 So, I know you probably came here to find out if I'm normal or maybe if you're normal or maybe learn about some of the technologies that might help you figure this out, but we're not going to do that, at least not now. First, I'm going to give a different talk. I'm going to give a mini meta inaugural lecture and I'll tell you why that is now. The real issue was I wanted to tell all my friends at home when the dean called me. The executive dean, Josie Frazier, called me and said, you know, congratulations. We're giving you a personal chair. And I was so excited. I called people up and said, look, the dean just called. I've been given a personal chair. And they said, yeah, so it's, you know, do people at the OU have standing desks or something? They didn't get it. I said, no, no, no. I've been made a professor. And again, they said, so what? I mean, you've been at a university 25 years. Everyone who teaches a university is a professor, aren't they? Because in North America, you just call everyone who shows up in front of a lecture hall professor. There's a bit of difference in terminology. So that didn't go down very well. So then I went on and said, well, look, in the UK, we do it differently. You don't just get the pre-nominal professor. You get to have a title and you get to choose it yourself pretty much. So I went, started going through possibilities for what my title could be. And as you've heard, I'm doing work in people measuring things about themselves. The field is called the quantified self. It's been sort of a bit of a cult for a number of years and now becoming mainstream. So we thought we might do professor of the quantified self. See how that went down. However, that was a bit off the wall. So we thought we might use one of the more conventional terms, life logging. So it could be professor of life logging. But in my user testing, that didn't go down so well either. So the more conventional one was professor of wearable computing because obviously you're wearing lots of the stuff. Or maybe professor of digital health. And then we started just adding words on. It started going a bit mad. And then it got too long because it wouldn't fit on a business card. And it began to just get a bit out of hand. Finally, because of the cyber security aspects of all this, we thought this one might go down well. And it actually occurred to me, Professor Heddington, I know we're doing the cyber security lectureship, professorship recruiting. We could actually put this as a possible title for that to encourage the right candidates. Then we were left with the title that all my colleagues call me because of the research I've been doing. And this is what the sign on my desk actually says right now. But in the end I went for the Sage Council of a long-term mentor, Professor Marion Petrie, who was also my best man at my wedding. And if you want, if she makes it here in time, she's at a funding panel right now, but she makes it here, then you can ask her about the cross-dressing. Oh, you're there. Hi, Marion. Yes, you can ask her about the cross-dressing theme at our wedding. Anyway, so we went with the more conventional professor of computing. So thank you, Marion, for making me see sense. So after I explained this to my friends, I said, well, at least you can come to my inaugural lecture. And they said, well, what do you mean? I said, you know, this is an inaugural lecture. And I said, haven't you been giving lectures for a quarter century? And I said, no, we don't do that here. So I then started realizing I had to explain to them what an inaugural lecture was. And I ended up doing a bit of research on this because when I came to the UK, I didn't know what an inaugural lecture was. So I'm going to give you a mini-meta inaugural lecture. So it seems to be a uniquely British thing, or at least common wealth-ish, some of the common wealth countries still adopt it. Canada mostly doesn't because they follow the American model. And it varies from place to place, but it seems to be a kind of indulgence given to newly appointed chairs. And they can reflect on some of the key aspects in their career so far. It can be used to thank the people who've helped them along the way. And I'm going to do that. And the orange text is to remind me to say thank you. So if you see someone's name in orange text, that's me saying thank you. And I guess the final reason is might be the reason you came here today, which is for me to share some of the exciting current research we're doing that might actually be relevant to you. So back on the indulgence part, I'm going to do the bit of reflection on why I became a computer scientist or why I think I did. I look back on what influenced me in my life. And I guess the first thing was I had a lot of support from family, as most people do in this area. But there were two people, I think, who influenced me to become a computer scientist. Computer science requires you to do both engineering and have a facility with the natural sciences. So it was a nice combination for me. I had an interest in engineering probably from my father, who was a private pilot. And he was always talking about flying and airplanes and the technology behind them. And so that really inspired my interest in engineering. And the other influence on me, I think, was his brother, my uncle, who is a geologist. He's still a practicing geologist at the age of 84. And he was always