ࡱ> uwtY bjbjWW ==ЎB]""""8Lj^^(/111111$ UU"";^"R/""""/W:,t"/B@*.e\% Internal Working Draft / Thought-Piece Script for Digital Living Room Conference Keynote 17 June 1998 Slide 1: Cover Page Slide 2: What is a Living Room? I first want to express my appreciation for having the opportunity today, to address this distinguished audience, and kick off what promises to be a fascinating conference on the future of the digital living room. A fitting way to start off this conference is to ask, quite simply, What is the living room? What makes it distinctive, and how should we conceive of the living room as different, special, as having unique needs that arent adequately addressed by technology today? Compare the living room to the office. In the office, were active, in lean-forward mode leaning forward at our desks, at meetings, at our computers. Whereas in the living room, were in lean-back mode, lying on a couch, sitting in a comfortable chair, relaxing with family and friends. Its characteristic of the office that we concentrate, and focus on productivity. Whereas in the living room, we let our guard down. The living room is the place where we socialize and relax with others, the place where we live. And theres a difference between the way we value technologies at the office vs. in the living room. In the office, the value of a technology to us is the functionality provided, period. But in the living room, the attention required by a technology, the degree it makes demands of us, diminishes its value to us. The functionality of a technology is important in the living room, as in the office, but it becomes less and less appealing, the more it forces us into lean-forward mode and interrupts our life. Slide 3: Imagining the Destination, and Steps Today With all the technological changes over the past decade, particularly in personal computing and the introduction of the internet, it might seem as though were already well into the next technological revolution. Theres no doubt that these technologies are very important: particularly the interconnectivity of the internet has the potential to substantially change our lives. But a point I hope to illustrate and drive home today is that these changes only scratch the surface, in terms of what can and will be done with technology. The so-called computer revolution only touches us occasionally, in focused ways; when we turn on a PC or use a credit card, and we have to contend with awkward, step-by-step techniques to interact with these very non-human machines. The things that do touch our lives much more frequently and deeply are the everyday objects in our homes. The mundane objects we use every day light switches, pieces of paper, clothing, bars of soap these things are the principal things we use to fill our everyday needs and satisfy our everyday desires. And herein lies the true opportunity of technology. Only when theres a seamless integration of technology with life, only when technology is more than a curiosity but an everyday and unsurprising way of satisfying our everyday needs and desires only then will we have seen the beginnings of the true technological revolution. Today Ill address two topics: First, some thoughts about the future. Theres no crystal ball to know what will be realized in the digital living room over the next decades, or what strategies ought to be pursued by companies to take advantage of that transformation. But Ill offer some thoughts about how we can reconceive of technology moving away from a computing-, and even a device- model of technology and consumer electronics. . . . . how instead, we can look towards a future where were no longer issuing instructions to machines in order to have them fill our needs, but rather where our needs are fulfilled without having to issue those instructions. We can imagine a future in which technologically-enabled can do this because theyll understand us more the way a human does, and understand things about us as individuals as well. Also, Ill talk about today. Only in light of the changing future can we discern the patterns in whats happening today, and distinguish whats important from whats incidental. Without showing specific and concrete examples of how were realizing this future vision, its nothing but empty words. So the second thing Ill do today is give you some examples of products, technologies, and ideas many taken from the work were doing at Philips, and some from elsewhere in which this future is being realized. Of course were early on the curve: thats where our opportunity lies. Its impossible to predict exactly what will happen on this curve in realizing the future. But the examples Ill show will give at least a glimpse of whats possible, and where we are now. Slide 4: Envisioning the destination: The unmediated fulfillment of needs The wrong way to figure out where the digital environment is headed, is to begin with todays technology and extrapolate from there. Its hard for technologists to avoid that, but theres no better guarantee that predictions will be wrong, whether the extrapolation is a straight line, or a hockey-stick projection. I propose we step back and start with humans; that we start with the nature of need-fulfillment. How is it that we satisfy the needs we have? Consider three different needs we have, and which we