NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/comp/humanfa/2678

Text File | 1992-11-19 | 63.2 KB | 1,157 lines

Newsgroups: comp.human-factors
Path: sparky!uunet!walter!hip10.bellcore.com!nielsen
From: Jakob Nielsen <[e-mail]>
Subject: List of accepted INTERCHI papers with abstracts (LONG)
Message-ID: <[e-mail]>
X-Xxmessage-Id: <[e-mail]>
X-Xxdate: Thu, 19 Nov 92 16:55:06 GMT
Sender: [e-mail]
Nntp-Posting-Host: hip10.bellcore.com
Reply-To: [e-mail]
Organization: Bellcore - Bell Communications Research
X-Useragent: Nuntius v1.1.1d12
References: <[e-mail]>
Date: Thu, 19 Nov 92 16:57:41 GMT
Lines: 1140

In article <[e-mail]> Jakob Nielsen,
[e-mail] writes:
>The following is a list of the 62 papers accepted for the INTERCHI
>conference out of the 330 submissions.
>
>The list below is the same as that posted earlier but now contains
abstracts.
>
>The conference will be in Amsterdam, the Netherlands, 24-29 April 1992.
>
>
>To request a copy of the advance program or if you have other questions,
>please contact the conference office:
>North American Office: [e-mail]
>fax [phone]
>European Office: [e-mail]
>fax: [phone]
>
>Whatever you do, *please* do not contact me. I am trying to get back to
>>
>Hope to see you in Amsterdam!
>Jakob Nielsen
>INTERCHI papers co-chair

This file contains the accepted INTERCHI papers arranged by session. For
each paper, the abstract from the submitted version us given. Note that the
authors may well change the abstract inthe final camera-ready copy which is
not due until January 1st.

Paper Session: Sharing Design Memory
   Session Chair: Michael Tauber, University of Paderborn
   Discussant: Thomas P. Moran, Xerox PARC

Preserving Knowledge in Design Projects: What Designers Need to Know
   James D. Herbsleb, University of Michigan
   Eiji Kuwana, NTT Software Laboratories
In order to inform the design of technology support and new procedural
methods for software design, we analyzed the content of real design
meetings in three organizations, focusing in particular on the questions
the designers ask of each other.  We found that most questions concerned
the project requirements, and focused on what the software or component was
supposed to do and how this functionality was to be realized.  Rationales
for design decisions were seldom asked about, but questions concerning
scenarios of use were very frequent. The implications of this research for
design tools and methods are discussed.

From Folklore to Living Design Memory
   Loren G. Terveen, AT&T Bell Laboratories
   Peter G. Selfridge, AT&T Bell Laboratories
   M. David Long, AT&T Bell Laboratories
We identify an important type of software design knowledge that we call
community specific folklore and show problems with current approaches to
managing it.  We built a tool that serves as a living design memory for a
large software development organization.  The tool delivers knowledge to
developers effectively and is embedded in organizational practice to ensure
that the knowledge it contains evolves as necessary.  This work illustrates
important lessons in building knowledge management systems, integrating
novel technology into organizational practice, and managing
research-development partnerships.

WHERE Did You Put It?  Issues in the Design and Use of a Group Memory
   Lucy M. Berlin, Hewlett-Packard Laboratories
   Robin Jeffries, Hewlett-Packard Laboratories
   Vicki L. O'Day, Hewlett-Packard Laboratories
   Andreas Paepcke, Hewlett-Packard Laboratories
   Cathleen Wharton, University of Colorado at Boulder
Collaborating teams of knowledge workers need a common repository in which
to share information gathered by individuals or developed by the team. This
is difficult to achieve in practice, because individual information access
strategies break down with group information --- people can generally find
things that are on their own messy desks and file systems, but not on other
people's.
The design challenge in a group memory is thus to enable low-effort
information sharing without reducing individuals' finding effectiveness. In
this paper we present the lessons from our design and initial use of a
hypertext-based group memory, TeamInfo.  We expose the serious cognitive
obstacles to a shared information structure, discuss the uses and benefits
we have experienced, address the effects of technology limitations, and
highlight some unexpected social and work impacts of our group memory.


