of the likelihood of potential diseases. I know this
is a long way in the future, but is this the sort of thing that you're..?

Jeremy Nicholson: Well, at the moment that would be extremely expensive. It would cost hundreds of
thousands of dollars per person. But the technology is moving quickly, so that eventually you'll be
able to get chip-based analyses, so it might come down to a few hundred dollars per person, in
which case you'd get a readout of your personalised gut microbes. It's science fiction at the
moment, but science fiction has a habit of becoming science fact quite quickly these days, so maybe
in ten years' time you will be able to personalise your nutrition, for instance, by knowing what
sort of gut microbes you have, and what the best type of diet-and indeed dietary supplements that
you might want to use-are for the particular individual.

Naomi Fowler: So, 21st century medicine will favour prevention over cure, with a more holistic
approach which diagnoses and treats through diet and drugs which target gut microbe abnormalities.
Over the coming years bio-molecular scientists will discover more of the secrets of each individual
gut bug and the links between them and potential diseases. I've been talking with professors Elaine
Holmes and Jeremy Nicholson at Imperial College London for The Science Show.

New approach to building design and transport

New approach to building design and transport

The MIT Design Lab is interested in design problems that have social significance and which don't
fit in the usual boundaries of architecture or urban design or engineering.

Centrally controlled light and air in buildings are inherently wasteful and don't cater for the
needs of the individual. New technology allowing individual control of light and air in office
buildings will increase efficiency and provide people with the environment that suits their needs.

Transport is also in Bill Mitchell's sights. The idea is to develop a lightweight vehicle that is
energy efficient, and works on the principle of one-way rental shared use.

Transcript

This transcript was typed from a recording of the program. The ABC cannot guarantee its complete
accuracy because of the possibility of mishearing and occasional difficulty in identifying
speakers.

Robyn Williams: MIT in Cambridge, Massachusetts, is a great centre for innovation. We have two
ex-pats living there at the moment, leading in their fields. First, from Melbourne, Bill Mitchell.
He isn't just an architect, not just a city planner, he's all of those.

Bill Mitchell: Well we're interested in design problems that have some level of social
significance, that really don't fit within the traditional disciplinary boundaries, they don't fit
within architecture, urban design or mechanical engineering, but spill in some messy way across the
boundaries.

Robyn Williams: So you're looking at the new technologies, the new ways of putting things together,
to see whether you can solve problems that otherwise have been intractable.

Bill Mitchell: That's right. Exactly. That's our big interest, yes.

Robyn Williams: So if you look at places like some parts of Australia-New South Wales comes to
mind, oddly enough, where the infrastructure is shaky and you've got some buildings where you could
spend a lot of money getting them up to date. What you could do is bring them well into the 21st
century, starting with buildings; what sort of things would you look for if you're examining, say,
lighting in a new building?

Bill Mitchell: Well, lighting is a huge opportunity at this point because lighting is coming into
the digital era, gradually. So now we have solid-state lights, for example, and LED lights,
solid-state devices, not a hot wire glowing there. So inherently they're pretty efficient. So if
you combine that with digital technology for individually addressing light sources, the difference
between a light and a pixel really disappears. They become precisely controllable points of light
that you can distribute around in a space. And this is a way of fundamentally rethinking lighting.
Instead of having a few large light sources with very crude controls, maybe one switch that
controls ten lights or something, you have a very large number of small light sources that are very
precisely controllable. So inherently this is much more energy efficient. And it's much more
sensitive to human needs. You can put a lot of sensing in there, you can really tune the lighting
to exactly what it needs to be doing at any particular moment.

Robyn Williams: In other words, you can save energy as well.

Bill Mitchell: Absolutely you can save energy.

Robyn Williams: And what about windows? Now that's something that really irks many of us who go in
to hotels, or even office windows where you can't open them. Why hasn't it been possible for such a
long time?