talking to the family about the natural sciences and inspiring us about what the latest thing he learned or the latest thing he'd seen that we thought we should know about. So the reason I'm telling you this, aside from telling you the background, is that for those of you who have influence over children, as parents and uncles, grandparents, I want to encourage you to basically inspire them whenever you can. Not because I think it's a good idea. I think if you do it out of benign self-interest, because we're an aging population. The NHS doesn't have the resources to support us. If you want to have a happy and restful old age, you're going to need to have a lot more scientists and engineers helping out. So please do encourage your children and nieces and nephews and grandchildren to study the STEM subjects and also to think about the Open University as well. So how did I end up in the UK? Well, I didn't know what Mark was going to say, a quick tour of that because he's talked about a bit of this already. So in 1991, I was doing my PhD at the University of Toronto in an area called software visualization. At that time, it was a niche area. Not many people working in the area. Mark was desperate to find a postdoc to work in the area for a grant and he couldn't find one. So he came to my then supervisor, Professor Ron Becker, saying, have you got a postdoc? I'm not going to complain. He's first year in his PhD, but he'll be great. You can have him for a year. Then you've got to send him back. I'll just mention, Ron Becker will be watching, I think, on the live stream from Canada. Ron is also still a practicing scientist at the age of 75 and he's got a new book coming up next year which is going to explain to us why machine learning is going to make a huge difference for everyone. So I encourage you to go out and look for that next year. But Ron came to me and said, I have this job for you at a place called the Open University. And in common with, I think, most other people in North America who weren't in education, I had never heard of it. And he said, it's in a city called Milton Keynes. And again, in common with, I think, most people in the rest of the planet outside the UK, I'd never heard of it. What I did here is you can have a paid holiday in England while working on your PhD. So off I went for a one-year job, as Mark said. And in the early days at the OU, as Mark said, a human cognition research lab. And this was an artificial intelligence lab based in the psychology department at the time. And I was there with Mark and two other people in the audience here. We've got John DeMang. John, where are you? There's John. And Enrico Mata. Didn't Enrico make it? No, he didn't. Oh, he's not well. Well anyway, this is what Mark looked like back then. And you can see the state-of-the-art Macintosh computer. There's John. He also had a bit more hair, as did Enrico. And some of us had a lot more hair then. So actually when I arrived, I also had a goatee beard, which was equally unfashionable. But thanks to the fact that it was pre-digital technology, I've managed to destroy all evidence that's ever existed. So nowadays, no one can do that. Anyway, so when we got there, we mostly did work on software visualization for the first year. And John and I co-edited this book. It says copy's still available. It's out of print. But it's still a widely cited book in the area. So that was all very exciting. I lobbied for the OU to be connected to the internet because when I arrived here, I was quite surprised. The OU, in common with most UK universities, was not connected to the internet. There was a tiny academic network that was unconnected to the rest of the world. And as Mark said, I built the first web server. And after about three years of a one-year contract, it was definitely time to go back to Canada and get back to my PhD where Ron Becker was patiently waiting for me. However, there was a job advert for a temporary lecture in computing. And my colleagues, Mark and John and Noriko, all said, well, it's a well-known thing that internal applicants are interviewed as a courtesy, even though you don't have a chance of getting this job. Just apply because the interview experience will be great. And then when you have the real interview later on when you finish your PhD, that'll be really useful for you. So I went to the interview. And to everyone's surprise, including mine, I got the job. But it was only a three-year temporary contract. So I thought, I can do another three years in the UK. I've learned how it works now. It'll be much easier. So in the first year, I went to my dean, Pete Thomas, and I said, this internet thing is really important. We're really going to have to look at changing practices in the OU because we had, at that time, every assignment came in by post. So the tutors would receive them by post. They would post them to the OU. They'd come in at 50,000 a week to a building just next to this one where clerks would open envelopes, key in data, put them in other envelopes, post them back out. And it was a huge industry. I believe the OU was one of the Royal Mail's biggest customers at the time. So with this money, we formed a small team. And some of these people are also here today. Marion, of course. Linda Price, Barbara Poniatowska, Mike Richards, still at the OU, and Pete Thomas, my dean. So we spent about two or three years and basically in that