Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Designing around the brain

UXAustralia
August 30, 2019

Designing around the brain

UXAustralia

August 30, 2019
Tweet

More Decks by UXAustralia

Other Decks in Design

Transcript

  1. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 TEA UGLOW: Thank you. It's not that tough. I do 24 hour flights all the time. I fee sorry for Carmine, and congratulations to Ruth and Jasa, I'm not sure what a Home Hub is, but I'm sure it's fabulous. My name is Tea and I run Google's Creative Lab in Sydney, and I have done for about five or six years. I'm going to have a little chat today about... One of the most interesting things is, despite doing that job, I've been doing it for 12-13 years but only recently started talking to UX audiences. I'm not even sure whether UX was a thing when I started. When I first started in '99, I was a designer, but I really should have been paying for the privilege of being a designer, because I had no idea how websites work, I had to have the notion of a server explained to me several times. And my design skills were...limited. I think I kept running away from working with computers. But I was part of the first dot.com boom , then I ran away for a bit and then in 2006 I joined Google which seemed like a giant American multinational corporation but only consisted of 5000 people. So the last 14 years have been something of a (inaudible). Designing around the brain. People have been telling me how to improve my presentations and one of the things they tell me is not to stand too close... Not to speak directly into the microphone... I will get used to it, don't worry. I've done it before. I want to give you a kind ofrRough overview, but it is a lot of words and a bit boring. And also very structured. And I don't really do structure. We have these new notions of structure that rea much more human – the neural network, which is very human content. I find out when I pull apart what I'm going to talk about, I end up with these rather more unstructured processes. Because you can draw on your networks nice paths and lines that go between them, but ultimately we're not really understanding, they are not following the sequential, logical, formulaic way of understanding the world. Just like we don't. But as we move forward, I think one of the most interesting things, and we're going to keep coming back to the fact that we have been designing for the tools that we use to understand information and we're going to have to start getting to design for the people who use the tools. Because the tools are becoming more and more fluent. It is nice that the Home Hub was given out. It's kind of a good point. Those things rely on voice UX. Has anyone worked with that? And one do conversation design? In five years' time it's to be half the room. Conversation experts, dialogue flow people. The graphics of voice design has yet to be invented and we don't know what it looks like. It is not one of my core specialties. My core specialty is basically mucking about. I do talky things, I do writey things, I make funny things, and I am gay for Google. I queer
  2. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 2 of 11 around the world. That is a thing now. You basically get paid for that, it's great. Some of my talks, just so you know, they are not like normal talks, and these are my favourite YouTube comments I have ever received. They say don't read the comments, but you should, right? Because if you're not failing, then how are you going to get any better? I have been making assorted disasters for 10 years. These are some of them will I. in that we learn something that we thought we knew but we didn't know we didn't know it. And that sense of learning gets lost very, very often when people get fixated on the thing they want to build, rather than the possibility of building something from a set place. It is really very much about whether you're aiming from somewhere or aiming from somewhere. Within my creative career I've had the opportunity to do that with all sorts of things, first of all when YouTube was a baby and we bought it for $1.2 billion, which seemed like so much money. Actually it still seems like quite a lot of money. But the you go. We re something for film, we developed the YouTube Symphony Orchestra, we went to Guggenheim with this new short form stuff. It is like an art form, like video art, we should document it. We did a project with NASA which kind of involved experiments and I think we were going to send a small child into space... As a prize, not as the experiment. And then they said no. We launched this browser called Chrome, I don't know how many of you were alive when there wasn't a browser called Chrome, but it didn't have any market share and nobody knew what it was so we had to explain to people what it was. Yes, you can have a new pair of hands and they are faster and they don't pick up germs. Which is most literally the product description. But it was so hard to try and explain that, so we ended up coming up with things like Chrome Experiment, to try and explore the potential. This is just at the HTML5 moment, when that potential for the browser was suddenly extraordinary, with enormous potential. So we did lots of experiments in that space. Then we had curious products like social products, which as you know, Google is renowned for. So there have been really interesting projects for me over the last five or six years which have been much more about how we interact and how we use our machines and how we interact with information. (Inaudible) and it got me more into theatre, more into the act of theatre, and my home space, which is like the arts. There is a wonderful project up there, the art project, that turned into cultural institution, arts and culture. Sometimes the little things turn into really massive things. And sometimes they don't. This is a project from (inaudible). It is really very old school, just a head pointing at a projector screen that you usingn. No're like this one we did last year. (Video plays) SPEAKER: I feel protected.