Bill Mitchell: This is a fundamental point about 20th century architecture, that really the
technology that supported 20th century buildings wasn't all that good. So air-conditioning systems
of buildings for example had relatively crude control systems, so being able to open a window
created all kinds of complicated problems. So the standard architectural response, and the standard
response of mechanical engineers was just to seal the building up. And this is not an
energy-efficient thing to do. It's not a pleasant thing from a human point of view. So with more
sophisticated control, you can begin to get a building that's more humane and responsive. And to
put it in very simple terms; yes, you can open the windows.

There's a very interesting point about this. People think high-tech architecture of the 21st
century is going to be some kind of Buck Rogers thing with all kinds of flashing lights and shiny
tubes and, you know, technology dominant. But really good 21st century technology disappears into
your pocket and disappears into the woodwork. Your phone just disappears. Computers used to require
fancy special environments. Now you can use your laptop anywhere. So the big paradox is really
high-tech architecture of the 21st century is not going to look high-tech at all, it's going to be
built around fundamental human requirements like light and air and view and sociability. And the
level of technological service will be very high but mostly you won't even see it.

Robyn Williams: When is this sort of building going to go up, when are we going to see it?

Bill Mitchell: Well, we're starting to see it now. These things happen gradually. And of course you
don't see a big revolution in architecture all at once because we have a tremendous inheritance of
old buildings that are going to be there for a very long time, so architecture changes slowly at
the kind of growing edge, where you're building new buildings and retrofitting old buildings. Like
for example all of the new buildings we've done at MIT over the last five years or so have operable
windows. Very simple thing but this is a fundamental change.

Robyn Williams: So today when we're speaking, okay, there's snow on the ground, freezing rain is
coming, and you want to keep the place warm and you've got two or three people who want to open a
window for some reason; what controls the temperatures in that part where they've decided to go
their own way?

Bill Mitchell: Well the traditional way to approach this is you have a few thermostats on the walls
and you have a central air-conditioning plant and centrally controlled lighting and that kind of
thing, so there's very limited capability to adapt to what people are actually doing and what they
need.

For example, if you're sitting down and reading, the requirements are different from if you're
exercising, for example, or having a meeting. So what you really want to do in the end is sense and
recognise what people are doing and then adapt the response of the building to what people are
doing at a particular moment, and the current climatic conditions. And that's actually a very
complex problem. Buildings seem simple because they're around us all the time, but behaviour is
extremely complex. So as we get more computational capacity, as we get more networking, as we get
ubiquitous sensing, we're going to be able to control these things in much more sensitive ways.

Robyn Williams: Okay, well there's a problem, because my people who want to open the window, even
when it's quite cold, are told by the person who's running the building that the machines, many of
them (they're not that old-fashioned) need to have a constant temperature otherwise they fail. What
do you do about that?

Bill Mitchell: You build better machines. And you go in the direction...this is a big
generalisation, I need to be careful about being cavalier about some very complex issues, but if
you think of the analogy of the internet, which is a very large number of intelligent network
devices that have a lot of redundancy and a lot of local capability to react and respond to what
people need, compare that to an old-fashioned mainframe computer system which had one big central
computer and some terminals or something. We're moving, I think, in the direction of that in the
control of buildings. So think of a very large number of decentralised actuators which control what
goes on in a building, very dense sensings scattered around, and a lot of networking and processing
capability to pull all this together, so then you can get localised response rather than big global
response.

Robyn Williams: If the rooms know that there's no one in them, they can then switch things off.
That's already happening, isn't it?

Bill Mitchell: You've got to be careful though. So this already happens, but one of the most
annoying things in a building I sometimes inhabit is you sit there thinking, and the building
thinks you're not there and switches the lights off. And this is really annoying.

Robyn Williams: Some tweaking to do. Now let's go out of the building and look at the transport,
because in your department here at MIT you're concerned of course with the whole picture. How is
transport likely to become revolutionised?