satisfy in different ways. Here are some needs I regularly have: Several times a day, I need to write letters, either emails or handwritten, or whatnot. I need to walk from place to place, around my home, to my office, etc. I need blood to circulate to my limbs. If that doesnt happen, Im in trouble. Now what I want to focus on isnt the details about how I go about satisfying those needs, but simply on how much attention I need to give in satisfying them. How much do I need to mediate that need-satisfaction by my conscious mind? Writing a letter is conscious activity: I need to think about the words Im putting down, actively formulate sentences, structure my thoughts. Sometimes writing a letter is easier than other times, but its never an unconscious, automatic activity. Walking is an interesting case. When I was learning to walk, I needed to think actively every time I took a step. And today, when Im walking carefully, say on ice or climbing a mountain, I need to think about where I put my feet. But in general, when I walk from place to place I dont need to think about the walking. When Im walking, most often the activity doesnt even cross my mind. And at the extreme is need fulfillment that I never consciously direct activity that I couldnt even consciously direct if I wanted to. My heart takes care of the circulation of blood. Interestingly, its the most important need of all these, but its fulfillment is always unmediated. Whats fascinating about these examples is that the more important activities are less mediated by my consciousness. And the reason for that is that those important activities are regular needs I have, and theyre predictable ones. When the road is predictably flat, I shouldnt need to think about walking; when its predictable that I need blood circulating i.e., always then I shouldnt need to think about that. The only time I should have to mediate the fulfillment of my needs is when theyre unanticipated, unpredictable. When I write a letter, theres no automatic way of anticipating what I intend to say, so thats why I need to mediate it consciously. But for needs that can be anticipated, its a waste of my conscious effort and attention for me to have to think about how to fulfill them. Our ordinary needs should be satisfied automatically. But most technology certainly the PC model is one in which we issue explicit instructions in order to get it to satisfy our needs. It doesnt know enough about us to anticipate our needs, so we need to tell it what to do. And just when we think its starting to become automatic the system crashes and we have to reboot Slide 5: Envisioning the Destination: Moving to an implicit, anticipatory model What Im suggesting is that technology in the future will move from an explicit, instructional model to an implicit, anticipatory one. Instead of issuing instructions as the norm, the norm in the future will be to have our needs fulfilled as the norm, and only as the exception having to issue instructions at all. Just like the example of walking I used above, riding a bicycle is a good way of illustrating the contrast. When first learning to ride, a child needs to issue explicit, conscious instructions to his or her legs to get them to move in the right way. Over time, though, that knowledge becomes fully assimilated, and so becomes implicit, with the body anticipating instinctively what the next move will be. We can see the development of computers as a kind of analogy to this development. When computers were first invented, programmers needed to issue instructions for flipping every gate, and at best in machine code, giving the processor explicit instructions for every move to make. We can see the development of computers as having them learn about us, making them understand more about our humanity and making assumptions about us to make them easier for humans to use. Given that humans dont think in machine code, weve programmed computers to understand us. Still, were very early in the process; computers understand us well enough that we can program in high-level languages, and can interact with them using graphical user interfaces. But thats still a long way from the stage where theyll have enough intelligence built into them that we dont need to be quite so explicit about what we tell them to do. One of the big lessons of the consumer electronics industry is that when the demands on the user are too high, the user will just let needs go unfilled, rather than bothering to issue all those instructions. Programming VCRs is too trite to mention. But theres a fascinating story that illustrates this message profoundly. Its actually a bit of a sad story, told by the neurologist and author Oliver Sacks. One of his patients was a man named Virgil, who had been blind almost since birth. Miraculously, at the age of 55 Virgil had an operation which largely restored his sight. So Sacks tells about what happened after Virgils operation. At first, it seemed as though everything was excellent: his vision, while not perfect, was still quite clear; he had good visual acuity. But that didnt mean he could see. While Virgils eyes were fine, he had never learned to focus, to identify objects, to track them as they moved, to recognize shapes. He had learned to identify things by touch, but never by sight. For Virgil, seeing