Paper Session: Interacting in 3 Dimensions
   Session Chair: Angela Sasse, University College London
   Discussant: Rory Stuart, NYNEX Science & Technology, Inc.

Facile 3D Direct Manipulation
   Dan Venolia, Apple Computer, Inc.
An experimental 3D interface is described, including rendering acceleration
hardware, a 3D mouse, and 3D interaction techniques.  A 3D cursor,
controlled by the augmented mouse, allows direct manipulation of 3D
objects.  Objects are selected by placing the tip of the cursor inside.
Objects can be moved in 3D, or simultaneously moved and rotated using a
technique called ttail-dragging.h  A method called tsnap-toh helps users
align objects.  The interface is designed without using explicit modes or
commands.  Sounds accentuate the interaction.  Details of the
implementation and informal user observations are described, as well as
topics for future work.

Fishtank Virtual Reality
   Colin Ware, University of New Brunswick
   Kevin Arthur, University of British Columbia
   Kellogg S. Booth, University of British Columbia
The defining characteristics of what we call "Fishtank Virtual Reality",
are a stereo image of a three dimensional (3D) scene viewed on a monitor
using a perspective projection coupled to the head position of the
observer.  We discuss some of the relative merits of this mode of viewing
as compared to head mounted stereo displays.  We report the experimental
investigation of the following variables:  1) whether or not the
perspective view is coupled to the actual viewpoint of the observer, 2)
whether stereopsis is employed or not.  Experiment 1 involved the
subjective comparison of pairs of viewing conditions and the results
suggest that head coupling may be more important that stereo in yielding a
strong impression of three dimensionality.  Experiment 2  involved subjects
tracing a path from a leaf of a 3D tree to the correct root (there were two
trees intermeshed).  The error rates were dramatically reduced in the head
coupled stereo display condition.

A Space Based Model For User Interaction In Shared Synthetic Environments
   Lennart E. Fahltn, Swedish Institute of Computer Science
   Olov St.hl, Swedish Institute of Computer Science
   Charles Grant Brown, Swedish Institute of Computer Science
   Christer Carlsson, Swedish Institute of Computer Science
In a distributed shared synthetic environment with provisions for high
quality 3D visualization and interaction, it is possible to implement a
powerful variant of a rooms/space metaphor based on the concept of presence
or proximity between participants in 3D space. This kind of model can be
used for interfacing a user to the machine, give overview and control of
applications, file systems, networks and other computer resources and for
communication and collaboration with other users in the networked
environment. We model proximity with a simple description, i.e. the
geometric volume of the immediate surroundings, the aura, of the
participants' representation in the synthetic environment. This proximity,
or aura, is used to establish presence at meetings, to establish
communication channels and to provide interaction.


Paper Session: Understanding Programming
   Session Chair: Jean-Michel Hoc, CNRS - Universitt de Paris 8
   Discussant: Thomas Green, MRC Applied Psychology Unit

Do Algorithm Animations Assist Learning? An Empirical Study And Analysis
   John Stasko, Georgia Institute of Technology
   Albert Badre, Georgia Institute of Technology
   Clayton Lewis, University of Colorado at Boulder
Algorithm animations are dynamic graphical illustrations of computer
algorithms, and they are used as teaching aids to help explain how the
algorithms work.  Although many people believe that algorithm animations
are useful this way, no empirical evidence has ever been presented
supporting this belief.  We have conducted an empirical study of a priority
queue algorithm animation, and the study's results indicate that the
animation only slightly assisted student understanding.  In this article,
we analyze those results and hypothesize why algorithm animations may not
be as helpful as was initially hoped.  We also develop guidelines for
making algorithm animations more useful in the future.