Bill Mitchell: Well a traditional way of thinking about urban mobility is that you have public
transit systems and then you have the private automobile to provide personal mobility. We've tried
to break away from that way of thinking and to really fundamentally rethink the issue of personal
mobility and try to develop systems that are even more attractive and functional than the private
automobile but much more energy efficient and better from the perspective of global warming, urban
congestion and so on.

A major project we've been doing with General Motors, for example, is what we call the City Car
Project, where we've developed a little, light-weight electric car that folds and stacks like
shopping carts at the supermarket or luggage carts at the airport. It recharges when it's in its
parking space just like your electric toothbrush, so you never have to think of plugging it in or
taking it to the gas station or any of these kinds of things. It's always full of energy. It
doesn't have to carry around a lot of batteries. And it works on the principle of one-way rental
shared use, so there are stacks of these cars around the city at closely spaced intervals. When you
want one you swipe a card, drive it away, go to wherever you want to go and then drop it off at
another stack.

Robyn Williams: Well, this is staggering, because I wrote a book (just to plug myself) called
Future Perfect, and the chapter on transport has a kind of dream that is almost exactly what you've
just said.

Bill Mitchell: Indeed, it's a very rational idea and it occurred to you, it's occurred to us, it
occurred to a number of people in Europe. Paris, for example, has a one-way rental bicycle system
now which is extremely successful. The thing that made this feasible in the end was of course
ubiquitous networking and embedded intelligence that enables you to really do the management and
control of a complex system like this in an efficient kind of way. That's really important. And
then the miniaturisation of technology; we were able to make our little electric cars with in-wheel
motors, for example, so very high-torque electric motor right in the wheel. So we get rid of
traditional engine, drive train, all of these kinds of things and make something that's actually a
very, very simple automobile and just about everything is in the wheels.

Robyn Williams: Now you said the battery's not a problem. Why not?

Bill Mitchell: Battery technology is still highly imperfect. There are advances; in our project
we're using some of the latest and most sophisticated batteries. But the fundamental problem with
batteries is that they have relatively low energy densities, you have to carry around a lot of
batteries in order to carry around a lot of energy, and their performance degrades over time as you
charge them and discharge them. And they create big recycling problems, all of these sorts of
things. So our approach to this has been not to look for some sort of magic in battery technology
but to minimise the use of batteries.

So with our strategy of recharging whenever you come to a parking space, or even perhaps recharging
on the road through induction loops in the road, it means you don't have to carry around a lot of
batteries and you don't have limited range, because you're always picking up energy. And this is
technologically feasible now. There are some technical tricks to doing the charging, to get enough
transfer rate and that kind of thing. But it's all do-able

Robyn Williams: What does the car look like? One of those sawn-off two-seater jobs?

Bill Mitchell: The ones we've developed so far are two-seaters. What we've tried to do is not to
make it look like a sawn-off traditional automobile because I think if you make it look like a
sawn-off traditional automobile, no matter how safe and capable it is, people are going to perceive
it as an inferior, unsafe small car. So we've really followed the kind of iPod strategy. You know,
an iPod is just an mp3 player, but Apple very cleverly, through careful design and very careful
marketing got people to perceive this as really a different product category, a new kind of thing.
So we've developed this and designed it as a new kind of personal transportation device and not as
a sawn-off automobile. I don't think it looks like a sawn-off automobile.

We've also done a motor scooter recently for Ssangyong Corporation, SSYC, in Taiwan that's about to
go into production. And this is very similar. It folds up too, it's like a transformer, it folds up
into a little package and you can drag it along like wheelie luggage. And again, there are very
practical reasons for doing this, but also we wanted to make a product that does not look like an
inferior version of the traditional product.

Robyn Williams: Because General Motors have had some financial problems announced in the last
couple of weeks. Presumably your projects will be immune to that, will they?