was a complex intellectual task, involving consciously remembering the visual features of things he saw, associating them with one another, and converting washes of light and color into visual objects. In short, every aspect of seeing, for Virgil, involved conscious mediation. Nothing about seeing was unmediated, as it is for us, who grew up learning those skills. The sad part of the story is that it proved too much for Virgil. Ultimately, seeing was too much work for the payoff, and eventually he effectively went blind again. In a much less extreme and poignant way, but still analogously, there are lots of people whove decided that its not worth the effort to use computers or sophisticated consumer devices. Even if they could use them if they tried, they rightly decide that its not worth the conscious effort. They simply make the decision that its not worth it to put so much mediation into satisfying whatever need the device is meant to satisfy. When needs require too much explicit instruction, we let the need go unfilled rather than devoting the effort to filling them. Slide 6: Envisioning the Destination: Ambient Intelligence The future of the digital living room wont be one where there are lots of gadgets scattered around the place, each requiring instructions to satisfy needs. Instead, it will be one where our environment satisfies our needs, mostly without our having to think about it. In the digital living room in the future, innovation is truly integrated with life, in which our environments become wiser, more comfortable, and more compelling. The products we sell today are responsive to button-pushes that we have to learn. The products our labs are developing today learn us, anticipate our needs, become PRE-sponsive. They will sense much more than button pushes, to better effect, at less effort on the part of the consumer. The first modern computers filled several rooms, then one room, then a 19 rack, then half a desktop, then a lap, then a pocket [Roel holds up the Rex, if you have one available or can find one] now they are about to seemingly become invisible, but in fact theyre going back to the form factor we started with; the whole building, plus all the devices inside it. We know how to design buildings and rooms for any application, and as such, each is a specific type of interface for the human function within them. Its simply inevitable that we give that interface some senses, some memory, and the ability to learn making it active, attentive, responsive, and anticipatory. The technology of the digital living room of the future Im sketching is: 1. Embedded: Many invisible dedicated devices throughout our environment 2. Personalized: The devices know who you are 3. Adaptive: Change in response to you and the environment 4. Anticipatory: The devices anticipate and satisfy your desires as far as possible without conscious mediation The snapshot Im suggesting is one in which technology anticipates our needs, in which the intelligence is ambient much like the light in this room, satisfying our need to see without our even being conscious of it, pervades the entire room. And as long as our needs dont change, the ambient light continues to unmediatedly satisfy the need. This means products with embedded, networked intelligence that are more realistic, in that they are more like how reality works, how life works it is constantly adapting to us in ways were not consciously aware of, like people on a crowded sidewalk. So, were moving from the Desktop metaphor to the Life metaphor, and since were actually putting these processors and sensors into our built environment, its no longer a metaphor its life augmentation, and its digitally personalized. Slide 7: Examples for today, and for the future. In order to get an idea of the sort of world which awaits us, Id like to talk about some interesting experiments and prototypes that are being worked on today. These projects are diverse, some will end up being key to future developments, while many will not. But all of them provide stimulating ideas and insights, and glimpses of our direction. Its critical to learn from our failures as well as our successes. In many cases, experiments and even commercial products that dont do as well as wed have liked and at Philips weve had such products often tell us the most about the users needs. The creation of tomorrows Living Room will be a highly iterative process, and we should be cautious with each step, but ambitious in our goals. A difficulty we face is the complexity of humans, and the enormous divergence we see in their behavior over personality types, age and culture. Anticipating human needs and desires is not an easy task. It requires massive investment in research. At Philips Research we have a staff of 3000 people who live in the future, amongst our six research laboratories in New York, Holland, Germany, the UK, France and Taipei. Rather than day-to-day product development, theyre doing fundamental research into everything from plastic semiconductors to user interfaces, from biometrics to wireless networks, and from 3D LCD displays to wearable computers. Of course we also draw on the work of other research powerhouses such as MIT, Stanford, and PARC. The future will be created not just by software and hardware engineers, but by social