  3. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 3 of 11 SPEAKER: t'Is the only thing we've got. SPEAKER: The essence of our family. SPEAKER: Hope, trust, the essence of our family. TEA UGLOW: This is a bigger thing. This is as big as any of my projects. Obviously some projects get fairly big, but I'm long gone by that point. It is not in my interest to take something and a prototype, take something beyond a space where it's like, this is what we do, this is it. And if it is of interest, which it is, but very rarely then. Usually two or three years later. It is fascinating to see this sort of simple UX, take the most distracting thing in the art gallery, and turn it into the thing they need to use in order to experience the artwork or hear the artwork. Make it work so that you can do multi-choice. You can have 500 people all watching their own version of the video with their own soundtrack. That leads to quite interesting things about how you move around space. This is pretty much all I do for a living will stop I do white boards that no-one understands. Then we give them to my team who are creators and. in their own ways who neither understand what the other is saying. I'm sure this is a common theme. Over here we have the engineers and over here we have the people that make the pretty shapes. People who build things and people with pencils. And allowing those two groups to communicate, you laugh at the idea that the designers are people with pencils. No, these are the serious ones and then the idiots. That interface is actually incredibly important. And the gulf between it is hard to bridge. And we know that in any irrelevant process, that is where you lose your product. That is where everything breaks, where the essence of this tool you have built becomes unwieldy, unusable, or just doesn't make sense. I have two small children. They are kind of different. One of them is very much a dreamer and the other is very much a thinker and engineer. Those are not binaries by the way, there are all sorts of people. I had the notion that there are one or another. But it is useful in this space because what to talk about is that my family orients towards one type of thinking. Which is fine, we always have and we always will. But it does mean that the child who isn't naturally inclined that way is slightly othered within their own family. And it's not long before they start to feel like they are a bit of a misfit, that they are not the right one and are the odd one out, and they start to try to re-engineer the way they think. And they're not reengineering themselves to think brilliantly, just to fit in with the family. This happened in your families, you know it does. It happens in your friends, your office,
  4. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 4 of 11 everywhere. And one of the things I would love to touch on is how we deal with difference. I like explaining about leadership. Because I come from a place I have never been able to deal with the hierarchy, leadership. Because there is someone at the top, normally a white guy, kind of telling everyone else what to do, when to do it, how to do it, how loud, when to shut up. And then you get to this place they have decided – their vision, what they're going to do. And I prefer a gardening approach, which is much more like find the pretty flowers, bury them in the soil, and then basically water them and talk to them. And then let them do their thing and then other people come along and say, "What a beautiful garden!" And I'm like, "Yes, I grew it myself." My team don't listen to me, the only point of absolute decision is when I tell them to stop doing things. Because there was a was a point where you have to stop being in control, allowing experts to do what they do really well. And I am not an expert in the things that they do. So there is definitely a point where I need to become a user because that is the nearest thing I can do to being helpful. So I am quite good at that. We all have different leadership styles. I find it's like the (inaudible) industry. We have a notion of expertise which is around what is important. In this era when science is debatable. We should remember that science is debatable. Science is, as his excess, as maths is, anything apart from ballet, debatable. And the way I often construe this and explain this to people is that, do you trust your doctor? Does everyone? Mostly. You probably do now more than you did 20 years ago, because they didn't know anything. And if you trusted your doctor from 100 years ago more than today, you would be an idiot. Because we have learned. But on that logic, if you could move me forward 20 years, without ageing, and if we could move me 100 years into the future, it would be useless to suggest that the experts from hundred years ago – the physicist, neuroscientists, behavioural psychologists, environmentalists, designers, anyone – would be more trustworthy in their expertise. Because it is an evolving thing. It is a continuum, in the same way that there are small children, and as a small child, I wanted things to get better. I was excited about what was coming next. And then I don't remember the bit when everything was OK, great, perfect. I don't remember that happening. But I do remember a moment of, "Everything is going to shit, why can't it be like it was?" And what occurred to me was that every single human who has ever existed as almost certainly follow the trajectory, which means that if you mass them all together, there is a huge wave of people going, "Isn't that exciting?" And that wave will pass on in time from cave dwellers to space astronaut – isn't life exciting? Isn't it amazing what we're going to do? Equally, just ahead of them, is a group of people going, "Oh, my God! Stop."