time scaled up from an initial pilot to showing large-scale internet teaching, doing lectures. We did assignment handling. We did automatic assessment handling. And this was well before it was a mainstream OU activity. Somehow during that period, they decided to make me a permanent lecturer, which was the closest you could get to tenure for the benefit of North Americans. Margaret Thatcher abolished tenure, but this is the closest we get to it as a permanent contract. So we did a lot of publications of all of that stuff, the internet teaching stuff. It's still one of the most cited material on that subject. But a number of things happened after we finished that in 1997. The first one is I got married to Linda, who had come to the OU to work on this project. Then I got seconded to KMI, as Mark mentioned. And then I returned back to computing and we had a new research director, Professor Bashar Nusebe. And he introduced me to a new research area of privacy. And together we supervised my first PhD student, Karima Dam. And at the same time I met another long-standing colleague, Arosha Bandara. And the three of us collaborated on our first major grant together called PRIMA, Privacy Rights Management for Mobile Applications. And this was all about using the newly popularized mobile phone, smart phone, not quite smart at that time, to control your personal data as you were roaming about. And Facebook was just becoming popular and only available on university campuses at this time. So we did a lot of work on privacy and then began introducing bits of what I call life-logging to the work. So we were looking at how people recorded data about their lives. We added... Oh, so we had two other PhD students at the time, also doing important work. Łukasz Jędrzejczyk did a lot of work on mobile privacy with controlling your location on your phone, who got to see it and who didn't, and developed a new interface called Privacy Shake, which isn't a dance. It's a thing you can do with your phone to turn the privacy on and off. And Kierty Thomas, also here tonight, please say hello to him afterward. He did important work on privacy requirements. During that time, as Mark said, he was a digital forensics module. Another student came along, Ian Kennedy, who's a part-time student and just graduated, and did important work on malware forensics. So the work in life-logging extended a bit beyond the person at that point. As Mark said, I had a hobby of energy monitoring and using renewable energy, monitoring solar power and electric vehicles. And that led us to another PhD student, Jackie Bourgeois, who also completed relatively recently, and also working on the MK Smart project with Enrico Mata, which some of you may know about. But the real change happened when a BBC Horizon producer came to me and said, I'd like to do a film of some of your next study, whatever it is, on digital health. And so we were just about to start a study anyway and allowed them to come and do filming of various segments. And that basically was the start of people learning about my interest in digital health and the work taking off. And that leads us sort of neatly probably to the lecture you came to see, which is, am I normal? So we'll begin with a bit of audience participation. For those of you, there are probably enough people in the audience who've known me at least five years. So if you've known me at least five years or you're at home on the internet and you can use the hashtag to reply to this, what is your answer to the question, am I normal? I see head shaking. So for those of you who don't know me well, I think I'll give you a bit more information and then you can judge and we'll see how it goes from there. So as you heard, I'm married. I actually have four children. The youngest is just about growing up and not quite left home. I'm also a grandfather recently, so I have one grandchild. And I know what you're thinking, you're far too young to be a grandfather, right? And you'd be right because I'm only 53. So that makes me middle-aged. Let's see, what else can I tell you? So I'm about 5'10". That's the 177 centimeters. I weigh about 13.5 stone. That's 86 kilos or 180 pounds if you're in North America. And unfortunately that gives me a body mass index of 27, which means that I am overweight. So that puts me there. So according to this, for some reason my pulse is 110. But my O2 sats are 98, so that's okay. So obviously I'm a bit nervous, but anyway let's ignore that bit. So that's fine. What else can I tell you? Let's see, let's try this. So this is my blood sugar, 3.9 and going low. Okay, so that's not good. Hopefully there's lots of food after this. What else? So I can tell you what I've been doing. So my phone helpfully records things like that. So I'll show you what it showed me from yesterday. It says I've got 5,000 steps. I did a run of 4.6 kilometers. I was in the car commuting for one hour and 27 minutes. And I was at work for 4 hours and 11 minutes. Okay, Vice Chancellor, just let me explain. So the app just records when I'm in the Jenny Lee building. I'm actually working lots of other places like the hospital, I was at the hospital too. So most professors do work 18 hour days, I promise you. So what else can I tell you? Well, so the passive sensors on my bed can tell you how long I slept. You won't be able to read that probably. But it says 7 hours and 41 minutes of sleep, waking once, 