scientists, artists, and designers as they carry on their constant dialogue with our culture. Some of our most beautiful work has come from Philips Design, whove been thinking for years about the role consumer electronics will play in society. Slide 8: Experiments in Natural Interfaces Clearly, one basic problem we must continually revisit is how to make our devices easier to use; how to approach our goal of unmediated desire-fulfillment. And this doesnt get any easier as we are able to cram more and more features into each device. At Philips Research we do quite a bit of work on multi-modal interfacesones in which the user has multiple simultaneous means of control, such as speech, touch, and gesture. Each of these is best-suited for certain functions, in certain situations, and the Ambient Intelligence I mentioned lies in all the modes working together adaptively and additively. Such multiplicity, parallelism, and redundancy is going to be the standard means of interfaces for Consumer Electronics. For example, imagine an e-mail device that reads you back your messages in the car. While driving, the e-mails should be read back to you and controlled by voice. However, if youre caught in a traffic jam it may make sense to be able to use a touch screen to quickly browse through the messages alternative and redundant forms of control. The user should be able to change modes, mix and match voice and touch unconsciously, without explicitly changing anything on the application; without breaking concentration on the task at hand. That is using technology to add real value in a product. Similarly, a television is likely to be controlled by a combination of voice, the standard remote, and gesture, recognized by computer vision. Each mode has its own advantages and will appeal to different users. In fact one way of looking at multi-modal interfaces is as a first level of personalization. Voice recognition in particular, is a major focus at Philips Research. Weve all heard extravagant claims for voice recognition in the past, and been disappointed. But this is truly a technology whose time is at hand, and we have been positioning ourselves to be the industry leaders. The silicon to implement reliable voice control for mass-market consumer electronics devices is still too expensive, but it wont be for long. And by locating the hardest parts of speech recognition on servers out on the network rather than packaging it into each device, we can deploy high-end recognition even to low-end devices. We must always be careful, though, of what problem were solving. An early Philips voice-activated remote control was designed to make programming VCRs easier. So instead of setting start and end times by buttons, you could do it by saying record channel 5, start 5.30, end 7.00. Trouble is, it didnt help people at all, because the buttons werent the problem. The problem was that people were still being forced to program a VCR as opposed to recording a program. What they really wanted to say was Record me Seinfeld tonight. Speech recognition needs a foundation of intelligence about the task domain in which the device operates. Another example of how even the cleverest of our researchers can have their expectations happened during our development of voiced-activated televisions. With considerable effort, we built a voice control system that worked with natural language. You could talk naturally to the television and it would obey, even talking back where necessary. Guess what? Our user tests showed that people didnt want to have conversations with their television; they were much more comfortable giving simple voice commands. Several users actually said: I want to command it like a dog, not another person. So the future isnt just a mouse and keyboard, its every form of human communication there is; the Life Metaphor. And as we develop our systems to make them smarter we must constantly test them on real users to make sure were solving the right problem. Slide 9: Experiments in Situational Awareness These smart devices will have to understand not just the way that humans interact, but also their environment and the fact that its constantly in flux. They must understand the notion of Place, and they must talk to each other. One key set of technologies we and others are investigating is home networks, and especially wireless connectivity. Wireless devices are far easier to install, and can be fully mobile. At Philips Research we have prototypes of wireless bridges that allow us to send video from room to room. Intensive research into wireless will help us eliminate the spaghetti which seems endemic to any audio or visual products. And far more important, it will enable the average person to feel confident about installing that home network, since it will be self-configuring. But it also has a deeper significance. Typically we think of mobile phones or pages and how far they reach to a base-station. Wireless has a long reach and infra-red a short one. But many of the wireless devices of the future will be rated on how small their range is, not how great. In a home where hundreds or even thousands of devices are trying to share the only electromagnetic spectrum there is, it is crucial to avoid interference between them. This can only happen if each device has a small range, such as 5 