  5. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 5 of 11 So that is a continuum. Saying that it used to be better is a stupid thought, it is literally redundant. Someone was always having that court at every moment in time. Those sorts of continuums are useful when we stop trying to do the right thing – truth, reality. This is how reality used to work – it used to be straightforward and you can look at things and they would tell you how reality worked. We used to watch TV, have these very linear forms, books and libraries. And you had to go to them – theatres. People would tell you. Over the past 20-30 years, we have established that reality is a thing. The information that colours reality is less certain, and it depends very much on one's perspective of reality as to what it is that you really see happening. And we see this all over the world. Reality is a perfectly valid, they are the same reality but they are perceived differently. And I have a big thing about truth and perspective. I am not a big fan of the idea of truth, because generally it's value based. For example, do you know the 'London Times'? It thinks it's important. A paper of record. That would mean one thing. That's an interesting truth. If I were to write that, it would be probably taken differently. You would read it differently. But they're the same reality. That is true and the other one is true. That is the title they actually wrote, and the reason I picked it out is that it has been written by someone who thinks the article is about trans relationships. What is a trans relationship? It isn't even a thing. It isn't a homosexual relationship, there is no such thing as a trans relationship. I suppose if you are trans and you go out with someone who is trans, but that isn't what they're talking about. The second part of that is that parents try to stop people, as if parents implicitly know best for their children. But as a trans child who wasn't spoken to by their parents, in fact, who had a government in the UK who literally legislated against speaking to me about how I felt, who literally put for 20 years a statute on the books that said we couldn't talk about that reality or perspective on reality – homosexuality cannot be taught in schools or discussed in culture. And that is probably why I ended up being mentally ill. So as a parent, I am more keen on being trans and not mentally ill than the alternative. That is all anyone sees – that headline. And in case you think it is an English thing, The Australian has recently put together an entire pool of articles on gender. And if you look closely, they're all the same perspective. There is no sense of their being alternative views, of being perspectives outside of the narrative thing. This is what we're doing. It is troubling! As UX designers, you have to deal with that. It's also untrue. I like to talk about it when I talk about data, because it's not how it works, I didn't go from being a white, cis-het male, stubbly person, to a woman, which I am. But it isn't enough data. There isn't enough context. More context here. You see that I joined Google a long time ago, wore a shirt, which is the only
  6. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 6 of 11 time I wore a shirt to Google. And I ate more food. And then I changed my name, which is a big deal. And I had the world's worst security photo. And then finally… There isn't enough data. I like data, I know you like data, so I put in a chart, it has an axis. It is a distribution curve, and I like talking about distribution curves in order to understand the diversity of the human experience, because one of the funny things is that we are increasingly inclined towards normative behaviour. Because we want to scale the products and services we make, therefore you look for the most applicable service, the people who will use it the most, the most normal people. And it's very interesting how most people are happy being normal. We do an awful lot to appear normal. But not so keen on being common. No one wants to be common or typical. But everyone wants to be included. Can't work it out. There are lots of things I don't know the answer to, by the way. Identified those. If you look at this more clearly, without the notion of actual data behind it, you will realise we all have aspects to ourselves which are not typical. If you don't, that would make you very untypical, or atypical, if I'm going to use English. And you can pile those up around the sides. And come to this point of realising that there is nothing that is normal, and actually all you end up with is a huge intersection. And your users hit all of those intersections, they are kind of bizarre. And even if you realise you are super normal and there is nothing you need to think about in terms of inclusion or awareness, even if you are a white, straight guy, at the very least, one day you will be very old and you will be excluded. We all hate it when people want to talk about that in design, but it's really important. This is Denise, Denise hit from me the fact that she is ADHD, until I