31 minutes of snoring. Sorry, Linda. Bedroom average temperature of... Even I can't read that. 19 degrees, humidity 51%. Resting heart rate about 52 while sleeping most of the night. So that's not bad. What else is there? So it says here that my car drove 47 miles. It used 8 kilowatt hours of electricity. The home used 15 kilowatt hours of electricity. Most of that came from the solar panels or the stored batteries. So it was mostly off grid or low carbon. And that means that the entire day was very green. I had an average carbon footprint of less than 150 grams of carbon dioxide per kilowatt hour, which was very low. Now that you've got all the data, once it falls in place, do you think I'm normal? No, everyone's shaking their head. So the question, am I normal, could be asking you two things, of course. Am I crazy? I mean, I'm collecting all this data. Someone who does that must be crazy. Or I could be saying, is the data normal? I mean, my blood sugar is low, my pulse is high. You know, should I be worried? And you're right. Spending time collecting the data would be crazy, except I didn't have to go through any effort at all, or mostly no effort to get it. It all came free, or mostly free, from apps on my phone or very cheap hardware. I didn't have to do very much. The smartphone that you carry has so many sensors in it. Gyroscopes, accelerometers, compasses, light meters, GPS. And the cost to you is nearly zero. So the question is, if I'm not crazy for collecting it, and it's easy and cheap to get the data, then am I normal? Well, we don't know, because all we have are these one-off measurements. And I'm sure lots of you have probably worn an activity tracker at some point in your life, or measured something, or even just measured your weight and written it down. And the problem is that single measurements, or short-term measurements, are almost useless. You get no trend. You don't know whether it's going up or down. You don't know whether it's a sudden change. Or like me, I'm probably nervous. That's why my pulse is so high. So you can't really tell. So you need to have something that lets you keep measuring and measuring and measuring. And ideally, there's no effort for you to measure, because you want the data to be there when you need it, and you don't know when you're going to need it. So I'll tell you a little story of when this was actually useful for me personally. And this is actual real data. I haven't made this one up. So I got probably one of the first Fitbits in the UK. I should mention other activity trackers are available, probably much cheaper. And if you want to ask my opinion, I'll tell you after the talk. But Fitbit were the first to come out, and I managed to get one before they came to the UK. And so I've been tracking my activity level for seven years. And I always was, at least until the early part of this decade, overweight. Not quite obese, but just generally a bit fat. But I was very active. I would be running all the time. I thought there should be no reason for this. I was very, very active. But what I did was I graphed the data and discovered that essentially I was allergic to UK winters. I thought I did running. But I would essentially, in the winter, when it's cold and wet, I would wimp out. I'd stay home or I'd maybe go for shorter runs or whatever. But the data was very clear. Those dips in the lines you see are the winter and they go up in the summer. And the weight would fluctuate completely out of phase. So I would lose weight in the winter and gain weight in the summer and gain weight in the winter. And once I figured that out, it was quite simple. I just bought the right clothing and I could run year round. And then I basically had a steady activity level and a nice low weight. So except for the last week or two when I'm being lots of cakes and that's why I think I'm overweight now. But anyway, so what is normal for you? And that's probably why you came to this talk. The idea that I'm trying to get across is that automatic and long-term collection lets you find out what your normal is. The so-called 10,000 step rule is just a convenient number. Your normal is whatever you can measure and find out is normal for you. If you want to then get more active, once you know your normal, you can increase and you'll know you've increased your activity. So some of the interesting things you can do with this technology in addition to the long-term measurement is you can measure your athletic progress if you happen to be a runner or in some other activity. You actually get the data and you can see how you're improving. If you're not improving, you can make changes. Other things that help some people is gamification. So some people broadcast what they do to social media so their friends can encourage them or compare yourself with others. And many of the apps, the one that I'm showing here is Strava, lets you compare yourself with completely unknown people so you can run a route and completely strangers who've run the same route you can compare yourself with them, people of the same age and gender and ability and then you can do fair comparisons. But the point is everyone is motivated by different things. You need to find what motivates you. So I know you're probably thinking that a lot of these things don't apply to you because you're not an athlete, you're active enough, you're probably not too fat, whatever, it may just