Bill Mitchell: Well, with all of these projects we're working with the research arm of General
Motors and with some of their concept car people and so on...the product side of GM is a whole
different complex thing. Our project, if you really do it on a large scale, transforms the
automobile industry. So instead of producing commodity products, which automobiles essentially are
now, under very intense pressure, it's a lousy business to be in, we shift it to being an
innovative mobility service business. This is the model of Google, which has been a very successful
model, in which you provide a base service (in this case urban mobility) at very low cost and very
efficiently, and then I think you can make a terrific business out of sophisticated profitable
add-on services on top of that. So this is our model. It's to really shift the fundamental basis of
the urban mobility business from commodity product to innovative service.

Robyn Williams: So on the basics of the car, how far does it go and can you take it out of town?

Bill Mitchell: Well this is designed as a city car. It's designed as an in-town urban car. You can
go as far as you want within the domain of the car, because whenever you stop, as I say, it picks
up energy. So the traditional way of thinking about this, that the vehicle has a sort of range
depending on the amount of energy it carries around with it, is not quite the right way to think
about this one.

Robyn Williams: Now, you come from Melbourne. You go back to Australia on a regular basis, you've
seen Sydney as well...if you imagine what Australia might benefit from in terms of these big ideas,
what would you like to see when you go back?

Bill Mitchell: This is not news, of course. Australia faces massive urban sustainability problems,
partly because of climate, partly because of dependence on energy sources that are not the right
energy sources for the long term. So I think the really big and exciting things to do are to really
fundamentally rethink our mobility systems, energy systems, the kinds of nervous systems and
control systems that manage a city, and to really think about how do you make cities like Melbourne
and Sydney particularly into real 21st century cities.

Robyn Williams: Yes, well despite the gridlock in Sydney, how do you turf the Sydneysider out of
his car, or her car?

Bill Mitchell: I don't think you can do it by force. And I don't think that's the right way to
think about it. I think you have to create something that's genuinely a more attractive alternative
at a kind of behavioural level and at the level of desire of what people want to do. And at the
same time it's more energy efficient, it causes less congestion, it creates less global warming and
so on.

So the idea of our little automobiles that we've been working on is not that this is some kind of
moralistic thing, you know, you've got to use these things because it's good for the planet and
we're all going to perish if you don't do it. We may all perish if you don't do it, but I don't
think you can affect behaviour by that kind of moralising in any significant way. So we've tried to
make these things so they have the right technical properties but they're highly desirable, they're
cool, they're fun, they attend to issues like...you know, one of the things with the urban vehicle
is it puts you on display and you want to look good. This is one of the problems with the Segue, by
the way. Think of the Segue, which was a very clever vehicle, developed a few years ago, projected
as a great potential revolution in the city. There are two problems with the Segue. One is there's
no real infrastructure for Segues. But the second thing is, on a Segue you just look like a dork.
There's no way out of it, there's absolutely no way out of it. It's not a cool vehicle.

Robyn Williams: Just remind us what it's like.

Bill Mitchell: Well it's a kind of little platform with a stick in the middle and you stand on it
and you just look very foolish. Unlike, for example, a surfboard or a skateboard where if you can
handle these things you almost can't help looking good. So this is a crucial component. Imagine The
Wild One, imagine, old Marlon Brando movie on Segues, it's not plausible. This is a point that's
often missed in discussions of urban sustainability and transit and all of these kinds of things,
that you're not going to succeed just because you have an energy-efficient transportation
technology. It has to be desirable. It has to work socially. It has to be something that people
think is fun and they want, they feel good using it and so on. This is a tremendous design
challenge but it's just as important as the technological challenge.

Robyn Williams: And the premier of New South Wales and the premiers of other states could order it
now and when would it be visible on the roads, do you think?

Bill Mitchell: We've built a half-scale prototype; we're building a full-scale prototype now. It
would take a couple of years and significant investment to get this into production. Our little
scooter, which operates in the same kind of way, recharges in the bike racks, Ssangyong is
currently intending to have that available in February of next year.

