The 21st-century citizen will see a proliferation of new electronic technologies: devices with heterogeneous interfaces and advanced services available at home, in the car, in the school, in public spaces. This connected world poses many challenges, opportunities, and risks, the greatest being that its full benefits will not reach most of the intended users. As computers and networks are increasingly integrated with everyday environments, this "ambient intelligence" creates new synergies and dependencies between individual users and a multitude of services and applications, which should become accessible through easy-to-use, almost instinctive, human interfaces. This is a new complex and interdependent ecosystem that will make or break the realization of a truly connected society. However, building and instituting this landscape require greater clarity about the interdependencies (conceptually and technologically) that constitute this ecosystem.
Along with interface design, one of the key enablers in the foregoing scenario is the question of digital identity. The need to identify individual users and provide them with intelligent personalized assistance based on their needs in specific locations and contexts remains the basic condition for reaching this new landscape. Intelligent personalized assistance (as opposed to customized customer relations management garnered through inferred behavior) must fulfill three conditions. It must:
- Help the user to find the right services to use.
- Handle the services for the user (by making novel user interaction techniques user-friendly and using self-managing mechanisms based on context-awareness).
- Establish trust.
To demonstrate the technological complexities of digital identity and ambient intelligence in the network, I present user scenarios based on location, life events, and social interactions.
Location-oriented sectors encompass the home, the office, and hot spots (such as cinemas, airports, shopping centers, and traffic), which a person may be on the way to or in. Each of these sectors has specific sets of requirements for dedicated personalized services.
The home is one example of this complexity: As domestic equipment gets more sophisticated and these devices have the ability to communicate with each other, advanced sensors will provide information that will enable remote control; the mobile telephone is the prime candidate for the home remote control device. Information, entertainment, and the whole environment will be adapted to the needs and moods of the user. The citizen will no longer be restricted to the internal space provided by architecture, but can be connected to the world through the "windows" provided by broadband and wireless connectivity and communications. Tools such as home-area networks and personal-area networks (PAN), image identification systems, location, bio-sensors, natural interfaces (gestures, voice, body movement), and automated learning machines will all merge, presenting a myriad of new possibilities. Yet, realizing these possibilities depends on the ability of the network to identify individual users and assign their profiles according to their preferred contexts and moods.
Although life events will always be location-based, they should also be understood as being linked to actions, tasks, or desires users undertake in their everyday lives. Being ill in a hospital is not just about being located in a hospital. The requirement for specific services depends on numerous conditions, of which the level of illness is just one.
Applying individual user profiles in context requires that these profiles be portable and linked to intelligence that delivers services based on an awareness that the user is, for instance, not at home, or not in their home country.
Social relationships, or communities, derive from interactions that are local, regional, or international in origin: the birth of a child and the sharing of his or her birth with members of a family across a global network defines one sort of interaction. Playing a sport defines another, in this case based on common interests. In these scenarios, these communities require different technologies and infrastructures to sustain them and different tools to establish, configure, and manage participation. Example of such technologies are peer-to-peer (P2P) and Web services. Adding the requirement that, as with individual user identities, group identities and profiles should be portable makes this one of the most complex areas for the future.
The problem with this kind of discrete scenario planning is that it can obscure the reality that an individual user can, and often does, play all these roles. During one day, a user will have different roles, perhaps encompassing work, his private life, and sports. Some of these roles may conflict. In business mode, the user might want to get information as quickly as possible and in a particular form, regardless of the price; in a private role, his preference might be affected by price but will be determined by personal value. To progress from digital identity to personal profiles will require developing a complex set of scenarios, and these profiles will require further categorization to facilitate delivery. Profiles should also be semi-permanent and dynamic.
Semi-permanent personal service profiles would be primarily based on a user's roles (e.g., work, family), community (described earlier), user/service profile (e.g., soccer, art), user behavior (update of user/service profile), context (what am I doing right now, where am I doing it, what device am I using to do this?), creation of content (e.g., information for my sports community), and localization (e.g., local services).
The dynamic part of the personal service profile is based on access technologies. This poses the question of which access technology is the best in the current context.
Various access technologies must be assessed against a variety of well-chosen parameters. These parameters should be generic enough to allow the technologies to be compared. The following list details the various parameters and their relevance in defining a scenario:
- Application (Quality of Service) support: negotiation between the application and the access capabilities must take place.
- Data rate: probably the most important parameter, data rate sets the limit on exactly which services can be supported via an individual access technology.
- User mobility: another of the most important parameters; sets a limit on the situations in which a service may be supported.
- Terminal equipment: the user will have a range of terminals, either as part of a PAN, or as supporting equipment in the vicinity.
Defining personal services follows a two-step approach. The first step, after identifying the individual user, is information retrieval, based on the preferences of the user and the role she is playing at that moment. If information is found to be valuable for her and it fits into her current context, a request is launched to establish the current access scenario, and the information is then scaled and adapted to best match it. Some examples taken from personal communication services include unified messaging, location-based call forwarding, and personal Web portal access.
Service architecture is the final element in the delivery chain, helping to realize services according to information, communication, and personalized needs of the user. The service platform will include the following important distributed components and make them work together:
- Personal service profile: consisting of a semi-permanent part and of a dynamic part called access profile (as previously discussed)
- Security infrastructure: integrated into the service architecture so that it is consistent and easy to use
- Management of terminal and user equipment: allowing command of machine-to-machine communication to support automatic synchronization between user mobile phones and the residential gateway
- Management of services: personalization of environments and services based on these features
It follows from this discussion that the user-interface design of future access and communications devices cannot be based on one-size-fits-all. More important, context awareness means user interface design will increasingly have to link or relate to embedded intelligence to ensure maximum flexibility, adapting to mood or task or location. Identifying the individual user; provisioning his or her profile; and delivering the right content or service according to preference, location, and context will mean little if that information cannot be easily assimilated or used.
The challenge is immense. Apart from the specific user-interface questions posed by this ubiquitous connectivity (and the security implications, which have merely been hinted at), realization of these scenarios will depend on different sectors of the industry acting together, developing common standards and cooperating at an unprecedented level. This remains the biggest challenge in a sea of challenges. However, the key message we should bear in mind is that user-interface design should be regarded as an intrinsic part of the design and deployment of these systems, not a convenient afterthought. Although digital identity and its management remains the fundamental building block for delivering an era of personal services embedded in ambient intelligent networks, user interfaces will be the foundation on which this structure stands or falls.
This is written in a personal capacity.
Dr Norman Lewis is Director of Technology Research for the European Internet services provider Wanadoo SA (part of the France Telecom Group). Prior to this he was the head of research at cScape Ltd. He was also the Director of Globalisation and Power in the 21st Century (GAP21), a forum for debating the social impact of information and communications technologies on 21st century society. He holds a Ph.D. in International Politics from the University of Sussex, U.K., and has lectured at a number of universities in the U.K. and Europe.
©2004 ACM 1072-5220/04/0300 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2004 ACM, Inc.