Don Norman Presents at Tecnico Lisboa: People-Centered Design: Why it Matters?

you and professor tom rodman is a well-known professor in this area is currently University in San Diego in between jobs was at Apple and I don't think we need to explain who we use by giving the floor thank you thank you so I want you to imagine that you are on a crowded Street in Lisbon and you want to cross the street and here comes some cars with no drivers how do you know that you can cross how do you know that they see you what is your crossing in the middle of the street sing so you want to say go go well how do you do that or what if they want they slow down and they tell you to go what how would they do that now I'll warn you it's very easy to think of solutions if you think of one car and one person but suppose that you are near technical some place where there are lots of students so there are many cars and bicycles and skateboards and motorcycles and scooters and many pedestrians and maybe many cars that have drivers the solution has to work in that case not just when there's one car and one person so I'm here to tell you about what we do in San Diego we started a group which we call the design lab and we decided that we did not need to do a design program a traditional design program because for many of you and certainly most of the world design is something that makes things pretty that makes them attractive objects and those are very nice and I like them and I buy them when I used but we're not very good at that in San Diego and the world doesn't need another school teaching that there are many good schools that already teach that so what can we do that's different we decide that we were going to work in a difficult problems we call them complex socio technical problems and in a few minutes I'll tell you about design X which is our way of describing it and so what we are doing is we're going to work on complex problems like the ones of driverless cars which is posing many many difficulties it's their legal difficulties political difficulties economic difficulties social difficulties and as you saw behavioral problems just knowing how do you communicate with a car that has no driver those are complex socio technical problems in their systems so we are doing what we once called user centered system design which of course stands for user centered system design UCSD which is my University so what we do is we work on these problems but we have divided ourselves into six different areas that we call centers of excellence and we work on different kinds of problems so for example we're working on activity centered visualizations so when we have big data and we now want to present it to you well many of the computer scientists computer graphics people some of you may be in the audience make very beautiful diagrams and pictures and illustrations and clever ways of showing all the data but without any thought about how it's going to be used and most of the time these are wonderful beautiful to look at but when you have a problem solved they're not the right presentation so trying to understand how to present something is difficult in the business world well and for that matter in the university my Dean tells me this she says that they came an answer what she needed to do her job and she explained all the different numbers and different statistics she needs and they said okay and then once a month somebody comes and gives her used to be a big printed book that Vic and nowadays they just give her a URL and so she said everything I ever need is there but you can't find anything so it's useless and every month they give her a new book or a new URL to update the statistics that she'd never uses and what you need to do is actually sit and watch her understand how she does her job and what kind of question she needs to have answered because she would like to have the answer in 30 seconds and as possible if you knew what questions she was interested in and that requires us to study the person and the activity and generate the diagrams appropriate and of course she does many different activities so she needs different kinds of data at different times that's one of our groups another one is people centered automation which is looking not just at automobiles which is automatic but also there's a real fear that automation is going to take away jobs and your jobs not the jobs of people doing mechanical things but your jobs thinking jobs programming accounting lawyers designers a new way of designing Autodesk is doing this the United States it used to be if I wanted to design Oh a partition in an airplane that separates say this class of passengers from that class of passenger would give it to a designer who would draw a picture sketch it using tools that would have to be way a certain amount and it would have to be strong enough because people bump against it and put their feet against it and so on so how do you make it attractive and it works so the computer program would figure out the forces but the designer would design it that's not how you do it today well it has you you will start doing it in the future what you will do is you say to the program well I need something about this size and it has to be using light as possible and here are the forces go and the program will generate a suggestion and another one and another one thousands of them and what you do instead is you say oh that's good that's although this is horrible don't do any more of those people are these and you keep sorting and it's learning by what you say is good and what you say is bad and changing what it's doing and then either gives you something that you say yeah that's neat or you don't ever find anything in which case you back off and you say hmm maybe I have to change the constraint or the goal or some of the specification but you don't actually ever draw anything in some sense you don't design but you are designing that you're giving the high level instructions in some of the new Airbus airplanes the dividing barrier was done with this program and it's 3d printed and the advantage of 3d printing is you can make parts that you can't make in any other way you can make parts that are hollow inside and so what you can do is make it very light because when you do the analysis of the of the part it turns out a lot of the stuff inside there's not add to the strengths that just adds to the weight so you leave it out well that's a new kind of design how would we do that in a way that people could interact with the technology how can a physician use a computer system to make them a better position not to replace the position but to make them better just like most of our tools a calculator does not replace me but it makes me better because what I do arithmetic or algebra or calculus I make mistakes so with a calculator I still have to formulate the problem and I have to interpret whether the answer is right but it can do all the drudgery you may not realize this but solving mathematical equations is not what mathematicians care about what you care about is the equation itself and maybe a demonstration of its correctness but solving it that's mechanical and takes a lot of time and people make a lot of errors so that machines do that how can we create machines that substitute for us where we're bad and help us where we're good we'll come back to that we have a center for health we're working a lot in health care trying to change health care and thinking of it as a system not as well known as a device here or a device there but a complete system we have a Center on large-scale learning how can I teach a hundred thousand people in a class and yet give you individual attention how do I teach a class with a computer where it isn't a class in mathematics or programming but is maybe in art history or philosophy so where there are no correct answers and the score you're we have to determine whether you are giving a sensible coherent argument we also are doing work a program on what we call work technology and society which is basically all of the above how all this comes together and one of them is we call it what we call it social computing where actually people are finding their own solutions and we help them so I had breakfast this morning with people at developing lis it's called patient-centered innovation innovation patience innovation where patients in the medical system solve their own problems and I was told the story of someone who was told he had a bad artery and he would probably die within 12 to 24 months and there was nothing that could be done about it well he was an engineer of the patient was and said well that's stupid if I was told that this we have a pipe and the pipe is going to break in the next 12 to 24 months I'd figure out the little