thought she was completely mental, and then when she explained it, it became completely normal. This is Annabel who worked on my team for a year, being Annabel, who is kooky, and it had nothing to do with the fact that she had aphantasia, which is a very interesting condition we don't create a mental imagery. And it was weird watching her work, looking back, but at the time, you fudge it along. I have structural issues with my brain, and when you start talking about mental health people start getting squeamish, but my health is my health, but it's my health. It is also worth remembering that the brain is amazing. We have amazing brains. And when we get caught up with this AI and technology, normally you are trying to get a machine to do one of the tiny things that your brain does so easily that you don't even think it's doing it. Sensory processing, spatial awareness, cognitive bias, the semantics of language. And that's because of a fantastically complicated array of qualities. I kind of think everybody should get into neuroscience because I'm into neuroscience. Because then you really begin to
  7. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 7 of 11 understand what is happening or what is not happening. You understand the things you didn't know you didn't know like neural networks. My understanding of AI expanded enormously once someone explained how a neural network works in neuroscience, as opposed to an engineer trying to explain it to me with little lines. Then you want to make it even more straightforward. You realise that these experiments are going on all the time. And you realise that they are doing those experiments, training their datasets, doing audio spatial awareness, looking at the object permanence, understanding not only that an object will drop and will drop and drop again, and a balloon will go up if it has helium, and go up and up, and that your parent will run and catch the balloon and catch the balloon until a certain point, there is a breaking point in that, due to behavioural science. But you are also learning about positive attachment and training your ears, understanding when you drop something, what that sounds like. You guys can understand that if I stand here – that was the creaky bit, do you hear that creaky bit? And there is a good creaky bit over here. And everyone in the room understands that that creaky bit is over there and this bit is over here. That's kind of cool. How? So we all know there are tiny hairs our ears and it is to do with our brain being very clever. We have had to train that because everyone's ears are different shape. So we are spending our childhood doing that training. If you really want to understand your users, you have a look at how complex that training is. And it's fun! I think all UX research should involve small children. Because we're all small children when it comes to the UX. We discovered this when we tried to make books. One of the projects I do fairly regularly, or have been over the last five years, is looking at ways you can take the physical infrastructure of a technology like books, taking the physical spine and the printedness on the page and looking at how that inhibits the object and the medium and the content. It is inhibiting if you have to start on page 1 and go in a linear sequence to page 270-blah. But once you get to the internet you are like, it is not really a reason for that. It is 500-year-old text. Even with these little machines like the Kindle, why? It doesn't have to do that. We are not rodents, we can adapt, have plasticity. So we built all these books and played with these ideas, moved the words in and out, that allowed you to move through a narrative that was literally geographic, based on street-view, or it was dialogue-based, lots of interesting problems. We did a book that uses your camera and all the data that you've got in your phone to create a ghost story, which is spooky, and also spooky. It is good for things to be creepy. It teaches us, in a different way, to have that conversation. We made a book which was very much linear, we made a book that was about that stuff I talked about, having multiple perspectives and being able to move in and out between different perspectives on everything from #MeToo to (inaudible). There is an emo section, and there is an AI section at the end which doesn't make any sense but I listed in it as it was really cute. "We