meters, and talks only to its nearest neighbors. Then, devices in adjoining rooms can share the same frequency without problems. This implies a network with many repeaters and nodes, or perhaps a totally decentralized network where each node is also a router. Wireless technologies allow mobility, where devices can change their behavior depending on where they are. Some time ago, Philips Design conceptualized the idea of Control Wands [holds up wand] as part of their Visions of the Future project. These are small, personalized, multi-modal, universal remote controls that are mouse-like in their simplicity. One of the applications that Philips Research have developed with these wands is Follow Me. With Follow Me if youre watching a sports program in the living room and get banished to the den by the kids, you can simply click the wand and carry the program with you. When you get to the den, click the wand again and the nearest display will come to life set to the same program you were just watching. The same principle would work with an audio track on a CD. The activity is something you carry with you, and that you no longer have to manually recreate at your destination. All of this requires a lot of research into sensors. The devices and their network have to be able to read and understand their environment. We are doing far more work in this area than I can go into now, but a few examples will suffice. At MIT there has been interesting work on what are called penny tags. These are very cheap tags that can be used to label and identify a multitude of everyday objects. Imagine that you could label your spectacles, your wallet and your pen with a tag costing less than a penny, and that the network could find it at any time. These tags are like the security tags found on goods in department stores, but embed an I.D. into their electromagnetic properties. Network devices can transmit a wireless signal, and detect the presence of a tagged object. Critically, these are cheap enough to be put on low-cost or throwaway household items. Such smart materials can act as cheap, dumb sensors for heat, pressure and temperature. There are many more examples, but the idea is that the devices that go to make up ambient intelligence will be continuously adaptive to changes in their environment. Slide 10: Experiments in Personalization Now our devices of the future are smarter, but not smart enough. They understand human communication and they understand the environment, but they dont understand me. Lets look at the topic of personalization. I shouldnt have to learn about devices, they should learn about me. They should learn my preferences, my desires and my habits, and anticipate my wishes. They should share this knowledge with other devices in my home but not with devices in my neighbors home, and not with anybody else on the Net. This knowledge has to be controlled by the consumer; a solution that compromises your privacy will be completely unacceptable. Key to personalization is recognition. My home must know that it is me, and not my son, who walks into the room, and be able to differentiate us reliably and non-intrusively. At Philips we have many research projects in biometricsrecognition of humans by machines. We have projects based on computer vision and others that use voice. We think that devices will use various combinations of these methods; each would be an example of an active, multi-modal user interface. Our Large-Area Electronics department has developed a compact, flat fingerprint detector. Standard fingerprint detectors are bulky because of the space required for an optical system. Our detector is solid state and printed onto a sheet of glass. This form factor allows its use in a variety of devices where it has been impractical. Were now working on other substrates such as plastic, and, for example, we foresee these devices being embedded in the steering wheels of cars. Once a device has identified the consumer, it can proceed with personalization. Personalization will typically be a mixture of explicit decisions and implicit choices, interpreted by a learning engine. Im going to show you a fascinating piece of research from our lab in Holland called Double Agent, which shows a combination of personalization and anthropomorphic Intelligent Agents. In a world where a home receives 400 channels of television, the problem is not what to watch, but to simply know what there is thats relevant to watch among the thousands of programs per day. Clearly, plenty of filtering is needed. The Double Agent project constructs a list of programs for a day, which reflect explicit preferences the user has set, such as no sports, as well as implicit ones inferred from their actual viewing habits. (What consumers say they like to watch, and what they actually like to watch are not always the same.) Any programs that meet a certain level of likely interest are automatically recorded on the homes media storage device. Each recorded program is then represented on-screen by an agent, these little creatures you can see here. Different shapes and colors represent different genres of programs. This one is sports, and this is a current-affairs documentary [pointing]. The actual title of the program is superimposed in white. The level of likely interest in the program is represented