D
elivering health services to the elderly is becoming a major societal problem in many countries, especially in countries where there are relatively fewer and fewer young people to take care of more and more elderly. The problem is already acute in Japan, and it is expected to be very serious in 10 to 15 years in several European coun-tries and in China. Managing and delivering health care in-volves large number of diversified decisions, ranging from allocation of resources to determining what treatment to provide to each patient at each given time.Elderly residents in assisted-living facilities require dif-fering levels of care. Some residents need minimal assis-tance, others have short-term memory problems, and yet others have more severe problems like Alzheimer’s disease so they require more supervision and help. At Elite Care’s Estates Cluster Residential Care Facility in Milwaukie, Ore-gon, pervasive computing is being used to increase the au-tonomy and care level of all of its residents, regardless of their individual needs.
Elite Care, a family owned business (elite-care.com), has been built from the ground up to provide “high tech, high touch” programs. Its advisory committee, which includes among others representatives from the Mayo Clinic, Har-vard University, the University of Michigan, the University of Wisconsin, and Sandia National Laboratory, has con-tributed a number of ideas that have been put into practice.
The entire facility is wired with a 30 miles network of (wireline and wireless) of unobtrusive sensors and other de-vices including: biosensors (e.g., weight sensors) attached to each resident’s bed; movement sensors embedded in badges worn by the residents and staff (wearable computers); panic buttons used to call for help; Internet access via touch screens in each room; video conferencing using Webcams;
and climate control, lights, and other regulated appliances.
These devices and others allow the staff to monitor various patient activity. For example, staff can determine the loca-tion of any patient, to tell whether he or she is in an expected area of the facility. Devices that monitor length of
absence from bed might alert personnel that the patient has fallen or is incapacitated in other ways. Medical personnel can watch for weight loss (possibly indicating conditions like impending congestive heart failure), restlessness at night (indicating conditions like insufficient pain medication), and frequency of trips to the bathroom (indicating medical prob-lems like infection). Also, close monitoring of conditions en-ables staff to give medicine and/or other treatments as needed, rather than at predetermined periods. All of these capabilities enable true one-to-one care, which is both more effective and less expensive.
One of the initial concerns with these monitors is that the privacy of the residents would be unnecessarily invaded. To alleviate these concerns, residents and their families are given the choice of participating or not. Most choose to par-ticipate because the families believe that these monitors pro-vide better tracking and care. The monitors also increase the autonomy of all the patients because their use reduces the need for staff to constantly monitor residents in person, es-pecially those with more acute care needs.
All of these sensors and systems are connected through a high-speed Ethernet (see Tech Guide 4). The data pro-duced by the sensors and systems is stored in a database and can be used to alert the staff in real-time if necessary.
These data are used for analytical purposes or for develop-ing individualized care programs. The same database is also used for administrative purposes such as monitoring staff performance in timely delivery.
A similar concept is used in Swan Village of Care in Bentley, Australia. At the present time such projects are experimental and expensive, but some day they will be affordable to many.
For Further Exploration: What types of data do these devices provide? How can pervasive computing increase the quality of elder care? How to consider the privacy issue?
Sources: Compiled from Stanford, 2002, elite-care.com, and ECC.
online.wa.gov.au/news (January 14, 2003).
6.10 INHIBITORS AND BARRIERS OF MOBILE COMPUTING 283
When mobile Internet users visit mobile Internet sites, the usability of the site is critical to attract attention and retain “user stickiness” (the degree to which users stay at a site). There are three dimensions to usability, namely effectiveness, efficiency, and satisfaction. However, users often find current mobile devices to be ineffective, particularly with respect to restricted keyboards and pocket-size screens, limiting their usability. In addition, because of the limited storage capacity and information access speed of most smartphones and PDAs, it is often difficult or impossible to download large files to these devices.
Mobile visitors to a Web site are typically paying premium fees for connec-tions and are focused on a specific goal (e.g., conducting a stock trade). There-fore, if customers want to find exactly what they are looking for, easily and quickly, they need more than text-only devices with small screens. In 2003, many WAP applications were still text-based, and had only simple black-and-white graphics. This made tasks such as mobile shopping difficult. Because all the transactions were essentially text-based, mobile users could not “browse”
an online picture-based catalog. However, more and faster multimedia are becoming available as 3G spreads.
The major technical and other limitations have slowed the spread of m-commerce are summarized in Table 6.5.
