Some Computer Science Issues in Ubiquitous Computing Presenter : Junghee-Han Mark Weiser Communications of the ACM, July 1993

What was Ubiquitous Computing in 1993 ? Hardware prototypes Issues of Ubiquitous Computing Hardware components Network protocols Interaction Substrates Applications Privacy of location Computational Methods Conclusion Contents 2

Very new to the field of Computer Science Ubiquitous Computing was 1 st defined by Mark Weiser at Xerox Palo Alto Research Center(PARC) in 1991 Three Waves of Computing  Main Frame Many Person - One Computer  Personal Computer One Person - One Computer  Ubiquitous Computing One Person - Many Computer Ubiquitous Computing in Ubiquitous Computing

Ubiquitous Computing is  Next-generation computing environment  Each person is interacting with hundreds of nearby wirelessly interconnected computers  Essentially invisible to user  New kind of computers of all sizes and shapes will be available to each person Ubiquitous Computing is  Quite different from personal digital assistants  Computers should be autonomous agents The Challenge is  To create a new kind of relationship of people to computers Ubiquitous Computing in Ubiquitous Computing

Virtual Reality puts people inside a computer-generated world Ubiquitous Computing forces the computer to live out here in the world with people VR vs. Ubiquitous Computing 5 Ubiquitous Computing

Hardware Prototypes The physical world comes in all size and shapes Three different sizes of devices Large-size: Liveboard Medium-size: Xpad Small-size: ParcTab A pervasive part of everyday life With many active at all times 6 Ubiquitous Computing

Large-Size Prototype LiveBoard Main idea was to simulate a office whiteboard Order of 1 per office Xerox product Hardware Prototypes 7 Ubiquitous Computing

Hardware Prototypes 8 Medium-Size Prototype Xpad Main goal was to simulate a personal notebook Order of 10+ per person Design focus The right balance of features The requirements for particular features Ease of expansion and modification Ubiquitous Computing

Hardware Prototypes 9 Small-Size Prototype ParcTab Information doorway Main goal was to simulate PostIts Order of 100+ per person Design Philosophy Put devices in everyday use Design Problem Size and Power Consumption Ubiquitous Computing

To construct and deploy prototypes  Need to readdress some of the well-worked areas of existing Computer Science Organization of a computer system  Hardware components  Network protocols  Interaction substrates  Applications  Privacy of location  Computational methods The Computer Science of UbiComp 10 Ubiquitous Computing

Low power  Work to reduce power rather than to increase performance Wireless  Permits reuse of the same frequency again  Permits transceivers that use low power Infrared Pens  Work over large area  Need not touch the screen  Operate from several feet away There is a need to balance performance and power usage Issues of Hardware Components 11 Ubiquitous Computing

Network Protocols Wireless media access  Use Multiple Access Collision Avoidance (MACA)  For fairness Same back-off parameter for all neighbor stations Real-time protocol  To support multimedia applications Mobility support  Device in one region moves other regions  IP can’t solve this: Mobile IP Enabling media access while continuously moving between spaces 12 Ubiquitous Computing

Interaction substrates Tabs  Have a very small interaction area  “Touch typing” that uses only a tiny area Liveboards  Using conventional pull-down or pop-up menus requires walking across the room  Location-independent interaction is need X-window system  User may move from device to device, and want to bring windows along  Window migration tools Removing technical barriers between user and device 13 Ubiquitous Computing

Applications The location of people  Information about Location can be deduced from logins or collected from an active badge system  Update the location database Shared meeting tools  Pen-based drawing on a surface  Several users simultaneously operate independently on different or same pages Location aware and allowing multiple users 14 Ubiquitous Computing

Privacy of Location In cellular system  Traveling pattern of cellular phone user can be deduced from the roaming data  Much worse in ubiquitous computing Preserving privacy of location  Central DB of location information  Store information at each person’s PC  Short-term accumulation of location information Transmission and sharing of data must be evaluated in a social context 15 Ubiquitous Computing

Computational Methods 16 Optimal Cache Sharing Problem  Optimal strategy for partitioning memory between compressed and uncompressed pages Requires methods to handle cache misses over high latency mediums Ubiquitous Computing

Conclusion 17 Ubiquitous Computing is  Making many computers available throughout the physical Environment  While making them effectively invisible to the user Ubicomp seems likely to provide a framework  For interesting and productive work  For many more years Have much to learn about the details Ubiquitous Computing