by the movement of the creature [creatures start to move], so a very excited and lively creature represents a likely high-interest program, while a slow and dopey one is a less-likely prospect. Viewers can gain a quick impression of the viewing landscape by watching a horde of these creatures [shows creatures moving towards camera]. The best programs for a particular viewer tend to stand out from the rest. Programs that have been recorded but not viewed stay stored for a few days. In an early version of the system, each day that passed with a recorded program being ignored meant that its creature became more and more tired looking. However we found that people would watch a program just because they felt sorry for the creature, so theres clearly still a lot more work to be done. Slide 11: Experiments in Ambient Intelligence So far I've described some experiments that push us forward in understanding how to -- and how not to -- design technologies which understand us and our environments, and allow us to interact with devices in the most natural and straightforward ways. Perhaps the most important aspect of the possibilities I'm sketching, though, doesn't just have to do with functionality or naturalness. Perhaps the most radical and important development I think we can look forward to is the move away from a single, "all-powerful" processing box, and away from devices altogether -- even if they're devices everywhere -- toward a world in which everyday objects, the objects that we don't even give a moment's thought to, become more and more intelligent. Maybe the best example of this is a simple light bulb. [holds up a light bulb]. There are scores of them in every house; now think of them as points-of-presence -- more than even nodes on a network -- becoming individual sites of intelligence. The pervasiveness of intelligence, the ambience of intelligence, will involve everyday objects even more than it will involve computing. To me, computation is important to this vision; consumer devices even moreso; but what's truly essential is the ambience, the universality and pervasiveness, bringing intelligence into every corner of the living room, even the most mundane parts of it. The economics of the future digital living room will reflect this change. It's not the economics of selling large pieces of high-tech equipment that'll be the business model of ambient intelligence. Clearly much of the differentiation in bringing intelligence to our environments will continue to involve huge investment in developing the intellectual property for understanding and fulfilling human and individual needs. In terms of deploying ambient intelligence, also, the economics must look very different. It can't occur, if we want to add even $1, even 50 cents, to the cost of a lightbulb. Ambient intelligence demands an economics for development, and a development for the pervasive points-of-presence I've discussed. To this end, Philips is developing a class of technologies that start to make this economics possible. We're working on plastic semiconductors, where for a fraction of the price of today's semiconductors we'll be able to embed intelligence into low-cost objects. We're also using conventional technologies to bring high-end computing to today's devices. The TriMedia chip can perform operations 200 times as fast as a Pentium, but costs an order of magnitude less. We and others are using TriMedia to bring an unprecedented level of quality to video, at an unprecedentedly low cost. Slide 12: Pervasiveness of Consumer Devices Whether the economics of the emerging technologies I've been talking about will be more like that of PCs, consumer electronics, or lightbulbs, what I'm describing will in any case involve a great deal more than one high-powered box sitting on your desktop. The living room today has hundreds of items, each of which satisfies a need or set of needs. Each of these items is a potential point-of-presence for embedded, adaptive intelligence. The force of this fact can be seen just by comparing the total number of points-of-presence in the home that all computer companies, all over the world, put into people's homes, compared to the number of items placed by one single company with the strength and diversity of Philips. All computer companies together, in 1998, shipped on the order of 20 million PCs to consumers. This sounds like a huge number, and it is huge, understanding it as the first major entry of intelligence and processing power into the home. But compare it to the points-of-presence of one company like Philips; the numbers start to look completely differentthe scale shifts. Philips ships 2.4 billion incandescent light bulbs each year. 30 million television sets and computer monitors. 11 million shavers. 