Robyn Williams: Mr Iemma, Mr Brumby and all, welcome to the future. Professor Bill Mitchell is
director of the MIT Design Lab. And while you're up, the ABC itself is trying to join this greener
future in a very public way. In a few weeks we're launching Green at Work, a website where you can
see many of the ideas such as those Bill Mitchell talked about just now, plus those Bernie Hobbs
and others at the ABC will be broadcasting. Green at Work coming up on ABC Radio National and
throughout the ABC outlets.

Low-cost robots for the production line and the home

Low-cost robots for the production line and the home

Rodney Brooks is looking ahead to the days of labour shortages in today's developing countries and
is creating robots for the ordinary person. Rodney Brooks points out that the PC empowered office
workers. Productivity increased. His idea is to empower people with robots. Manufacturing may be
brought back to western countries. Robots can also help around the home. The key is cheap
computation.

Transcript

This transcript was typed from a recording of the program. The ABC cannot guarantee its complete
accuracy because of the possibility of mishearing and occasional difficulty in identifying
speakers.

Robyn Williams: So now let's cross the MIT campus to meet a legendary Aussie. The robot man behind
iRobot household slaves and others destined for a factory near you.

You're listening to ABC Radio National, where we've come to the Massachusetts Institute of
Technology to catch up with an old friend from Australia, Rodney Brooks, who comes back on many an
occasion, but when we've talked about robotics in the past it's always been in abstract. I'm
actually now in his laboratory. It's a wonderful space-age sort of module, I suppose you'd call it;
huge, very high ceilings, lots of benches, coloured robots everywhere of different sorts. One of
them has got eyes and is holding coloured balls, and there's wiring and...Rodney you're no longer
running this giant lab at the moment, what's your actual function here?

Rodney Brooks: Right now I'm on sabbatical, and I'm trying to figure out how to do research again,
and the most surprising thing to me is that after ten years of being a full time bureaucrat I'm
back at programming computers and doing computer vision. Computers have got a lot faster in the
last ten years.

Robyn Williams: You can still find your way around?

Rodney Brooks: Yes, but I press the button and I've got the answer, right there...it's fantastic.

Robyn Williams: It is. So you've really got to be on your metal, literally. Most of these are made
out of all sorts of different materials. In front of us, straight away, there is an orange one that
looks like one of those devices that you see in places where robots build motor cars. Is it close?

Rodney Brooks: That's exactly what it is. That's a commercial one we've got there, just as a
place-holder. Can we walk over there and I'll show you what's going on here?

Robyn Williams: Yes, let's go over there.

Rodney Brooks: This corner of the lab here is my little Chinese factory. With one of the other
things I do in life I founded a company called iRobot and we build vacuum cleaners, robot vacuum
cleaners. We've sold well over three million of them world-wide. And so I started spending a lot of
time in factories in China as we were building these devices, because it's just not cost-effective
to build consumer electronics products in Europe or North America or Japan anymore, or Australia,
that sort of thing doesn't happen. And I started thinking, well, we're going to run out of Chinese.
Labour costs are going up there. As the standard of living goes up we're seeing people who used to
work on these production lines saying, well I want to go to college and get the degree. And after
you move through Vietnam and then you run into Burma, and on the other side you've got Thailand
which has come up in its economy, and we're just going to run out of places to exploit. That's what
western democracies have been doing for the last few...

Robyn Williams: So they can get their own labour shortage...

Rodney Brooks: They're going to get the labour shortage, and if we need to figure out how to change
the structure of labour...you know, in the same way that the industrial revolution changed the
structure of labour and then electric motors really changed the structure of labour because you
didn't have to distribute all the mechanical turnings through belts and stuff and the mills
changed. I think it's now time to think about robots for the ordinary person.