  8. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 8 of 11 tried to get an AI to do part of this book, and it's rubbish. It's the most beautiful part of the book but it doesn't make any sense." We did a book with blockchain which is about ownership and provenance. For me, that is the meta point of the project, but in the book itself it is self-destructive, where you have to remove words. You can't do that with a printed book. Ironically, we made a bookcase, it was people don't understand what you're talking about, we made literally a bookcase with the books in it and they stick out. We gave it to the British library and they really liked it. I really like the British Library because they are dealing with the fact that these are books and they are duty-bound to archive books. But they are entirely temporal and ephemeral in nature. How do you archive a book when it is working off Street View which gets updated every two or three years? Ten years later, the whole story is completely different and impossible. You can't find things because they've been physically removed from the world. So you can't carry on with the story. How is that meant to work? Again, I have no idea. But I like that they were willing to take it on as a challenge. This is a storybook that you place in the kitchen and it is audio-based. And a lot of the work we have been doing is audio-based, about moving around space. Why is it every time I talk about the idea of augmented reality, people think I mean... Augmented reality? That is not augmented reality to me. My kind of augmented reality is a sliding door. When you walk towards it, a computer senses our intent and opens the door and then it closes it again. It has augmented your reality. I like a more literal meaning of augmented reality, it is that a machine can augment your reality. As we were discussing, with the ears, your ears are incredibly sensitive, they pick up and dispense with an enormous amout of information incredibly effectively, especially when you are asleep, which is handy. That is why you have to close your eyes, because your eyes can't even do that. Your ears are still going. Your brainwave is slow, but it is still working outwards, if you hear something going, you will wake up. I am particularly interested, I'm going to show an AR with a screen thing, but also how we move through space. Like, what that becomes, when actually it is the reality of the world that is your user experience, rather than the reality of the screen. Because this screen is the book. It's all to old text. Just a way that we had to get the content to you. And that's not going to be the way we get the content to you. We can see that. Otherwise people would be making better screens. We are not making better screens. We are working on things we used to do before screens – gesture, voice, intent. How we orientate ourselves in space. I work with cultural groups and a lot of it is to do with those things. I play in the spaces. We did a theatre performance, a group called PunchDrunk, with kids, about where they are in space. Giving a computer phone to understand actually where you are and what you're looking at, location and orientation in space, is actually really hard. There were a lot of different ways to do it. You think of the obvious ones, GPS which is naff because of the army,
  9. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 9 of 11 all of the RFID suite of things, different wavelength in different forms. What we discovered is that the best wavelength is the light spectrum. So we ended up using the camera and using the AR tools on the camera to map it, because we're only hacking off the back of other people's things. My team is tiny. We don't do anything to we take other people's tools and play with it. Recently we did a project called Display, working with a dance troupe, where we wanted to place the mental experience of the dancers inside them. So as you moved around the performer, you could hear them as well as see and experience the dancer. We work on this with Liz Jackson, who I think spoke to you yesterday and is a huge hero of mine even though she can come across as a bit spiky. I hope she wasn't too. I will show you a short video, you can kind of see how it works. In this particular performance there is a period called the sculpture garden where everyone stands still. We put people in these rigs, and they are still rigs, because we're not there yet. But you have your little ear pods. These are the spaces you are moving into, when you walk around the space and understand the world around the user experience. (Video plays) SPEAKER: People look at me and make assumptions. SPEAKER: It's not something you can turn off. SPEAKER: My grace, physicality and vulnerability are all on display. SPEAKER: It scares me, because if people knew how angry and tired I was, would they like me? (Music) TEA UGLOW: It is literally the only onboarding (inaudible) I've done, when people talk about what they felt. Engineers talk about their feelings? I don't know how many of you do onboarding and user testing, but getting engineers to talk about their feelings, how they felt finishing the project, it was a magic moment for me, eye-opening. I'm going to bring us back to the little blue line, like that normative curve. One of the problems we are facing – you are facing, I'm too old now – is that this normative curve is applied not just to society, not just to design, it is applied as a principle as to how we train our machines. How we build those neural networks. Just like we as children have spent years and years doing our data testing and validation and training our datasets and our senses, so have AIs, and they are training on the internet because that's all they've got, on historic data.