be not for you. But let me tell you about some of the ways we're using this technology now to make important changes in healthcare. One of the ones that we're just about to finish is one involving joint replacement surgery. So as you may know, as you age, your joints begin to wear out and they become quite painful and so much so that you don't really want to move very much because every movement is painful and that leads to more health problems and so on. So we've been able to fix this for about 50 years now with joint replacement surgery. So knees and hips are the most common ones. There are about 160,000 done in the UK alone last year, around 400 hospitals doing them, very routine. It's quite major surgery, as you might imagine, but most people feel better afterwards and they get the movement back and they get their life back. However, about one in five of people who have knee replacement surgery are still in the same amount of pain a year later. So for one in five people, it's not very effective. Hip surgery, it's a bit lower, but it's still not 100%. So what we want to do is find out what makes the difference in people recovering from this kind of surgery because we actually don't know. If you went and had joint replacement surgery now, your surgeon, depending on who he or she is, will tell you something completely different from the next surgeon. There's no evidence and that's what we're trying to figure out. So we've been running a two-year project with Milton Keynes Hospital and we've been working with Ollie Pierce, orthopedic surgeon, who is here tonight. Somewhere, where is he? Yes, on the back, okay. So we've been doing a two-year project. We've been monitoring people who've had total knee replacement surgery. Our 35th patient is going to have her surgery on Tuesday, the last one, and what we're doing is giving them essentially a fit bit, promote two weeks before they have their surgery, along with a box they plug into the reader at home, and we just monitor the activity level before surgery and six weeks after. And in addition to this, they monitor their pain, they measure how much movement they have during that period, and what we're doing is currently analyzing that data so we can give personalized advice to people after they've had their surgery and tell them how active they should be or how inactive they should be. Some people might be overdoing it, but the point is with very inexpensive technology to give people surgical practice and give people real evidence for how they should behave after surgery. The other thing that's important for any kind of surgery and recovery from it is pain. And everyone's pain is individual. There is no normal pain. On a scale of 0 to 10, my three might be your seven. So it's important to understand how much pain someone is in and give them the right kind of pain relief so that they can move again, and if you're in too much pain, you're not going to be moving. It's possible by having a nurse come around every couple of hours and ask you how much pain you're in and then if it seems to be going up, they'll give you more pain relief, but nurses are very busy and often that information comes too late. So we designed a device we call the pain pad. Note how cheap it looks. It's actually a plastic box with a number pad on it. And all it does is it beeps at the patient every couple of hours, whatever frequency we set, and just ask them to push a button to see how much pain they're in. This is transmitted wirelessly to our secure server. And we can use this in two ways. One is for research purposes to see, look at their pain relief history, see what the pain was, see what was working and what wasn't. But the more potential use of it that we're investigating next is to send live pain graphs to the nurses station. So as a person's pain is starting to go up and up and up, the nurses can spot this in advance and give the pain relief before it becomes so bad for the hospital next day. So we're hoping to save lots of money for the NHS by getting people out of hospital sooner. And the next step on that project, by the way, is a pill logger. We still haven't got the prototype of that yet, but the idea is as you take your pain relief pill, it sends a signal and logs when you took your pain relief and we can combine the pain level and the pain relief sending and get people really, really coordinated with their pain relief. The next thing I want to tell you about is this project in the last four months now called Stretch. It's an acronym and we had to make up an acronym for most projects you have to make up an acronym to get funded. So this stands for Sociotechnical Resilience for Enhancing Targeted Community Health Care. And so in English, that means that we found that older adults tend to spend more time in hospital than they need to. You may get hospitalized for something simple like a urinary tract infection. And then they cure you of that. But then they're kind of unhappy sending you home because you're well enough to go home, but perhaps the care at home, you don't need full-time nursing care, but you need some level of monitoring a bit beyond what a normal home would have. So this project is about giving people some smart home technology, some wearables to give them the confidence to go home and recover in their own home, and also connect with what we call the circle of support. And this is the ordinary people, not