Can I explain a little analogy; many of the listeners wouldn't remember this but you and I remember
that back in the '70s computers were not things that ordinary people touched. They were often in
the back room, and an office worker might get a fan-fold printout of various numbers that they'd
look through. But if they wanted a different set of numbers it had to go through a whole
engineering staff; a systems analyst, a programmer, a punch-card operator, a computer operator,
etcetera. It took weeks until you got the new sorts of reports, if you were lucky.

When the personal computer came along, around 1980, ordinary office workers became empowered to
have direct access to computation. And through spreadsheets they started becoming their own
programmers, their own automation engineers. They didn't say, oh, now I'm a programmer. They
instead said, oh, I add up this column of numbers and then I take 5% of the total...and they did
that on the spreadsheet, and so they automated part of their job and increased their own
productivity. And over the last 25 or 30 years, computers have got more and more capable, we've
connected to information sources...it hasn't replaced the office worker, the office worker still
sits there and orchestrates the information flow, but they do tremendously more, so they increased
their productivity.

Right now, robots, where they exist in industry are backroom engineering operations. Ordinary
factory workers never get near them because they're too damned dangerous. And they never get to
control them. They're a separate thing happening. So we want to build robots that an ordinary
little enterprise, three or four people, could order a robot and it would arrive the next day and
they'd take it out of the box...they wouldn't look at a manual, they'd set it up and they'd show it
what they wanted it to do and two hours later it'd be doing simple cases of simple tasks. The
people who programmed it would be also working on those same tasks. They would be right there to
correct it when it went wrong, and they'd increase their own productivity. And I think that can
trickle through the whole of manufacturing and even a lot of service work. So it's increasing
productivity for manual workers, is what we're trying to do.

So what I've done here is set up a little Chinese factory with a conveyor belt which normally would
have a lot of Chinese workers sitting on each side and as a part came along on the conveyor belt
they'd visually inspect it and see if their task had been done or not. If their task hadn't been
done they'd pick it up, insert the screws or whatever they had to do, put it back. And the way the
line is balanced is if it's a short task there might only be two people doing that one. If it's a
task that takes a long time there might be eight people doing the same task, so you sort of balance
this line on the conveyor belt and all the workers just sit there and do their tasks, and in a good
factory they're allowed to go off for a bathroom break. But what I'm trying to do now is; can we
build a robot that we could just roll in and do some of those tasks, and that the people program?

Robyn Williams: And how have the Chinese responded to this wild idea?

Rodney Brooks: I haven't told them about it.

Robyn Williams: Oh, I see, right, okay. You're being pre-emptive.

Rodney Brooks: No, this is actually, I think, for bringing manufacturing back to western
democracies if you like, increasing the productivity of the person working for the supplier for a
car company or...I don't think we're going to get straight into car companies, but suppliers, and
an increase in the productivity so that there's less outsourcing to China and other places, and
industry can stay in the countries more and then, because it's local, it gets to be much more
localised in what it's building. It doesn't have to be mass, uniform for the whole world. Local
communities, I hope, can still get the productivity levels you need for cheap goods but have it
unique for the location. One of the things I hate is if people talk about going on a trip to go
shopping. Every airport in the world has exactly the same stuff in exactly the same stores.

Robyn Williams: So if only you'd been on-line in advance, maybe Adelaide could have kept its motor
industry. Anyway, we're looking to the future.

Rodney Brooks: One of the things, to make these work, these robots have to be very low-cost. If
they're expensive they won't work. So we have been playing around with building prototype robots,
and you see the robots with human form here and they're very expensive and they're mechanically
precise, and then we've taken cheap pieces of material and trying to get similar sorts of results
out of them. The key thing is that computation continues to get cheaper, and cheaper and cheaper.
So what we're trying to do is replace mechanical precision and precise components and expensive
components with really rather poor components.