  10. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 10 of 11 We know this to be a problem, but even when it is not historic data, we have interesting problems, like a program called Quick Draw, have you done it? That's enough issue to pretend that you all have. Basically, it tells you what to draw. This is a hammer, of forces a hammer. And it decides whether it recognises it or not and it is improving depending on what you draw. So it learns not from you drawing it but from millions and millions of people drawing it. It is doing exactly the same thing. Training and testing and validating and learning and getting better and better. I can't even remember what I was drawing. A blackberry. We end up with these huge datasets. The really interesting thing about them is, you get enormous sets of data around what people draw. So the data around a banana is so normal that even the curve looks like a banana. But the data around shoes, much more complex. Because actually, to begin with, (inaudible) but as the computer learns, it discovers that it is abnormal and maybe those were not what people meant when they drew a shoe. Maybe that is a different kind of shoe. So literally, within two months, stilettos or high heels were no longer shoes. We taught them out of existence. And that is what we talk about when we talk about dataset and training and testing and inclusion and erasure. It is not easy, it is intentional, it is not anyone's fault. And we have these huge public access data sets, which I use on a regular basis and slightly anxious about. Because they are an issue in and of them self. If the data doesn't exist, you don't exist, and if you don't exist, you have no way of introducing your data into the dataset. And even when you do introduce your data into the dataset, it has to be so significant that it isn't accurate data anymore because you are a tiny part of the dataset. I can explain at length how important that is, but we wanted to do a test the other day with a watch where you allowed your watch to talk to a phone which talks to a clip. So when you look at my photos database, it looked at my watch and tells me that I know them. So we can do something I don't do, but it's using technologies. And now there is this issue… And then, actually, my friends go, "But then it can predict the name." And I said, "Well, of course, it has to give me the name, or I wonder who it is." "Basically you are giving me a watch that can tell me the name of the person I am standing in front of." "Yes." That is every awkward moment sorted. And actually you follow that thought experiment through and you can see how easily that would happen and how easily the dataset of my friends who happen have voted one way or another in the American election, happen to be Masons, happen to be queer. And then you have a watch which is basically a queer-bashing watch, and you can walk down the street and get a buzzing thing and see someone who is gay and kick the shit out of them. Which isn't very helpful, because there are lots of places in the world where I am not only not welcome, I am not allowed to exist. So it kind of matters, unless you are really normal. And as we have discussed, you are not really normal.
  11. UX Australia 2019 (AUUXAU3008D) Main Room, Day 2 – 30th

    August, 2019 Page 11 of 11 We are working on projects to work on that. You will see lots of things happening. But it is a problem for all of us, because you have to apply it every day, it isn't something you can just think about and get. You can't let it float by and be someone else's problem. I don't really like takeaways. I have been standing up here talking to 40 minutes, if you haven't taken anything away, what have you been doing here? But the people who tell me about presentations say, "No, no, no, takeaways." So here's the first one – if something you meant to happen happens, lucky you, but it's kind of boring. The other thing is, everything is structured, there are things which are absolute, but there is this obsession with things that are subjective. Things that are subjective are the structures of our society and they can't be changed. Thems are the rules, that is the law. It's like, OK, but that kind of means everything is adaptable. So fucking with the structure is kind of my main takeaway. That is how we have done anything interesting. By the way, that's the accessibility logo from Android. I am not quite sure how they do that. Remind everyone you meet that they have the most extraordinary thing in between their ears. Because we forget. We think we have the most extraordinary thing in our pocket, and it's a toy in comparison. And you do you. I can't stress this enough. And I love the source material, because it is a good UX gag. There is no point us building a normative world for users or anyone else, because there isn't a single user who is normal. Every brain is different, every person is different, everyone you meet is different, and whilst that isn't helpful to you, whilst it would be a lot easier if I could just tell you to build it like this, that is what most people are like, it is just not a good idea. So you do you, and please bring it to work. Thanks very much.