patch I could put over it so let's figure out what I can do it to myself and he spent a lot of time learning about blood vessels and the issues and he invented of a device that he could be inserted over his aorta and it would be vented from bursting and he was told he had 12 or 24 months to live and that was ten or twelve years ago but he did it by himself so how could we do more of that how can we get citizens to solve their own problems people with diabetes have to measure their blood sugar level and then determine whether they should get a shot of insulin or eat some candy or something sweet get more sugar well that's awkward for those of you who have diabetes in the audience and I suspect there must be some given the percentages it means you're either taking a blood sample periodically throughout the day four or five times or if you have a continuous glucose monitor you have to pull out the device and look at it periodically and that decide what to do so a bunch of patients decided that they didn't want to wait for somebody to invent an artificial pancreas so they built their own they found a continuous glucose monitor that you could wear would always show the results but only on a special leader so they reverse engineer the computer code or made it available so you can see it on your cell phone or anybody could look at it with protection and then someone found the company that made an insulin pump and figured out how to make it work from the internet and then it's a group of them all got together and wrote the code to connect those two so it automatically inserts insulin when it's needed and not when it's not needed the sugar is your you're worn to eat a piece of candy or take a sugar sip because if we don't have a sugar pump yet but they did this all by themselves and in the United States of a company wanting to do that they would need to get permission from the federal drug drug authority and do clinical trials which would cost it could cost as much as a billion dollars and it can take four or five years it's very difficult to get permission but by having the patience do it they did it immediately and one of the companies will work that we're working with Dexcom made the continuous glucose monitor that this person took apart and demonstrated to the world how it works and a set of complaining what they did is they hired the person who said keep doing that because we got so many more advances by having the people experiment on themselves that's a new way of doing things new methods of Education new methods of automation new methods of innovation so vision things were working on at UC San Diego and we call these problems we call them design X because the number of years ago a number of us were on the Advisory Board of a new school it's it's the College of Design of innovation that told you University in Shanghai and so we were sitting around making our teas and decided that we all wanted to work on these complex problems and that isn't what designers are taught to do but to us said this is for the new generation of design design is not making pretty things design is a way of thinking of solving complex problems and above all making sure we're solving the right problem so it's a way of thinking and it's people-centered we put people first so we decided we're going to call this design 2.0 and then we discovered there already was to design 2.0 so he said design 3.0 well there already is and they have weak side four and a few people anyway we decided to call it peace on X and we actually published a paper on it but we also ran a conference so the next year we've had a conference of design X in Shanghai and we brought a bunch of people who were working on complex problems and I had written a paper for the conference in which I did an analysis of complex problems and talked about how difficult these were to analyze and understand there's a whole bunch of work in existence theory and especially what's called soft systems theory that looks at musicians and tries to understand them and it's very clear that the traditional way of approaching these problems is to try to solve the entire systems problem and come up with a solution well we had this conference and we had a couple of really interesting papers and one of the most interesting paper was a Royal College of Art in London where they had spent a lot of time thinking about the Angelyn system in the city of London and they realized that there was a much better way of doing ambulances than what they had because that right now the ambulances were located in the hospitals and they would be called out with a problem and they might not even have the right equipment with them and it would take quite a while to get to the patients and then to come back to the hospital and so what they decided after a blood study was that they should redesign the ambulances d'italia more equipment and more telecommunications informations devices so they could radio back and to the hospital and show pictures and medical information so they would be ready for the patients and moreover the hospital's the ambulance should not be at the hospitals they should be where people needed them so they decided a ambulance should be located throughout the city and they had historical data about where the incidents occurred so they presented this big plan they won some prizes for it the Mayor of London loved it and said yes that's what we should do it was a really nice paper but and the question period I said that was a really impressive study that was done a few years ago so in fact ten years ago so you didn't tell us what the results were tell us how effective that was and he said nothing happened it was never implemented and the difficulty is you can't implement things like this and the difficulty is not analyzing the system the difficulty is getting it done and I discovered that there's a very famous paper in political science quote must not mumbling what's the right word I wrote it down the words that I put what I wrote down so this is my talk this one page and I really shouldn't need it muddling through muddling through and what is meant by muddling through is that the prison you can't implement solutions to these big problems is that any solution is going to take like ten years on top of that is going to be very expensive and so what happens is all these forces come out against you you probably have seen this when someone introduces a new transportation system or a new structure of the university oh let's combine several each University to make one big University to take an arbitrary example and what happens is that lots of people are against and all the people who are four are actually quiet but the people who are against are ferocious and it's also going to cost a lot of money and it may cause harm because if you're going to deal with transportation system you're gonna have to tear up roads and streets and you may even absolutely move homes so make room for it and it may benefit the public but it hurts people in the middle on top of that because it takes ten years there could be all sorts of unknowns that happen and so it'll probably cost a lot more money than you planned and take a lot longer than you thought if you are able to do it at all and so please be the coach which is called muddling through is do your analysis figure out where you want to go and then try to find some very small steps that moves you toward it because if it's small not very many people object if you maybe that nobody objects it's also easier to get the money it's also easier to do and you can also do it in a relatively short time and if it fails well nobody notices if it succeeds that makes it a bit easier to do the next step and so what you do is you basically keep taking small steps always going towards the goal but because you're doing small steps over time as the requirements change you can modify the goal you can change it you can be flexible until that's our philosophy you're doing our work we decided that if working on healthcare or an education or problems like that there is no solution but we can make things better and so what we try to do is look for opportunities to make some changes so let's go back to the automobile example so how do we solve the problem well we started trying to understand how cars signal to suggest trees we call the road users because it's everything you can imagine anything it's in the road and we look at how it's done today and it's done by high movement I gave in some cases by deliberately looking at the driver of the car and seeing that the drivers looking at you or in some cases especially in some countries like Mexico the drivers of the car are very careful never to look at any other car driver or pedestrian because if the pedestrian thinks you haven't seen them the