So here's an example: we've got a shaft that's being driven by a motor and we want to have a spring
in series with the output so that we can have an arm that's safe for a human to interact with. In
our humanoid robots we have thousands of dollars of bearings and springs and precise stuff. And
here in my hand I've got something that probably cost about ten cents. There's a couple of pieces
of brass with some strips of rubber in there. It's elastic but it turns out to be non-linear in a
horrible way. It has what's called hysteresis, because rubber has memory. So trying to control this
with a conventional straightforward controller and knowing what state it's in, it's a mess, but
computation continues to get very, very cheap. So here in my other hand this is actually a
two-year-old little tiny circuit board. It's got a 400 megahertz personal computer on there with a
web server and running Linux and everything else...

Robyn Williams: Size of a wafer.

Rodney Brooks: It's called a gum stick, actually, because it's the size of a big stick of gum.

Robyn Williams: Oh, that's right, chewing gum.

Rodney Brooks: And this was $100 two years ago, so you can imagine it's much cheaper now. This is a
400 megahertz PC. Not so long ago that was a top-of-the-line thing. Well, hey, you shove this next
to this few cents' worth of actuator and you do computation to compensate for the non-linealities,
compensate for the hysteresis, and suddenly you get a nice, linear system, beautifully
controllable, because you can use cheap, cheap, cheap computation.

Then you get motors which are really cheap. This is the motor out of an electric window winder in a
car, so it's really cheap and if you play with the gearbox you can hear the backlash in those
gears. It's a terrible gearbox. Cheap, cheap, cheap again - use computation to model what's going
on, compensate for the mechanical imprecision, and then we built this robot arm for just about $100
in components, which has the same effective characteristics as an arm that we used to build for
$20,000 or $30,000.

So we're trying to get the cost way down, get the useability way up, again using computation, so we
have cameras lying around here and various systems. And we used to use expensive cameras, now we
use $30 web cams and compensate in the software for the terrible lenses and things. And so we're
trying to build really low-cost robots that are safe for people to interact with by having these
series elastic elements so it won't...that orange one behind you, if it was running it could knock
your head off, I'm afraid.

Robyn Williams: Oh, I see, it looks pretty hefty.

Rodney Brooks: We're just using that as a stand-in while we're developing the other arm, and we've
just got the first finger on the end of it here. This is a printed finger, we printed this in a 3D
printing machine and this is going to be covered in touch sensors and will be sensitive enough to
reach into a bucket of screws and pull out a screw to drop it into a product and then screw it in.

Robyn Williams: It looks like a sort of Lego finger.

Rodney Brooks: Yes, the colours of these 3D printers are fabulous.

Robyn Williams: May I just wander over here to just briefly say hello to this nice robot with a
chest and arms, it's the one that's holding the coloured balls.

Rodney Brooks: Yes, this robot is called Domo. It's a terrible name. It's from the old song Domo
Arigato, Mr Roboto, but it means thank you in Japanese. The name of the robot is Thank You.

Robyn Williams: Blue eyes. Doesn't look like Japanese eyes.

Rodney Brooks: This robot is completely safe to interact with. It has two arms, it's like a human
in form. It's got two cameras up above which give it stereo vision. And around its cameras we have
things that look like eyeballs. Why would we possibly have that? It serves no function for the
robot. But what it does, is it lets a person know where the robot is looking. Because we're really
good at estimating where someone's eyeballs are pointing, so we make it look like eyeballs.

Robyn Williams: That's what the white stuff is for, to enable you to see the pointing...yes.

Rodney Brooks: So if I'm trying to show you some task I'll hold something out in front of you and
I'll keep looking at your eyes to see if you're looking at the thing I'm trying to demonstrate to
you. And so we're trying to look at how a natural interaction where you need no training, you just
keep glancing at the eyes, you know what the robot's looking at, know what it's paying attention
to, and you can know that the interaction is hopefully going okay.

Robyn Williams: And is this something in the past. Is it Jurassic era robotics, or is it still
something being worked on?

Rodney Brooks: This is a robot that we started about four years ago and it's getting to the end of
its life. We're on the final project with it right now.