Several ethical and legal issues are unique to mobile computing. For example, fashion retailer Benetton Group SpA was considering attaching RFID “smart The Usability
Problem
TABLE 6.5 Technical and Other Limitations of Mobile Computing Limitation Description
Insufficient bandwidth
Security standards
Power consumption
Transmission interferences
GPS accuracy WAP limitations
Potential health hazards
Legal issues
Human interface with device
Complexity
Sufficient bandwidth is necessary for widespread use and it must be inexpensive. It will take a few years until 3G is in many places. Wi-Fi solves some of the problem.
Universal standards were not available in 2003. It may take 3 or more years to have them.
Batteries with long life are needed for mobile computing. Color screens and Wi-Fi consumer more electricity, but new chips are solving some of the power-consumption problems.
Weather and terrain problems as well as distance-limited connection exist with some technologies. Reception in tunnels and some buildings is poor.
GPS may be accurate in a city with tall buildings.
According to mofileinfo.com, in 2002 there were only about 50,000 WAP sites (compared to millions Web sites). WAP still is a cumbersome process to work with.
Potential health damage from cellular radio frequency emission is not known yet. However more car accidents are related to drivers who were talking (some places bar the use of cell phones while you drive). Also, cell phones may interfere with sensitive medical devices.
Potential legal issues against manufacturer of cell phones and against service providers exist, due to the potential health problems (Borland, 2000).
Screens and keyboards are too small and uncomfortable and tedious for many people to use.
Too many optional adds-on are available (e.g., battery chargers, external keyboards, headset, microphone, cradles). Storing and using the optional add-ons is a problem to some.
Ethical and Legal Issues
tags” to its Sisley line of clothing to help track shipping, inventory, and sales in the company’s 5,000 stores worldwide. (Also, the tags could help prevent shoplift-ing.) The idea was to integrate the RFID tag into the clothing labels. Using the tags, the store would know where each piece of clothing is, at any given time.
However, privacy groups expressed concern that the tags could also be used to track buyers, and some groups even urged that the company’s clothing be boy-cotted. As a result, Benetton backed away from the plan, at least until an impact study is done (Rosencrance, 2003).
According to Hunter (2002) privacy is in a great danger in the world of ubiq-uitous computing because of the proliferation of networked devices used by indi-vidual, businesses and government. The Elite-Care project described in IT At Work 6.5, for example, raised the issue of protecting information collected by sensors.
Also, privacy is difficult to control in other types of context-aware systems (e.g., see Jiang and Landay, 2002). As indicated earlier, security is especially difficult in Wi-Fi systems.
For pervasive (ubiquitous) systems to be widely deployed, it is necessary to over-come both the technical and ethical/legal barriers associated with wireless com-puting, plus overcoming other barriers unique to pervasive computing. Davies and Gellersen (2002) provide a comprehensive list of both technical challenges, social and legal issues, economic concerns (including finding appropriate business mod-els) and lack of experiences in deploying ubiquitous systems. They also cite research challenges such as component interaction, adaptation and contextual sensitivity, user interface interaction, and appropriate management mechanisms.
As with any other technology, especially a new one, there have been many fail-ures of both applications as well as entire companies in mobile computing and m-commerce. It is important to anticipate and plan for possible failures as well as to learn from them. The case of Northeast Utilities provides some important insights.
According to Hamblen (2001), Northeast Utilities (located in Berlin, Connecticut), which supplies energy products and services to 1.2 million customers from Maine to Maryland, embarked on a wireless project in 1995 in which its field inspectors used wireless devices to track spills of hazardous material and report them to head-quarter in real time. After spending a year and a half and $1 million, the project failed. Some of the lessons learned are:
● Do not start without appropriate infrastructure.
● Do not start a full-scale implementation; use a small pilot for experimentation.
● Pick up an appropriate architecture. Some users don’t need to be persistently connected, for example.
● Talk with a range of users, some experienced and some not, about usability issues.
● Users must be involved; use bi-weekly meetings if possible.
● Use wireless experts if you are not one.
● Wireless is a different medium from other forms of communication. Remember that people are not used to the wireless paradigm.
Having learned from the failure, Northeast made its next wireless endeavor a success. Today, 15 field inspectors carry rugged wireless laptops that are Challenges
in Deploying Ubiquitous Systems
Failures in Mobile Computing and M-Commerce
6.10 INHIBITORS AND BARRIERS OF MOBILE COMPUTING 285
connected to the enterprise intranet and databases. The wireless laptops are used to conduct measurements related to electricity transformers, for example. Then the laptops transmit the results, in real time, to chemists and people who pre-pare government reports about hazardous materials spills. In addition, time is saved, because all the information is entered directly into proper fields of elec-tronic forms without having to be transcribed. The new system is so successful that it has given IT workers the confidence to launch other applications such as sending power-outage report to executives via smart phones and wireless information to crews repairing street lights.