pedestrian will be out of the way it's a pedestrian or the other car thinks you've seen them then although they can go ahead because you won't hit them so there's a lot of signaling that takes place and sometimes people use lights and sometimes people use the horn lights are interested because it's not obvious with the light flash means if i flash my headlights what does it mean there's a very famous story about in britain if you two cars are coming together on a narrow road and you're going to cross the narrow bridge if you flash your lights at the other car it means please go ahead and if you're in mexico and you flash your life it means i got here first i'm going so now imagine there's a mexican driver and a british driver coming together so lights are ambiguous so how would we signals you actually take a look at some of the commercials mercedes-benz had this wonderful commercial where the the driverless car pulls up and that it it flashes on the ground in front of it a walkway the pedestrian knows they can walk across the street well none of these things work so when you start having what we call scaling up a lot of drivers a lot of cars a lot of road users sure you can cross but there's another car coming and it's going to hit you so I really can't tell you that it's safe to cross so how do we do that and so the more we thought about that the more we realize that it should be the other way around which the car should say what his intentions are now right now they're only two signals on the car that give anything at all about the cars intentions one of them is a turn signals the turn signals is basically the only thing that really is pure intention I intend to turn left or I intend to turn right the other is the brake lights and the brake lights in a sort of combination of this is the state my foot is on the brake pedal but simply this when you first start the brakes you aren't really slowing down yet it's also kind of I'm intending to slow down or stop but the problem of the brake lights is that's on the back but if I'm trying to cross in front of a car or if I see a car coming at me I want to know if it's slowing it will be really good for the road user to have the brake lights on the front so that's one issue but even so there are more signals we might want again I'm about to stop about to go I'm going how do you do that so as we did this more and more and studied it and we get lots of filming of people around and we we started driving a car with no driver around the streets of the city and the street to the University and we discovered among other things that there's huge cultural differences we all know there's a big cultural difference between driving in India or China or Vietnam and California and there's a cultural difference in driving in California in New York and there's cultural difference between here and Spain and Italy and Lisbon and it's it's really interesting those very very different cultural differences but we discovered cultural differences in the city of La Jolla and in the University of California which is in La Jolla because in the university the students own the streets and when it's time for them to go they go if that's across the streets they cross the streets and they're reading their phone they're reading their phone and as they read the phone they do is they read the phone and they know they're at the curb until they do that and figure out where the curve is and Bob lay still read and they never ever look at the cars and the cars are expected to stop and give them away and when it's between classes 10 minutes between classes of a car coming to an intersection that's it Makar we'll never get through but the students keep going and if there's little pause and the students other students are there the run fast to catch up so while the cars are so soft so you get thrill but you have to push your way through now your different ways of pushing your way through that on the college campus you have to do it very slow and gentle no way when you're in downtown San Diego or La Jolla you can do it a bit more forcefully when you're in Shanghai you just do it and in fact if you look at the movies of Shanghai pedestrians order to Vietnam what happens is basically you're waiting for a big bus to come and when a big bus comes and makes a right turn say that what happens is all the people on motorcycles and cars and pedestrians going on behind it in the shadow of the bus because the bus is breaking the way and then you can go along how is that going to work with automatic vehicles how will you handle those cultural differences well the one thing we decided was that is better be standardized if the world does not if every car manufacturer tries to signal differently it will be chaos so even if the signals are not the best if they're all the same it will be better and we think the intentional signal is what's most important because if I know what the car is going to do and then I can control I decide I know what to do so how do we get standards done it takes years together standards of proof but it's necessary so actually today is Friday and tomorrow I fly home so I'll be home Saturday and Sunday I fly to San Francisco and we have a big standards meeting all day we fought together four or five automobile manufacturers and a bunch of researchers and few people from the Standards Committee to start this process of standards is that design yes it is for this design of the sort of design X which is a social technical problem it's require society it requires politics it requires negotiation it's going to require laws and regulations it's a very complex problem that's design X let me tell you about the medical problem talk about cancer because we're working with the cancer people we're working with about four different medical school departments and in the hospital of the clinics so a child has a problem pain in the leg and it takes the child a week or two before it bothers to tell her mother and the mother well it'll go away but after a week or so the mother says well maybe we should go to your physician so they go off to the physician but it takes another week or two to get the appointment and to actually get in to see the physician and the physician says I don't know you need an x-ray and so now we wait a little while to get the appointment for the x-ray and you go and get the x-ray and then you go home and the radiologist has to leave the x-ray and then has to call up the position and explain what is sleeping and then the physician calls the family up and the family has to make a new appointment to come in so we've already had how many different specialties we've had the position we've had the x-ray technician and we had the radiologist at least three and now the physician says we don't know you what you need is a biopsy so here we go again another appointment and then a pistol sample and then it's read by the clinician and report it back and now another appointment to go back and hear the result there well it looks bad I think you should see an oncologist and so they go see the oncologist another set of appointments another special thing the oncologist says you need an MRI and now another set of appointments and another set of specialists etc and then they have to decide how it should be treated if it's cancer it's by usually it's chemical or it's surgery or radiation suppose they decide it's radiation well now you have to go see a radiation oncologist and actually it's a child so the pediatric radiation oncologist and they have to decide how you will be treated and the traditional treatment by the way is a small dosage about 15 minutes a day for 30 days which means six weeks you have to come in every day for six weeks to be treated and that's really hard they say the mother works or they don't live close to the clinic but that's the treatment and when you treat it you have the pediatric radiation oncologist you have the on-site radiation oncologist you have a medical physicist you'll have a dosimeter you have a therapist so you have a few other technical people when you're all finished this thing can take six months before you start treatment and twenty different medical specialties every time there's a transition from one clinic to another one specialist to another there's a chance for ever errors occur in the transitions when you do a diagram they love to draw this diagram the box showing you're doing this and then the box showing this in a box here the box here in a box there with arrows connecting them you may have drawn these boxes yourselves and nobody ever asked what's inside the arrows the arrow yes it's a transition from here to here but it implies the most important part of the transition communication communicating everything that was learned here to the people here