The purpose of our project is to provide toolkits for building applications that support collaboration between people in distributed environments. In this paper, we describe one such toolkit, called DistEdit, that can be used to build interactive group editors for distributed environments. This toolkit has the ability to support different editors simultaneously and provides a high degree of fault-tolerance against machine crashes. To evaluate the toolkit, we modified two editors to make use of the toolkit. The resulting editors allow users to take turns at making changes while other users observe the changes as they occur. We give an evaluation of the toolkit based on the development and use of these editors.

A great interest has developed in recent years in building tools that allow people to collaborate on work without the need for physical proximity. One such class of tools, group editors, allows collaborators to view and edit a shared document simultaneously from their workstations. Building group editors, however, requires solving non-trivial problems such as providing adequate response time for edit operations and yet ensuring consistency with concurrent updates, and providing adequate per-user undo facilities. We have implemented a toolkit, called DistEdit, for building new interactive group editors and for converting existing single-user editors into group editors with minimal changes to their code. The toolkit allows different users to use their favorite editors (e.g., Xedit, Gnu Emacs) to edit a shared file and observe each others ' changes as they occur. The toolkit provides fine-grain concurrency control, fault-tolerance, synchronization of views, and support for per-user undo. ...

this paper, we describe a paradigm and its associated collaboration artifact to allow flexible support for asynchronous collaboration. Under this paradigm, a user session with an application's user interface is encapsulated into a data artifact, referred to as a session object. Users collaborate by annotating, by modifying, and by a back-and-forth exchange of these session objects. Each session object is composed of several data streams that encapsulate audio annotations and user interactions with the application. The replay of a session object is accomplished by dispatching these data streams to the application for re-execution. Re-execution of these streams is kept synchronized to maintain faithfulness to the original recording. The basic mechanisms allow a participant who misses a session with an application to catch up on the activities that occurred during the session. This paper presents the paradigm, its applications, its design, and our preliminary experience with its use. Introduction

This proposal outlines a plan for the design, deployment and operation of a high speed campus network at Washington University based on fast packet switching technology that has been developed here during the last several years. This new network will support ubiquitous multimedia workstations with high-resolution graphics and video capabilities, opening up a wide range of new applications in research and education. It will support aggregate throughputs of hundreds of gigabits per second and will be designed to support port interfaces at up to 2.4 Gb/s. Initial implementations will emphasize 155 Mb/s port rates, with higher rates introduced as the demand arises and as economics permits. We propose to move this technology quickly into an operational setting where the objectives of network use and network research can be pursued concurrently. Contents 1 Motivation and Objectives 4 2 Applications 5 3 Networking Questions 7 4 Creating the Network Components 8 4.1 ATM Network Technology ...

The trend in computing models has changed from thin-client (text terminals) to thick-client (graphics terminals) to standalone personal computing (PCs), and then back to thick-client (client/server applications) and thin-client (Web-based applications) again. This trend is now leading us to the so-called stateless-client computing, which is an ultra-thin-client model that frees the client completely from preserving any application state. This paper explains how we integrate synchronous and asynchronous sharing of workspace with Virtual Network Computing, a stateless-client computing technology developed at the AT&T Laboratories Cambridge. Experiments and applications have demonstrated that our collaborative system is feasible for operation in current and future computing environment.

Distributed multipoint applications for group interaction across wide-area networks, such as for simulation and telecollaboration, are becoming increasingly popular. While reliable multicasting has made significant advances in recent years, effective mechanisms to synchronize and coordinate work within large multicast groups and across long distances are still lacking. Synchronous sharing of resources, whose operational semantics prohibits parallel usage, typically creates race conditions among users, which can be resolved through an access discipline called floor control. Existing solutions on floor control, implemented either at the session or application layer, are mostly proprietary, limited in scope and not scalable. Furthermore, no performance comparison of floor control protocols has been attempted to date. We present a novel taxonomy and comparative performance analysis of known classes of floor control protocols, ranging from socially mediated control to protocols operating on ring and tree topologies. We find that aggregation and selective transmission of control information in a tree structure is the most promising solution with regard to scalability, efficacy, and robustness. The principal operation of such a tree protocol is outlined, which dynamically organizes participants in a multi-level control tree and aggregates resource sharing directives on the paths between interacting stations.

In this paper, we describe scheduling and synchronization support for a novel multimedia document, referred to as a session object. The session object captures a voice-annotated, interactive session with an application --- it contains audio and window streams. This paper addresses media scheduling and synchronization issues for the support of faithful replay of session objects when subject to timing variability at the replay workstation. The replay is supported by an adaptive scheduling algorithm. The algorithm preserves relative interstream synchronization between window and audio streams. Run-time temporal deformations are applied over the schedule of the window stream. We show that the inter-stream asynchrony floats under statistical control as a function of the scheduling interval. The mechanisms could be generalized to the replay of streams that are subject to timing variability. Our object-oriented toolkit, ReplayKit, enables an application to become replay-aware through access ...

Domain-independent, high-level, flexible, and efficient user interface development environments (UIDEs) are required for easing the task of developing multi-user interfaces. This paper describes several principles of designing such environments, gives motivation for supporting them, and identifies existing systems that follow them. Some of these principles are also well-accepted principles of designing single-user UIDEs---multi-user interaction provides new reasons for accepting them. Others are single-user principles that have been adapted for the multi-user case, while still others are new principles that have no counterparts in the single-user case. These principles address the functionality and architecture of a multi-user UIDE and the programs supported by it. Keyword Codes: C.2.4; D.2.2; H.1.2 Keywords: Distributed Systems; Tools and Techniques; User/Machine Systems 1. INTRODUCTION Most of the current principles of designing user interface development environments (UIDEs) [5, 15...

We have constructed a proof-of-principle system for supporting collaborative forensic medicine. The early prototype is built on ABC/DGS, a graph-server and collaborative hypermedia system built in the UNC Collaboratory. A second prototype is underway that has more flexible control of multi-person creation of, and access to, the shared patient data and pathology artifacts. Created with Dewan's Suite, this version maintains consistent yet different independent views of the underlying data, and moderates access through these views. We conclude by describing a planned third prototype, to be built not on ABC, but on a modification of the WWW httpd distributed data server.

The ability to undo operations is a standard feature in most single-user interactive applications. However, most current collaborative applications that allow several users to work simultaneously on a shared document lack undo capabilities; those which provide undo generally provide only a global undo, in which the last change made by anyone to a document is undone, rather than allowing users to individually reverse their own changes. In this paper, we propose a general framework for undoing actions in collaborative systems. The framework takes into account the possibility of conflicts between different users' actions that may prevent a normal undo. The framework also allows selection of actions to undo based on who performed them, where they occurred, or any other appropriate criterion. The proposed framework have been incorporated in DistEdit, a toolkit for building text group editors. Based on our experience, we discuss methods for maintaining the undo information in a groupware env...