Across this specialization we've covered a number of different kinds of designs. We focused a lot on the web and on screens, but I don't mean to offer the impression that that's the only kind of design. In fact, I wanted to close the specialization by looking forward a little bit, and thinking to the coming decades we're going to see an increasing diversity of smart things, and talk just a little bit about some design strategies for addressing that. One way to think about this coming age of ubiquitous computing where we have an increasing number of devices is if we plot the number of devices per person the desktop age is somewhere around one and you've got a PC on your desk maybe a laptop on your desk. As you might add a phone or a tablet or something else this number starts to creep up, and that's kind of where we are today. And what I think is interesting about this is when things become computation powered not only is it good for the semi conductor industry, but it's good for us too that the designers role means there's a chip there, interactivity there, computation there that you can use to modify the user experience. So if we go back to our graph as computation becomes embedded in everything we have a huge opportunity for thinking about the future of design. As we talked about in crowd sourcing design melds to the world of atoms, the world of bits, and the world of culture. In our social computing section we talked about the culture side of this. In a lot of specialization we talked about the bit side of this, and what I wanted to close with is thinking about the integration of those three. The internet of things is really the Internet of smart social, computational, and physical systems that together form new types of collective intelligence. Here's an example I really like of the coming age of ubiquitous computing and what that can mean. This is a design by Gad Shaanan's firm Gadlight. It's called the Yofi-Meter, and it's a Bluetooth connected diabetes meter. What I think is interesting about this is that it unifies all the disciplines of engineering and design. There's the physical aspect, there's a screen, there's material science and fluids all in a tiny package. And it transforms something that used to just tell you one piece of information that was ephemeral to something that's durable, and the fact that this information can be shared with people, aggregated over time, and analyzed brings all sorts of new design opportunities to the fore. To make this concrete I'd like to give three examples that I think are today kind of representative of the frontier of smart things. The first of these are devices that are embedded in the home like the Nest thermostat. The second one are devices that go on the body like a FitBit, and the third is computational augmentations to existing devices. I think of these like smart barnacles like the Amazon Dash system. The interaction design for devices like this looks in many ways very different than how we might design for a graphical user interface. In particular, with a graphical user interface it's a device that's at the forefront of your attention. You're interacting with it for minutes, hours, sometimes many hours. With these devices the interactions are really short and in that way the best user interfaces are the ones that you may not even notice or that just seem completely obvious. In the late 1980s Mark Weiser was a research scientist working at Xerox PARC, and he pointed out that the most profound technologies are often the ones that disappear. They weave themselves into our everyday lives until they're invisible, and since that time it's been an aspiration of the field of computer science and human computer interaction to try and make computers that disappear. That support everyday activities that are invisible. Take the Fitbit for example. A core piece of functionality of many wrist worn devices is that they count the number of steps that you take. There's a huge amount of inference involved there. The machinery has gotten impressively good. The sensors get better every year. Still, there is some amount of inference because these sensors are intrinsically noisy and there is real challenges, and so how do you expose that to the user? Do you express the uncertainty? Do you try to hide it? These are the new design challenges to think about. Often you may be aggregating in the user experience. Multiple difference sensors. For example, there might be an accelerometer that's worn on the wrist and you're using that in conjunction with the GPS sensor that might be in a phone to try and infer something about the user's activity. As Don Norman talked about when we were discussing gestural user interfaces one thing that can be difficult is to know the users intent. Did I bump up against something automatically or was I trying to shake something in particular? Sometimes when I listen to music on my phone there's a gesture where if you shake it it will randomize the songs, and so if I happen to bump the phone all of a sudden it will switch to the next song. It can be difficult to know the user's intent when there's an inference step involved using sensors, and because this new set of devices is just coming online there's an extreme lack of standardization. Some of that is intrinsic. How do you standardize across a button on your washing machine and a device worn on your wrist, and lastly one thing that's important is really the topic of a different specialization, but it's something that designers need to take into account from the get go. Is thinking about user's privacy. This isn't something that you can bolt on at the end because many of the design decisions that you make have huge implications for privacy and security of systems, and you need to engage with those issues at the very beginning of the design process. The good news is that the hardware is getting increasingly easy to find in the program and there's a huge community building up around hardware hacking. What's exciting for me is that many of these people are designers and artists rather than more traditional computer scientists. There's a real sea change afoot that I think is going to hugely effect the user experience of ubiquitous computing, and I have no idea how 3D printing is going to effect this. Will we have a moment where in our future we hit a print button, or are Amazon and other electronic retailers in essence of 3D printer albeit one that has a day or two latency between when you hit the print button and when it shows up at home. I showed an example of a sensor that was affixed to a washing machine. The tech press surrounding these devices and our own experiences about these devices is often inside the washing machine where we're alternating between the hype and the enthusiasm of what's possible with the disappointment of this is just a gadget, or the connection doesn't work. And so as a designer you always have to beware the drawer. That drawer of unused devices that seemed neat, but don't in fact deliver any value. I mean, the designer's role here is not just the embarrassment of being the device in the drawer, but there's a lot of e waste that's generated from devices that the design wasn't thought through. They don't delivered on the value. If we improve the design, then we can better ensure that the things that we produce are things that people actually use. One thing from Mark Weiser's time in the 1980's up through the present that we've seen consistently is that when censors are involved dashboards nearly always follow. Dashboards are the user interface design of showing quantitatively an analysis of your behavior, sleeping, meeting room calendars, stocks, whatever else that you might be interested in is. In this specialization you learn techniques for being able to distill down the user interface to show just the essential aspects, and this is particularly true with dashboard design that a well designed dashboard can turn a useless gadget into something that provides valuable information. We're at a really exciting time in design. Design is front page news. It's the topic of Hollywood films, and as you well know yourself enrollment in design classes both in person and online has skyrocketed. I opened this video by talking about the convergence of atoms and bits in culture, and one example of this is the Nest thermostat. In addition to being a beacon in many ways of where design is headed I also think it has metaphorical significance for design. We could take all the enthusiasm for design right now and just take the temperature of the current situation, but I think that would be a mistake. There's a piece of advice from a philosophy professor that really stuck in my head. When America's current president was inaugurated this philosophy professor Cornell West gave him the advice don't be a thermometer that takes the temperature. Be a thermostat that sets the temperature. I think that's our opportunity in design. Designers have an incredible potential to be the thermostat. To be able to set the temperature that makes the world a better place, and so I'm encouraging all of you to take the skills that you've learned in these classes, and as you move into the capstone think about what you can do in the design world to set the temperature. I'm really excited to see what you come up with.