and the medical system it doesn't work none of all the information is there even in the complete medical record because if I'm a physician and I say here's what the problem is but you know I have this slight kind of queasy feeling that it might also be this I don't put that one in the medical record but if I were the only person seeing the patient I would remember it and it might come back later on to be important so that's one of the problems who are looking at and we're working with an arrow with the National Cancer Institute which says we don't want you to treat just that issue because to us cancer is a lifelong disease now once you get cancer is there at home you discover you have it you go and you get diagnosed something you get treatment and then you do not go back home cured in cancer you go back home to what's called survivorship now you may survive the rest of your life and that more and more people do which still you have to be monitored because you never quite know so it's that whole cycle we want to look at it as a system so we're working on that and actually even this morning I've had interaction with the National Cancer Institute in Washington discussing the progress of what we're trying to do so those are the kinds of problems we work on now what is the method that we use the best is that we called or giving it several names over the years user centered design human centered design and today we're saying people send a designer I've also said activity center design but it's all the same thing user centered design is what was commonly called that we still talk about user interface but I decided I don't like calling you users so I started to call you humans but after a while I say I don't ever call you a human I call you a person so that's why people say great design but for some things actually when we're trying to design for many many different people or even for one person doing many different things yes we want to Center on the people but what's really more important is the activity so to my mind activity center design is a kind of people center design there's even more specific depends upon what you are doing which is also wide by the way if I'm designing an automobile for Vietnam or for I don't know Lisbon the activity is still the same and therefore the automobile the very same automobile works in both places the way the cars are driven are dramatically different but in terms of the design factor the design is the same because that accepts the basic activity is the same so what is activity center design or person centered design what we start off trying to study the people and when we talk to engineers now and they have great trouble because when a new product is being designed the new product team is assembled the engineers of the programmers want to get to work right away because there's not much time so the program is always thought programming and the California tears when I start specifying dyes and and equipment but how do you know what you're designing how do you know what you're building so one of the first most important questions we ask is what is the real problem we're trying to solve and this drives managers in companies crazy because the manager comes and tells you the problem that is to be solved and the designers say now let me think about it so it turns out that when you start trying to figure out what the real problem is quite often the problem you've asked to solve is not at all important and quite often if you can solve the real underlying problem the problem you're asked to solve disappears goes away I mean here's a slightly different example but let me tell you about the jet engines that are made by rolls-royce or General Electric it make these huge big expensive jet engines that they sell to the airlines what's more important itself to the manufacturers like Airbus and Boeing and they used to ask them what their what the specifications were in the course and they would and the airline the airline's who buy the airplane by the way often get a choice of which brand of lauder they want on it or engine and the companies suppose rise to started said we realize that they didn't care about the engine so we don't sell down the engines anymore we don't sell our engines we say we will guarantee you I don't know a thousand hours of flight per year that's what we will sell you and that's actually much better for the airlines but now they don't have to hire technician to repair the engine to take them apart then I'll still worry about when they need service Rose Royce takes care of it all Rolls Royce puts the engine on the airplane the end the airplanes have all sorts of sensors on them would send back information to rolls-royce so they know exact state of the engine so they can let it go longer without maintenance than normal or they could make it be heaven immediate maintenance if necessary so they know the state of the engine in fact there was a recent crash of an airplane in Indonesia that got lost nobody's ever found an airplane in there they lost contact with the airplane watching the engine was still in contact with the manufacturers so they know actually where the airplane was roughly because they got the information from the engine maker so that's solving the right problem and I was just entering the equipment maker and Modena Italy where they make this wonderful machine that puts pills in boxes it's a credible machine costs about ten million dollars and what you do is you pour the pills in on one side and then you have big rolls of paper and plastic and aluminum foil and out the other side as fast as you can see there's this box is shooting out which are a box which is printed and spanned to the date and inside it has the cardboard sleeves with plastic bubbles in them and inside each plastic bubble is one or two or three pills whatever the amount required was and it's the whole thing is completely automated the company sells them to companies that make pills and now they're starting to say we're not there to sell it to you it's going to be like the engines we're going to say will guarantee you a million boxes a year whatever number you make and we'll take care of all the maintenance stopping borne you'd start doing that now you realize the design is going to change the design of the engines have to change because we need all those sensors to give us the complete information so we know where we should do it the same with the film making machine say with tires big truck tires a good year now in the United States are not sold to the truck companies they sell them a certain number of miles of where and the tires themselves are highly instrumented sending radio signals to the truck which has a larger Wi-Fi connection so they always know how good the tires are whole different way of doing things so that is what is starting to happen and that's activity-based but it's also let's figure out what the real problem is that there is a solve so we do what the Japanese call the five why's when we're giving a problem to solve we say why is that a problem when they get an answer we say why is that so and then we say why and we keep asking why do we think we have what is the fundamental problem that we have to solve so the first step is solve the right problem the second step is what how do we work around the people what is it the people need and then we also have to figure out what and focused solution is so we have two phases phase one is where we start trying to figure out what are the possible basic fundamental underlying problems and then we come back and decide what we think we've got and sort of sort of as a diamond do we expand what the possible underlying problems are and now we figure out what the problem is and now we try to think about all the possible solutions and then we try to converge upon one that's feasible sensible affordable reliable doable and that's what we call our process but there's one important part of that which is we don't trust our results we're designing for people it's very difficult to know what people will do so we actually have to test and so we try to test anything that we're suggesting and get back a result right away now normally when we tell engineers we need to have a prototype that we can test and see if it's the right idea well failure it will take us six months to finish coding or six months to finish building and our answer is no we want to test tomorrow and so what we are really good at is theater we're really good at making something that looks like it's working but isn't I mean those of you who remember the old poem pilot it was called very first small little thing you carry around in your pocket the way it was modeled at first was basically you just got a piece of wood which is the same size that you hope you could make