necessarily clinical people around you who support you. They could be your family, your adult children, your grandchildren. It could be neighbors, community health workers, your GP. There's a whole network of people who provide varying levels of support. And what we're looking at is giving them so that they can intervene. If, say, someone goes on holiday and the person needs a visit from someone, we can detect they haven't visited in some length of time. Or perhaps they've stopped drinking. We detect they're drinking less liquid, which can lead to a urinary tract infection and send them back to hospital. So by having the right kind of monitoring technology in the home, we can intervene early and get the right kind of data to the right person, whether to the medical staff or to the neighbor, so that the person is well supported and can carry on living there. I should also mention that I mentioned the importance of the STEM subjects and we have a recruiting problem in the STEM subjects in the UK. But that is not the end of everything because all of this work requires the social sciences as well. We spend a lot of time and a lot of our projects, as I'm about to tell you about, working with social psychologists. And they help us get the understanding of the people and the technology right because it's never always about the technology. So, what are some other things we can do? Well, as you may know, you saw me check my blood sugar earlier, which was low. I'll just check it again and see if we're going to do better. Yep, it's up, so we'll be okay. So I don't really need to check my blood sugar. I have a perfectly working pancreas. It produces enough insulin to regulate my blood sugar. But as you may know, diabetes is becoming a serious problem in many parts of the world, including the UK. In some parts of the Middle East, almost half the population have type 2 diabetes. So it's a growing burden on many healthcare systems. Not just on the healthcare system, but on the individual. Because in order to control your blood sugar level if you're diabetic, you need to do a lot of dynamic calculations all the time. So, as you probably know, when you have carbohydrates like sugar, your blood sugar level goes up. When you take insulin, your blood sugar goes down. If you exercise, your blood sugar goes down. If you have a stressful experience, your blood sugar goes up. And diabetics need to make decisions about all of these things as they're going through their day. Should I have a bit more of this bread because I'm going to exercise later? Should I take an extra shot of insulin? Or will that be counteracted by the fact that my boss is going to yell at me in an hour and then I'll be stressed? So there's lots of complex decision making going on. And we have a brilliant PhD student, PhD Katz, who's been building models and quantified self-systems using these technologies I've been talking about to help give diabetics proactive advice on what they should do based on their current context. Not what they should have done a few hours ago, which is the kind of information they usually get from the systems they have now. If you manage to look at some of the demos outside in the hall before coming in, you might have met Theo Giorgio, one of our PhD students just completing. He's doing work in gait rehabilitation. So if you have a stroke or some kind of brain injury, or even just have surgery on your lower limbs, you're going to walk funny. And learning how to walk properly again, it's almost like learning to walk again as a child. You really have to be trained again on how to walk properly. And what Theo has been doing is using the kind of sensors you find in your smart phone to measure people's gait asymmetry, how asymmetric their gait is, and then by haptically queuing them. That's basically tapping them on the ankle or on the wrist at a regular interval. At the pace they should be walking to, they can be entrained to walk properly again. So the data looks like this. This is actually my data, which is kind of boring because it's completely symmetric. I mean, it is actually slightly asymmetric. I'm not perfect. But if I had a stroke, I would have seen a big separation between the blue and the red lines, which are each leg moving up and down. And so the work that Theo has been doing is having tremendous results on people who have basically learned to walk properly again. Now, most of the work I've talked about has been about health-related life-logging. But I'll just do another audience participation thing again. I'll put these on. And so does anyone in the audience know what they are? Yeah, so the ones with the hands up are under 30. Yeah, you don't count. So these are Snapchat spectacles for those of you over 30. They are essentially a camera, which is what the little light was going around. You've now been videoed and uploaded to my Snapchat account. But no one looks at it anyway, so don't worry. So you may think that's a bit weird, especially if you're a bit older. But you may also have seen GoPro cameras, which are things people wear when they do extreme sports. They'll put a camera on their head and then film themselves going about. So what we found is that filming, or even just still photography in public places and posting publicly has just become more and more common. It's something that is part of the Facebook generation, the Snapchat generation, the Instagram generation. And