18 billion semiconductors. The so-called "computer revolution" has been a revolution already. But now that it is beginning to migrate both intelligence and senses into our devices, our homes, and our lives, it's an entirely new game. Slide 13: Conclusion Perhaps the key thing we must recognize is that the technology is a means, not an end. The consumers were designing these products for are not early adopters, and they have zero to no tolerance for complexity. Fortunately, there are so many of them that our hard work will pay off. The living room will only be digital insofar as it recognizes, builds on, and embraces our humanity. Our homes and tools will complement us, adapting to our environments, and anticipating and fulfilling our needs. Only when theres a seamless integration of technology with life, only when technology is no longer a curiosity but an ordinary and unsurprising way of satisfying our everyday needs and desires only then will we have seen the beginnings of the true technological revolution. Technology can enhance life but doesnt have to take us out of it. . . . because the Living Room isnt about technology, its not about convergence, its about your six-year old daughter giving you a kiss when you come home from work. Useful quotations, which can be included at any of several points in the speech: ______________________________________________________________ . . . one of the principal uses of design is the articulation of difference. the writer Ralph Caplan (use this quote to say that personalized C. E. devices is an adaptive, digital way to articulate personal lifestyle differences.) For Caplan a personally important place "has a measure of you in it." . . .Sounds like what the sensing and adapting consumer electronics devices do. _____________________________________________________________ "Our senses . . reach far beyond us. They're an extension of the genetic chain that connects us to everyone who has ever lived; they bind us to other people and to animals, across time and country and happenstance." Diane Ackerman, A Natural History of the Senses. "Writers cannot presume shared knowledge, so they must be explict where a speaker is implicit; precise and careful where a speaker can be careless; streamlined and sparse where a speaker can be redundant." Ellen Dissanayake, Homo Aestheticus. Where Art Comes From and Why. (This is pertinent to the speech in parallel to the way that gesture and language, when parsed by your smart television or stereo, are more casual and unconscious ways to operate the devices; requiring less focus and conscious intent of the human user.) . . . the intercommunicating nature of perceptions "While normal waking consciousness works to simplify perception, allowing us to act quickly and flexibly by helping us remain seemingly oblivious to almost everything except the task in front of us; simultaneous perception is more like an extra, or a sixth, sense: It broadens and diffuses the beam of attention evenhandedly across all the senses so we can take in whatever is around us -- which means sensations of touch and balance, for instance, in addition to all sights, sounds, and smells." Tony Hiss, The Experience of Place "Islamic literature delights in the image of the reed as pen for the calligrapher and instrument for the musician, both revealing man's inner thoughts to the different senses." Sheila Markham, "Islamic Calligraphy" What constitutes place? Interconnectedness. A simple and profound aspect of place is our reciprocal relationship with our environments: Not only do we change them, but they change us as well. "Our relationship with places is a close bond, intricate in nature, and not abstract, not remote at all: It's enveloping, almost a continuum with all we are and think." Tony Hiss's book The Experience of Place (This quote seems to fit well with the speech; talking about how our new Ambient Intelligence helps integrate us into a space.) In order to envision the digital living room, in order to have a chance at glimpsing the way the home and our most personal environments will be evolving over the next decades, we need to expand our vision beyond the living room. What the digital living room is, will affect and be affected by how our whole environments change, how we live our lives in and out of the home. So I understand the digital living room conference, and the object of this speech, to be both broader and deeper than the digital living room alone. PAGE 1 PAGE 13 INTERNAL WORKING DRAFT / THOUGHT-PIECE '|} PQ3 9 ""o###$$+$A$222v46[6A88888 99E9R99V;_;<<<-=@BzE)F>IrItIxINN OORRRS$Z3Z[[Qa]aZb\bQfUfiik"kCk>*CJ CJhnH 6CJhnH  56CJCJ55CJ6CJCJ6T'(Zgh|}tu D E PQ'(Zgh|}tu D E PQ67>-. $"%"n#o#     q xyCDOP 767>-. $"%"n#o### & Fo###$$&&l(m(,*-*s+t+>-?-//V0W02222u4v455A8B8889A999 ;;<<-=/=>>@@BBPDQD|D}DyEzE)FXGYGsItIIIJJJMKMOORRSSMSNS3T4TgVhVqYrY]]``aaaaaaabb6d7dggxiyi`#$$&&l(m(,*-*s+t+>-?-//V0W02222u4v455A8B88899A999 ;;<<-=/=>>@@BBPDQD|D}DyEzE)FXGYGsItIIIJJJJMKMOORRSSMSNS3T4TgVhVqYrY]]``aaaaaaabbb6d7dggxiyilloopppppss1v2vyyzzZ{[{{{B}CkFkikqkm mmmnn|ixV5Me%7:F>>>Вђגْؒڒےݒޒ0JmH0J j0JU5CJ CJhnH CJ6CJ1yilloopppppss1v2vyyzzZ{[{{{B}C}MNxyIJ_`{|RShiTUV7Ӈ>?>?@uvg@ghij֏=>Вےܒݒ  ZB}C}MNxyIJ_`{|RShiTUV7Ӈ$&d >?>?@uvg@ghij֏=>&d  & F&d &d Вےܒݒh&`#$ / =!"#$% [$@$NormalmH 0@0 Heading 1$@&CJ4@4 Heading 2$@&5CJ0@0 Heading 3$@&5<A@<Default Paragraph FontJC@JBody Text Indent   9B*hnH ,@,Header  !&)@& Page Number*B@"* Body TextCJ, @2,Footer  ! AAAAADCkKT#9JbB}LNPQRSVWXo#yiMOU D!!/X2$t RV̕|" 2$ gvEc* ܩb@0(  B S  ?zqqttxvvvw@wOwRY GNЎ.AI 8 7>cd ''K(Z(13,31344V9a9:BEBCC _PID_GUIDAN{9C1E9F29-004D-11D2-BE83-00008610BD29}  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXY[\]^_`abcefghijkmnopqrsvRoot Entry F`M{-U.x1TableZWordDocumentSummaryInformation(dDocumentSummaryInformation8lCompObj&'()+,-/05678jObjectPoolU.U.  FMicrosoft Word Document MSWordDocWord.Document.89q