it at and you carry it around in your pocket all day and when you went to a meeting you took it out of your pocket and you took out the little plastic scribe but you may believe you are writing on it well that tells you whether the size was going to work and then for other things we make plastic models or we make foam models or wood and we use PowerPoint to demonstrate electronic circuits you just draw the picture on PowerPoint or you just draw a picture on paper and I put the paper screen in front of you and say what you do you touch that okay and I kick the paper away put downward the new screen will look like how do we test the driverless car driving around the city or in the through the streets of the university well what doesn't really have to be driving us it just has to look like it's driverless so what we did is we went to the store the upholstery store the automobile upholstery store and we found the upholstery that was already used in the car seats and we bought a whole bunch of it and we made it what we call a seat suit so it's a car seat suit and you the driver puts on the same upholstery that is being used by the seat and sits there and looks through the weave of the upholstery and we have two little holes at the bottom maybe stick your hand through and hold the bottom of the steering wheel and so if you look inside the car if you look carefully you could see the driver's seat was thicker than the passenger seat but no one looks that carefully so when you look in the car you don't see anybody so that's how we're testing what it's like and how you to signal from a driverless cars or pedestrians so we use theater a lot because this prototyping is really important now that's sort of basically what we do and what we're doing and there's one more point I want to make though in this talk which is because most of you are engineers and a standard statement among engineers and I know this because my first career was engineering I started off at MIT s and let's go engineer I went to University of Pennsylvania I got a master's degree too let's go engineering and I didn't switch over to apply awhile later and in engineering basically this will work perfectly the one for all those people people really mess up what we're doing but they do the weirdest things and what's neat today is those real things are kind of fun my job is to figure out a term a something that is really so effective that doesn't matter what here did you do with it will work and you may end up using it the ways I never dreamed which is wonderful that's the real ciyon success but the first thing is that we have to focus upon people and we have to therefore work with people and people people who understand people have to be part of the design process from the very beginning you can't add this at the end these either is also what the sort of help you decide what you should be building in the first place and I try to remind engineers that while they're engineers design build create things well it's for the benefit of people in society isn't it so don't you don't should we know more about people in society and even in theory classes the theory eventually is going to be used in a practical form that design is a strange discipline because most of the disciplines of the engine in university are analysis and we get better and better analysis and professors are usually promoted by being really good at it which means they have to be more narrow they become specialists and more and more specialized and so we have specialized departments and then within each department we have specialized disciplines and design though is different because if you want to build something you have to understand all the different disciplines and put them together we have to understand the Social Sciences the engineering Sciences we have to understand materials and technologies we have to understand business and its really attractive and we have to take all this into account in doing the design but design is whether engineering or not is for people so the first thing is that the second is you have to go watch people I feel that I'm an expert in people's behavior and even so every time I watch people doing a task or I would we devise something new and I'm trying to assess it and watch I learn something I am always surprised I'm always surprised by what people will do and engineers have a tendency maybe some of you to be very logical they say how come you can't use it it's really simple let me explain it to you well but you're not going to be there when you're using it so you have to actually build for people the way they are and there are two things we talk about expert users and novice users you know everybody is really an expert at something and they're also an expert at the thing they're trying to do with your tools there are novice in using your tool but don't confuse that with not being an expert of something but in addition I have to remind you you all know about IQ scores right and probably all of you have pretty eye I shoe scores that's why you're here I want to tell you that half the people in Lisbon have an IQ less than 100 half the people in the world have an IQ less than 100 because that's a definition by definition the IQ the meaning is a relatively symmetrical distribution in the meaning is 6 10 100 means half the people about half the people are below and we have to design for everybody and we have to do that by watching and on top of that not everybody is fully functioning we talked about designing for the elderly and I say that stupid we shouldn't be trying to be elderly we should design for people who might have problems or who might have motor control problems but some of them could be young as well as old a lot of the elderly are quite functional I mean what's the definition of elderly someone over 60 65 I'm 81 I hate to be classified as a builder waited for a elephant so what some people can't see very well and it isn't just the elderly right I mean woof actually I now can see very well without the glasses for vegetables I don't have lenses I had cataracts so they took out the lens and put in pieces of plastic and the main reason I need glasses is in those days they couldn't correct astigmatism but also now there's a piece of plastic that can't focus so if I want to look at something that's not far away I need this to help me but almost everybody is when you're need glasses basically after they reach the age of 45 roughly because the normal lens gets hard and no longer focuses so lots of designers especially young ones love to use tiny little print and because print is ugly Apple especially thinks pretty ugly it destroys the beautiful subsisting of the devices so we use small print so it isn't so visible and thin and light grey prints on a light grey background so it doesn't disturb the elegance the hotel I mean now is wonderful it's the hotel white board it's got this beautiful well I came in and there's the controller for the TV said did you try it last night could you work it and there's a controller for the TV set but where is the TV set oh come look there's the wall of mirrors making one of them is a TV set so the total boom boom nothing I finally crawled up the front desk and they said bill sending somebody up and it turns out the middle mirror is the TV set okay but it doesn't work no no you have to point them right down there at the bottom in the center of the bottom in design we call signifier first of all you have to discover abilities how you know what you can do and second of all there should be little some signifier which comes from semiotics but that says this is where you do it so I'm going to go back and tell the hotel just get yourself a little colorful dot green red whatever color you like and put it right there militia game for it and they just have a little notice someplace it says to use a TV set aim at the red dot because it's what I asked the person helpless does anybody else have this problem every day he said and last night we know checked into the hotel so while we were there we asked the people signing in to anybody have trouble with the TV set all the time they said whoever he signed it was someone who really made it look beautiful it is pretty but if you want to use it how much time do they have to spend these people coming up all the time to give them with their help of the people but you see if you would have thought about that or tested it there's a trivial solution which I just told you we're ready in fact for a long time I used to walk around with a little punch strips of colored little dots keys I never know whether I should put the key in and turn it to the right or to the left because each one is different so I used to go around little green dots and put the dot on the direction that I had to turn the