so we decided in one of our research projects called Privacy Dynamics, to start looking at this and to see how people control their privacy when everyone is taking pictures all the time. So we studied groups of people who were sharing visual life-logging images. We gave undergraduate students a wearable camera. We intentionally made it ugly and bright, so it would draw attention to the people wearing it. And so that people around them would ask them questions about it. But strangely, people didn't ask questions about it, because at least in Britain, people are very shy about doing that. They assumed it was a medical device and they didn't want to embarrass them. Little did they know their photo was being taken. So anyway, we examined what people did with these photos after wearing them for a week when a group of people all wore them, a group of people all living together wore them. And the interesting thing was these were all undergraduate students. They were very protective of the privacy of strangers. So if strangers appeared in their images, they would not share them or not share them anyway. They would not let anyone see them. And then we found different social norms started developing. In the groups of life-loggers, there was a social norm that developed that it was okay to wear your camera when you're around the other people with cameras. But when you're with the normal people, you were much more protective. You might switch it off or hide it or whatever. So it was quite an interesting behavior we found. So I've told you about all these wonderful things and all this data being collected and the obvious question to quote Jeremy Clarkson, what could possibly go wrong? Well, one of the major problems that we have is a problem that the older people and the audience might recognize as the beta and VHS, the beta and the beta VHS problem which was a form out of videotape for those of you who are under 40 which you had to basically choose one provider or another. And such is the case with many of the devices that you buy today. If you buy a Fitbit, your data is locked into Fitbit's ecosystem and it's quite difficult to get it out. We call this a data silo. The other thing that can go wrong of course is the company that gave you a nice cheap wearable and the free access to the web page and everything else. They can go bust and you could lose your data or more likely they will sell the data in their bankruptcy sale to someone else and you have no say in that. The contract you signed by implicitly using it says they can do whatever, the company that buys it can do whatever they want with it. So it can be sold on without your permission. It can be hard to switch providers and it could also be sold if the company is hacked as we've seen many high profile hacks lately then it could be sold on the open market. So this is what we call a problem we call inverse privacy and if you have a Tesco club card or any loyalty card you also have this problem because the company you have the loyalty system with knows more about you than you know about yourself and such is the case with all of these wearables. They're only giving you the data that they think is useful to you and the other data they're keeping secret from you even though it's yours. So we've got a project that's been looking at ways of fixing this. It's called Monetize Me. It was joint work with professors Kirstie Ball and Maureen Meadows formerly from the OU Business School they've now gone on and we've developed a couple of things that can help people get control of their own data. One of our brilliant research software engineers Austin Owens-Dake here in McCormick has built a system called PACRAT and what it does is it connects to these providers of data it sucks the data out of them and puts them in your own personal silo and that is the advantage of one you get to keep your own data but two, you could also run analytics on it or someone could run analytics for you and find interesting connections between your data if that's what you want to do you could even sell your own data to someone if you want to because it's yours. We're looking also in this project about the business value proposition of privacy and how companies can be encouraged to basically make keeping your data private So what else can go wrong? Well privacy is a very nuanced thing you know many people probably under the age of 30 quite happy to share absolutely all their data on social media some other people a bit more cagey and some people just want to decide on a case by case basis so one of our other brilliant research engineers Vikram Mehta who was also doing demos earlier has built something called a privacy band now this is just a prototype here so the big box on the side is a very, very important but the idea is that instead of having disruptive graphics that come up on your screen or beep at you or interrupt you you get a subtle itching sensation on your forearm from the privacy band and depending on the pattern of vibration it tells you what you know what is being possibly accessed and you can make a decision to have that data released block it or prevent privacy breaches just by a subtle scratching action and the other interesting bit of news is we won an award for this two weeks ago at the IET Innovation Awards for Cyber Security so that's the photo of us with our award and the certificate that's Bashar, Arosha, Vikram