Claus Unger

Using Handheld Devices in Synchronous Collaborative Scenarios

Abstract: In this paper we present a platform specially designed for groupware applications running on handheld devices. Common groupware platforms request desktop computers as underlying hardware platforms. The fundamentally different nature of handheld devices has a great impact on the platform, e.g. resource limitations have to be considered, the network is slow and unstable. Often, personal data are stored on handheld devices, thus mechanisms have to ensure privacy. These considerations led to the QuickStep platform. Sample applications developed with QuickStep demonstrate the strengths of the QuickStep environment.

Methods for Protecting a Mobile Agent's Route

In the world of mobile agents, security aspects are extensively being discussed, with strong emphasis on how agents can be protected against malicious hosts and vice versa. This paper discusses methods for protecting an agents route information from being misused by sites en route interested in gaining insight into the profile of the agents owner or in obstructing the owners original goal. Our methods provide visited sites with just a minimum of route information, but on the other hand allow sites to detect modifying attacks of preceding sites. Though, under noncolluding attacks, all methodes presented provide a similar level of protection, they differ w.r.t. performance and the points of time when an attack can be detected.

Protecting a Mobile Agent's Route against Collusions

. In the world of mobile agents, security aspects are extensively being discussed, with strong emphasis on how agents can be protected against malicious hosts and vice versa. This paper discusses a method for concealing an agents route information from being misused by sites en route to collect profile information of the agents owner. Furthermore, it is shown that the protected route resists attacks from a single malicious host and from colluding malicious hosts as well.

Group Rendezvous in a Synchronous, Collaborative Environment

Before a session in a synchronous, collaborative environment can really start, various actions have to be performed: users have to be informed about planned sessions, network access paths between session participants have to be determined, etc. In the following, we call all these actions group rendezvous. In existing groupware systems, the group rendezvous is often neglected, i.e. it is assumed that a team has already formed, session information has been distributed and network paths are known. In reality, this assumption often does not hold: members with dial-up connections are not permanently online, if they get their network addresses from an address pool, e.g. via a network access provider, they are difficult to find even when they are online. This paper describes a fully decentralised group rendezvous system, which addresses these problems. Besides session distribution, the system has a component for resolving variable network addresses. It has been integrated into the synchronous groupware system DreamTeam [RU98].

Flexible synchronization of shared groupware objects

In this paper we describe an object manager, that has been developed for simplifying groupware development. The object manager offers a variety of services for controlling an applications shared data, including flexible object distribution, flexible coupling of user interface and shared data, latecomer support and configurable concurrency control. After a brief description of the object manager in total, we focus on a configurable concurrency control scheme. An example and directions for future work conclude the paper.

Fault Tolerant Geographical Addressing

Geographical addressing and resource discovery are important building blocks for creating mobile location-aware systems. Mobile systems rely on wireless network technologies, which are prone to frequent failures. In this paper, we present a protocol for maintaining a self-organising routing backbone that supports geographical addressing and resource discovery. The routing backbone detects node failures and automatically repairs its infrastructure, while maintaining the operability of the network at least in partitions an thus maximising service availability. The system operates in a peer-to-peer fashion, i.e. the administrative tasks are distributed between all participating nodes, taking into account their capabilites, location, and mobility. In this paper we explore maintenance aspects of the network that are caused by node failures.

Abstracting operators for hierarchical planning

Publisher Summary This chapter discusses abstracting operators for hierarchical planning. The operator hierarchy produced by the algorithm is a type hierarchy, that is, the abstract operators are not merely macro operators representing a fixed, possibly still unordered set of concrete operators, but in fact are new operator types subsuming all possible combinations of their constituting more concrete ones. The approach significantly diminishes the number of operators from abstraction level to abstraction level and this method allows a much more flexible refinement and, therefore, is independent from the designers view of possible actions in the domain. The post condition of a concrete operator describes which post states can be caused by the operator if it is applied to a pre-state, which meets the corresponding precondition. With regard to abstract operators, and their use in a planning system, post conditions serve two purposes. They have to describe post state conditions: (1) which are inevitably caused whatever refinement by the constituting more concrete operators will be chosen, and (2) which are able to be caused by one or several appropriate more concrete operators. While in case of a concrete operator, both purposes are met by the same post condition, in case of abstract operators, the purposes conflict.

Engineering for Human-Computer Interaction

Wearable computers are beginning to appear in research environments and soon will be commercially available. We present several different philosophies behind wearable computers and then focus on a particular project at Carnegie Mellon University involving the development of a wearable designed for on-site maintenance of large vehicles. This computer is discussed in terms of its projected use (both solo and collaborative), its development process (rapid), and its physical characteristics. The paper concludes by discussing some implications of wearable computers on current user-interface evaluation techniques, on user-interface development environments, and on models for cooperative work.