key and automobiles we got two cars and one of the new gas tanks on the right but on the left you could never remember which no green dot realize it which side nowadays actually the cars that have a little so the gas tank was an arrow saying which side the opening is on think about people last point we tend to build things that we can build and then it doesn't quite work every part so the rest we leave for people this means that we're building machines like an automobile and even the most advanced quality autonomous cars today are not fully and so the person has to watch and take over and that means the person has to behave like the Machine and we're particularly bad at being machines we're very bad at things that require high accuracy a higher that require a lot of attention for periods of time when nothing is happening that require repeated activity the same one over and over and over over again with high precision or high accuracy we're very bad at those things our memories are not perfect they're imperfect we're really good though at creativity we're really good at knowing what to do when there's something situation suddenly is completely unusual that never has been experienced before we're really good at lots of lots of different things we're very flexible we're compliant and our memories may not be accurate but you know in the real world they don't have to be accurate what our memories are really good at is especially as abstraction is and recognizing or this situation is kind of like something I've had before and knowing how to respond and not knowing that don't remember the details is actually helpful in matching some experience to some previous experience but the machines take that away from us so here's something we all know that people are distractible and that's why we get a lot of accident because we're so distracted and I say that's false we aren't distractible what that is doing is taking one of our great virtues that we are very attentive to things that happen around us we notice nutrient oh look at that and we put them together and we are creative that way that's what really adds to our creativity our curiosity that were sensitive to changes but when we're forced to act like machines this kind of curiosity and sensitivity things that are happening around us is a deficit so we call it distraction and we blame people because they're distractible so we build our machines so that people have to act like machines which is a people are bad at and that when people are bad at doing things that they're bad at we blame them and we say you see people make all these problems so we have to solve them by getting rid of the people and I'm saying no what we need to do is figure out what people are really good at which is actually the sort of thing that computers are bad at and that computers do the stuff that we're that the stuff that requires precise memory and precise activity and precise responses and let us take care of the importance and just like a calculator takes care of the mechanical parts of calculation but I have to think about the issues really hard in writing my equations and why not do that and make a much better what I call in a fact in our lab we call human technology teamwork you know what the best chess player is in the world it's not deep blue the computer programmer and it's not it's not a Russian you know world's Grandmaster it's a combination of good chess players and good chess machines not the best chess players in the world are not the best chess machine the kind of chess machine you can buy any of you can buy you combine those together and a team of them can be the best human player and can be FIFA loop and that's a wonderful example of technology human teamwork were active collaborators each of us doing things that were good at but the combination is better than either of us could do a lot and that's what we're searching for and that's my talk thank you [Applause] [Applause] to see Donna making a beautiful talk no swallow in perfect timing I feel old so we have 20 minutes for questions we have some people with microphones we will [Music] I think there hi thanks for the great talk so my question is related to learning and education in the classroom so in your opinion in terms of people center design what would you change in the traditional classroom or in other words what the traditional classroom gets most wrong how can we start muddling through that problem Thanks well I told the head of our University the first thing I would do is get rid of departments and and I would really think we should change the whole educational system but before I tackle that let me just try the question there are uses for lectures and I mean I have to say that I just gave one but what I think a lecturer is good at is basically what I just did it's good for meeting new people it's good for giving you a general framework is good for motivating you it's not a good way to learn a topic in depth to learn a topic in depth you actually have to struggle at it work at it you have to really work in fact you learn best if you are in trouble you get into jet trouble and you figure your way out and that means teaching instructing we have to figure out how to get people into trouble where they don't know what to do but it has to be just the right amount of trouble because they you don't want them to give up you want them to persist until they get it and that's when they will learn but if it's too much in trouble they'll give up so that's a problem so I feel that lecturers should be really occasional in a class maybe even never and the reason I want to get rid of departments is that they become too specialized and quite often in sort of out of touch with the world and what I want what I prefer is problem-based solutions and problem-based education and for that matter why not problem-based Department and what's nice about a problem organized department if I want to work on environmental influences or something I would need to have chemists and physicists of vertical scientists and maybe psychologists and sociologists all working on that same problem together it forces going across the boundaries now we still need specialists absolutely we need to have people who are experts into given topics because without the experts in the specialized area we're not going to solve the problem as well so we need a combination of these two now in the classes what we like what I love to do is and I it's getting more and more Commons problem basis so one thing I like to do in class is one day one I give the students a problem to work on and they go and it has to be one that I that they will find interesting so I will either try to find one that I think everybody will find interesting or I'll select two or three for them to work on or four and they can choose whichever they find the most interesting and they get to work on it but they will variably get stuck and here's why I like that is because if I give you the fundamentals first the fundamentals are pretty dull and if I give you the philosophy that's really go in the history that's dull and so but you can't solve the father without knowing the fundamentals but they're dull when I just give them to you but if I give you an exciting problem notice how I opened up the talk with what I thought was a pretty interesting problem the car was no driver knowing what each across the street that beginning to give you a framework and then when I could come back and describe the method that we use you can begin to see how that might apply if I start up with a method you wouldn't know what I was talking about or why I was talking about it and so by giving you an interesting problem and you get stuck then I have you leave the fundamentals and oh it makes sense so that's one thing I would do and that people are now experimenting a variety of things problem-based courses yeah what's called the flipped classroom so the material which today is taught in lectures and then you have homework you do it at home we flip it so we don't have lectures anymore we have we have assignments that you read at home or more and more computer and video now a lot of the so called MOOCs massive online open courses have lectures and I think that's horrible because if lectures don't work well in the classroom why does it get any better if you watch it on TV it's a little bit better because if you have trouble understanding something you can stop it and back up and listen again but aside from that the best thing is to have very short pieces of material and then basted into maybe some problem exercises based on it where those problem exercises are not meant to test your intelligence they're meant to force you to review what we just heard or saw and think about it and that helps retention and if you didn't get