and myself and there are patents pending I think in the US and the UK at the moment so wearability of devices it isn't just for people because humans are easy to evaluate if someone doesn't like a wearable device they'll tell us, we can interview them or they can vote with their wallets by just not buying it but what about animals? wearable technology for animals is not new we've been doing it for about 50 years we've been using it biotelemetry in the analysis of wildlife but we can't find out from the animals whether it's affecting them or not and why is that important? well, you probably want your cat to feel not to be in pain but wearability doesn't just mean that your animal is uncomfortable if you're studying wildlife it can mean bad data so we need a way of getting information from the animal about how the wearable device is affecting them so another of our brilliant PhD students Patrice Apache also in the onstage, she can talk to you afterwards about her work is doing work at measuring animal wearability and what this is leading to is informing the design of people who are building wearables not just for cats and dogs but for livestock and for people doing animal wearables to track wildlife because it's very important they get the right kind of data so comes down to the question do I think you should life log record this data about yourself well it's me so I'm going to say yes but only really if it's low effort I think the real key is that it has to be low cost, not just in money terms but in your time low cost in terms of privacy so you can do lots of things to control your privacy, set pseudonyms don't use your real name, that kind of thing and still have access to your data and you want something you can set and forget because you need to have the data when you need to have the data you need to have collected it years in advance in many cases so if you want a history of data if you want to be able to detect trends then you need to have been collecting it beforehand and if your doctor asks you has been this way, you can look at your data and find out or you can set algorithms that automatically watch your data and look for a trend and if it sees you gaining weight for example, you can get a warning of this because you don't become obese overnight it happens slowly over time and we're very bad as humans at recognizing very slow changes in ourselves so the other thing I would counsel is make sure you don't get locked in don't go to a system that doesn't let you get locked in some way and have some control over it because otherwise you'll be throwing it away I imagine probably some number of you have had a wearable device a fitness tracker, a Fitbit or other brand can I just see a show of hands if you've had one ever in your life and warning so that's at least three quarters of the audience now put your hand up again if you're still wearing it after eight weeks that's about a third so we have a mean time to back of drawer calculation most people have this it's about eight to twelve weeks actually and they forget about it and they stop collecting the data so if you're going to do it, find a system that works for you obviously be careful about showing your data less to be used against you there have been a number of cases in the US of people making insurance claims and then finding that their fitness tracker betrayed them because they claimed to be injured and the fitness tracker was saying oh yes they were active and they shouldn't so they were arrested for fraud so another thing, don't become a slave to your data I've done studies of people wearing fitness trackers and I found three types of people in my studies those who give the tracker back to me at the end and say thank you those who ask to keep the tracker because they want to wear it not every day just every once in a while to just check themselves keep themselves honest and use to give the tracker back no matter what buy replacement trackers and check their steps every day I think Celia one of my participants from my first horizon study her husband was in hospital in emergency and she hadn't reached her 10,000 steps at five minutes to midnight and had to run around his bed six times to get up to 10,000 so don't become a slave to your data and the final thing that I'll leave you with the final thought is if your service is free then you are the product that's being sold so just consider the cost of everything you buy I'll say my final thank yous then obviously my family for putting up with me wearing every gadget imaginable for the last 15 years the nickname at home was Inspector Gadget all my academic colleagues it is a bit of a cliche that we just let all the post docs and research engineers do all the hard work and then we take credit for it but it is at least partially true I'd like to especially thank my colleagues in the software engineering and design group particularly Professor Bashar Nusebe and newly announced Professor Arosha Banderra if you see him afterwards congratulate him he got his chair two weeks ago both of these people have been instrumental in all of my work and I wouldn't be here without them I'd like to thank the support staff in the School of Computing Communications Leslie, Danielle, Debbie and Sarah and the OU-COMS team who have really helped put this together and thank Lucy for allowing me to do my inaugural and thank you for indulging me today thank you very much