Robyn Williams: So the whole field is moving pretty damn fast. You've come back in the middle of a
computer revolution and also a mechanical revolution and a cost revolution as well. We always ask
about the home, of course, and that's the difficult thing because the home is unpredictable. If you
wanted a robot to do the Hoovering it would have to find its way around all sorts of things.

Rodney Brooks: Well here's a robot to do the Hoovering right here. This is actually one of my
commercial products. They're on a schedule, they come out every night and they clean this place. So
let's listen... [switches on, robot starts up]

Robyn Williams: It's moving out, and it's turning round. It's like this great big disk, it's going
in one line and then turning a corner, and it's got brushes and it stops, and it goes round your
foot. It bumps into something and manages to avoid the obstacle and not knock it over.

Rodney Brooks: I'm just going to send it home...

Robyn Williams: So it can find its own path back?

Rodney Brooks: Now it's heading back towards the dock. So we have these programmed and they come
out every night and they clean the floor and then they go home and dock themselves and recharge
themselves. And sometimes we forget to clean out their bins and they get full and get stuck
somewhere.

Robyn Williams: How long have you had that?

Rodney Brooks: Well we started selling these about five years ago. This is the third generation,
the 500 series, and you can buy these in Australia.

Robyn Williams: So apart from vacuum cleaning, what's on the horizon?

Rodney Brooks: Well, I think that's an interesting thing. You know, you look at you and me with our
grey hair and we're starting to lose our hair. And you look at the demographics of the world coming
up, and there's going to be a lot more older people and a lot less younger people. So in the
western countries people are worried about how we're going to fund the pensions, or we call it
social security in the US, and it's been a big political issue. But more than that, who's going to
provide the services for all us old baby boomers? And so the competition for services is going to
go up because there'll be relatively less people at working age relative to the number of older
people, and that's going to cause some real disruptions. You see that a little bit in the US, a lot
of people moving to Mexico for their retirement because their pension cheque can buy more there. In
Japan people are moving to the Philippines and Thailand because they can afford full-time nurses,
et cetera. I think we're going to see a lot of pressure on where that service comes from, and
that's going pull on any sort of increase in productivity that robots might be able to provide.

Robyn Williams: So when I get really old and grey, and let's say for a spell, anyway, I'm living
alone, could I have a robotic nurse, perhaps, doing at least some basic chores?

Rodney Brooks: Well I think a lot of Japanese companies are really working on that. Exactly whether
it's going to be in time for you and me I'm not so sure. If we look at 50 years, I'm confident that
it's going to be there. But there's going to be pressure for that, and I think the baby boomers are
going to want to stay in their homes a lot longer than other people have and they're going to be
slightly healthier, so they will stay longer. So there's going to be a lot of different pressures
coming together.

Robyn Williams: Well you've given me a range of the sorts of things you're working on now that
you're back in the lab. Is there anything you've left out?

Rodney Brooks: I'd still like to discover the secret of life. What it is that makes systems living
versus non-living, but I must say that in my older age I'm feeling more productive on practical
applications than deep philosophical questions.

Robyn Williams: Now a final question about being here in MIT where you've been for so long and
where the provision is so spectacular...just walking around the mighty campus you can see a range
of things. I've just been talking to an architect actually from Australia looking at completely
different forms of transport and building design and materials. How would you compare Australia and
America as a place to work on robotics? What do we need to get going back in Oz?

Rodney Brooks: Well actually you know Australia has some fantastic work going on in robotics for
mining and field robotics, so there is some great work at Sydney University and CSIRO, and other
places. I think the secret to those places is they're really tapping into industrial funding and
know how to work with industry so that the applications come out of their rather pure work that
they do in the university. That's something that MIT's been doing for over 100 years. So we work
with companies all the time, and that's where the funding comes from. We don't get an allocation
from the government, we're out there competing every day to get someone to pay us to have the fun
that we have.

Robyn Williams: Could you be tempted back?

Rodney Brooks: I come back to Australia often, and yes, we'll see.