it right no problems go back and listen again and try it over again so it's a different way of teaching and we call it the flipped classroom because at home you now do the equivalent of electors and then the problem is you come into class and you work the problems and here's where the teachers whether the professors or teaching assistants walk around helping you solve the problems and so it's you get the help and you needed but if professors give the same lectures year after year after year roughly then what I could be televised and better yet it would be done better because you get a televised you can spend time to perfect what you're saying to give demonstration and show demonstrations so it might take actually close well I know it takes 10 to 20 hours to do a reasonable half hour lecture if you're going to be televising it so we actually have a bunch of courses like that available I have a short course that covers two chapters of my book the design of everyday things it's available in your dasa t and you can finish it in and easily in a couple of days but it has a lot of exercises along the way and what you're asked to build and create things and put the results up on the internet and all the other students who've been taking the class have their results and you can start comparing what you do what they've done like here's one of the exercises make a box and you can put down and leave and go away and other people will be attracted to the box and will pick it up and turn it all over and look at all the sides and you're not allowed to have any instructions that's one of the exercises and it's actually quite a successful one it's amazing where people come up with a really clever ideas and then watching what the other people have done cleverly teaches you a lot so is that an answer more questions I have to wait for the microphone so I do I don't select to ask a question thank you for your lecture with a great fun I have actually two questions the first one is when do you think I should be ready with the skills to cross the street in the secondary any time you cross the street because what you're saying is what is it what are these all fully automated car is going to be but actually I think it's more dangerous today there will be when those occur because one of the things that is true about these fully autonomous cars is there's actually a danger that they have today they obey the law the Google cars have had many accidents and I with one or two exceptions every accident was they were hit from behind by a car didn't expect them to go so slow or to stop so the other thing is I just wrote a little peacefulness oh it's the automatic car how neat I know is programmed not to hit me I'll just walk across the street right I don't have to pay any attention so you don't have to worry thank you I'll try not to the second question is I couldn't agree more with the idea of doing away with departments but it sounds a little bit like spreading ambulances throughout London which small steps would you recommend along the way changing the university system will own the educational system the change in the university system is a design x problem absolutely and when I when I give this idea to our provost and university head and even my beams they they smile at me they might even agree with me but come on down they say but I think the way to do it slowly is first of all we are naturally moving towards the flipped classroom and we're also naturally moving towards problem-based education not everybody by the way there is no single method of education that's going to work for all people in all subjects so when I'm an advocate of problem-based education it's not going to work in everything and so it's but in the kinds of areas I'm working as that would be better but it also requires a different kind of classroom so we're building a new building called the design innovation building and we're going to have a number of different departments and some from visual arts some from us one from what's called the Innovation Centre and what we need our Studios because when you're doing problems we want we like the problems we've done by groups and the problems can take a long time or a short time and we want the students to be able to come in any time or even though there's a class assignment they ought to be able to be doing your work and working on building something and leave it and go away and then they come back it's still there and so that's a really difficult scheduling problem but there are already existing lots of solutions you can look at we've looked at design centers around the world and one way is you have these lots of these boards and you can put post-it notes but also you can post your stuff or fold it up or put it into lockers and they're on wheels so you can wheel them off to the side and keep them out of the way but bring them back when you need them so problem-based is by starting to have problems based classes and most professors by the way once they start doing them start to like them they may resist it at first but they end up coughing not everybody but also really liking them the other thing is and once you start doing problems they I think that's sort of the the technical term is the camels nose under the tent but what this means is you've made a little step which makes it easier to do the next step in the next step for the next step that's how I that's one way we're trying to start the other thing is by the way our design lab is I have a difficulty with it is that I'm not in any assigned area so when I have some problem I can't go to my Dean for help I have to go all the way to the provost which annoys him sometimes but but when I said well maybe I should be reporting to some Dean or a couple of Dean's he said no that that this was set up by the Chancellor on purpose not to be in any department or in any school because we work with the School of Social Sciences and with engineering and with what's called global policy and strategy and with Arts and Humanities and with the School of Business and it's really very nice that no one thinks that we're in one thing versus another so we can work across and we can put together problem-based well in our research what we do is to go around the campus and find the relevant people asking to try to convince them to come work with us on this interesting problem and faculty like that first of all and second of all faculty like when we say we also have money to support your students so that's how we're doing it yeah one question not here in other words you hi and I agree with you that standardization is the solution for autonomous cars but the problem with standardization is it's not profitable Apple hasn't removed the Lightning port on their iPhone because they want to sell more lightning adapters and lightning Chargers and my question is how do you Stan the term standardization profitable take Tesla for example while everyone is left behind debating what's best for people how do we make cars safe how do we implement this in the large scale they're just painting the way through to autonomous driving and they might one day become the apple of the Apple of autonomous driving so how did you make that I know when I was at Apple Microsoft was a dominant player and we were weak and so it was my job actually to cause standards to happen and so I went to the FCC in Washington and we got the Wi-Fi banned and I went to the FCC in Washington and we said we the TV standards you're bringing about the High Definition TV are no good for computers and we got them to change the standards so that we actually had some things we had something that the computer people could live with believe me each of these exercises was non-trivial it took for TV it took two years and a lot of great a lot of work and in addition I tried hard to get standards so I got for the computer system so I got together with IBM and for it so long I don't remember names but most of the computer manufacturers and we were all trying to bring in standards and the one person who wouldn't play was Microsoft and Microsoft came into our meeting one day and said well if you really want standards you should just all use Microsoft and so and one of the lessons there is the leader never wants standards and the followers always want standards now that Apple is the leader it doesn't want standards now that's not completely true I have a feeling that the new standards of a connector is USBC which Apple already uses in his MacBook Pros and so actually Apple is blessed got rid of everything in the MacBook Pro except for the USB C so if you want to have a power adapter all you have to have a new kind of transformer so plugs in USB see if you want to go to area HTML that's another standard projector you have to have added app sure that goes from it if you want to go to a normal USB that's a double standard yet have an adapter to those of the USB seeing us it's a mess but this is actually a better thing that some day will happen because first of all it's very small second is very powerful be charged very fast and it can handle everything all sorts of data rates which therefore means high definition television not not when it's called high definition but what's called 4k or even 8k television and pay guess why it doesn't matter which way you plug it in so Apple is still experimenting with using it and that may happen but you're right the joke goes the nice thing about standards is that there are so many of them but I disagree in your comment about Tesla I think Tesla's reckless Tesla does not understand that there are people in there and so for example they got people to trust their driving so well that one person killed themselves because the over trusted it's very difficult they have people to trust things but the danger is you have the older trust that in fact they may get killed and I've watched some of the tests or stories about their programmers are not using if you're doing something that's life critical the programming practices you follow better be life critical programming practices which the military is established which the aviation community has established which the medical community has established but the automobile the Tesla and the the consumer goods don't follow it because assume the goods people are quite happy oh we released it or because people want it fast and it has lots of errors that's why every week it seems like you have to upgrade or modify your operating system that's not going to work at the lights critical safety thing the number of Tesla vehicles out there is still tiny my new I've driven a Tesla or I would say I met the test would drive me it's kind of scary is it safe for yeah but it doesn't feel safe it's wobbling all over the road because it can't see the lane markers it doesn't quite know what to do so I'm not happy with Tesla Tesla's getting better they're learning Tesla says well we do beta tests we sell our vehicles and all the drivers are beta testers if the drivers lives we're talking about so no I'm not happy with that by the way the standards that we have if you do the standards right they don't tell you how you must do something they tell you what the results should be so the proper standards are the result the functional standards it should be able to pass these kinds of tests you can make you can do it any way you like and believe me when it comes when it comes to profits yeah you make some profit selling the accessories but that's not where your profits come from notice the third parties that make lots of this stuff anyway next question yeah hi thanks for your talk and for your amazing books and you talked about the project of literally sided passion innovation and then I have something to share with you and one question I'm a designer and I have a disease a chronic disease and the few years ago I discovered that we had a lot of problems related with the data recording kinds of things but the problems that we talked about it so then the other example that will speak this book here and my question is being a designer or an engineering is the more easier for us to have the big pictures of the problem that we are facing deal diving in the $33.00 disease or doesn't matter oh the more information we have the better absolutely we have to sell examples we have experienced I'm not an expert in in medicine I'm an expert in behavior but that means I must work closely with experts in medicine when I stand be set up centers of excellence I find a faculty member who who's an expert in that area who's excited by itself in your health I don't leave the center for health realize Spencer a friend of mine who's a physician leads it and we have a wide you know a large range of positions that are helping us do this including the department chairs of all the departments the department chair of emergency medicine of radiology of oncology there and public health they're all working with us and they also are in charge of the clinic so we also work with a physician is running the clinics it's absolutely necessary to have a large understanding that doesn't mean though that the every designer must have this but we must work with people who have that because of the question okay last question here you're running out of time actually I shouldn't allow this because the director lost it in the time our advisory board in the last Richmond we're going to offer an exception I'm trying to compensate okay to told it was a very very inspiring talk I'm curious about the part you mentioned the the signaling from the cars to the people and all they interpret each other and and I know that for example so we humans exchange a lot of information without actually talking without speaking so there's a lot of bytes going back and forth just by hearing these other channels and and and people are looking in the brain in your sites want to understand a bit of how we do this because we are not always aware do you think this has an important role in your research in these kind of problems to fully understand how humans do this magic information exchange without knowing and understanding it we need to develop the understand is better than to address the communication between devices or cars and people so much of human behavior much of our behavior especially expert behavior is subconscious we're unaware often of what we're doing or if we are aware of what we're doing we're often unaware of why and so it we try hard though to try to understand what people really are doing and we do a lot of video recording and we don't ask people what they're doing if you ask people you get you get responses but they're not very helpful and what so we do a lot of observations but the real issue though is that we don't necessarily want to build the autonomous cars to work just like people in fact I think it's a mistake to try to copy people it's the old story is you know which we've airplanes don't fly the same way that birds do they use the same aerodynamic principles because that's physics but they obviously don't flap wings and they dart the wings are rigid that are flexible etc so what we need to do is find solutions that work that are effective they don't have to be the same as people and one of the promising directions that we're exploring this is filling the research bases we don't know the answers is motion that we discovered for example that if we were coming up to a crosswalk - please stop our car two meters behind the crosswalk that's a clear signal nobody even notices it by the way this and we only noticed it after analyzing a lot of videos it's a clear signal that I'm letting you go and if I sort of edge up and I'm now one meter all through the crosswalk that kind of a signal well maybe I'm stopped but if you watch people they're kind of nervous going around that car and the same is motion and when I think I've been waiting long enough and people aren't ready to cross and I need to I'm going to go well motion by small motions or a little jerky we're even thinking of being able to lower the Siq article they kind of let me through and one thing I noticed in Milan was they were credibly I didn't realize that the Italians are so polite in driving their cars the light turned red and all the cars stopped and there was a lady walking across the street she started with it was green for her but it turned red again when she was only halfway across the street so the cars are patiently waiting for her to cross the street and there was a motorcyclist who just really rolled he really wanted to go and so what he did is he went with his engine but it those are actually not bad signals they are signals that I'm doing something but it's safe right because it wasn't moving and so we're looking at a lot of the real subconscious cues that people use and that the automobiles might be able to use okay we are running out of time but we do have post lanterns served in the sponsorship and working I would like to invite everybody to go there before I would like to get to thank the sponsors karsenault poster tangible and otherwise recovering the water and drink and also in particular I think she already left luge for such participating their team also from washes that helped in the logistics thank you very much and thank you [Applause]

Channel: Design Lab

Views: 21,842

Rating: 4.967742 out of 5

Keywords: UC San Diego, Design Lab, Design, ICT, Automation, Healthcare, Cognitive Design, UCSD, Tecnico Lisboa, Autonomous Cars, Design Thinking, People-Centered Design, Don Norman

Id: kgRjZYhDf5U

Length: 83min 42sec (5022 seconds)

Published: Fri Jul 28 2017