Reducing the Variability of  Programmers' Performance  Through Explained
Examples
   David F. Redmiles, University of Colorado at Boulder
A software tool called EXPLAINER has been developed for helping programmers
perform new tasks by exploring previously worked-out examples. EXPLAINER is
based on cognitive principles of learning from examples and problem solving
by analogy. The interface is based on the principle of making examples
accessible through multiple presentation views and multiple representation
perspectives. Empirical evaluation showed that programmers using EXPLAINER
exhibited less variability in their performance compared to programmers
using a commercially available, searchable on-line manual. These results
are related to other studies of programmers and to current methodologies in
software engineering.

Mental Representations of Programs by Novices and Experts
   Vikki Fix, University of South Dakota
   Susan Wiedenbeck, University of Nebraska
   Jean Scholtz, Portland State University
This paper presents five abstract characteristics of the mental
representation of computer programs: hierarchical structure, explicit
mapping of code to goals, foundation on recognition of recurring patterns,
connection of knowledge, and grounding in the program text.  An experiment
is reported in which expert and novice programmers studied a Pascal program
for comprehension and then answered a series of questions about it designed
to show these characteristics if they existed in the mental representations
formed.  Evidence for all of the abstract characteristics was found in the
mental representations of expert programmers.  Novices' representations
generally lacked the characteristics, but there was evidence that they had
the beginnings, although poorly developed, of such characteristics.


Paper Session: Typing, Writing and Gesture
   Session Chair: Catherine G. Wolf, IBM T.J. Watson Research Center
   Discussant: Donald A. Norman, Apple Computer, Inc.

Touch-Typing with a Stylus
   David Goldberg, Xerox Corporation
   Cate Richardson, Xerox Corporation
One of the attractive features of keyboards is that they support novice as
well as expert users. Novice users enter text using "hunt-and-peck,"
experts use touch-typing. Although it takes time to learn touch-typing,
there is a large payoff in faster operation.
In contrast to keyboards, pen-based computers have only an novice mode for
text entry in which users print text to a character recognizer. An
electronic pen (or stylus) would be more attractive as an input device if
it supported expert users with some analogue of touch-typing.
We present the design and preliminary analysis of an approach to stylus
touch-typing using an alphabet of unistrokes, which are letters specially
designed to be used with a stylus. Unistrokes have the following advantages
over ordinary printing: they are faster to write, less prone to recognition
error, and can be entered in an "eyes-free" manner that requires very
little screen real estate.

Half-QWERTY: A One-handed Keyboard Facilitating Skill Transfer From QWERTY
   Edgar Matias, The Matias Corporation
   I. Scott MacKenzie, University of Guelph
   William Buxton, University of Toronto and Xerox PARC
Half-QWERTY is a new one-handed typing technique, designed to facilitate
the transfer of two-handed typing skill to the one-handed condition.  It is
performed on a standard keyboard, or a special half keyboard (with
full-sized keys).  In an experiment using touch typists, hunt-and-peck
typing speeds were surpassed after 3-4 hours of practice. Subjects reached
50% of their two-handed typing speed after about 8 hours.  After 10 hours,
all subjects typed between 43% and 76% of their two-handed speed, ranging
from 23.8 to 42.8 wpm. These results are important in providing access to
disabled users, and for the design of compact computers.  They also bring
into question previous research claiming finger actions of one hand map to
the other via spatial congruence rather than mirror image.

Incremental Recognition in Gesture-Based and Syntax-Directed Diagram Editors
   Rui Zhao, University of Paderborn
Diagram editing is an attractive application of  gestural  interfaces  and
pen-based computers which promise a new input paradigm where users
communicate with computers in  diagram  languages  by using  gestures.  A
key problem in building gesture-based diagram editors is the recognition of
 handsketched  diagrams.   Existing approaches concentrates either on
gesture recognition or on parsing visual languages, there has been a lack
of integrated  recog- nition  concepts.   This  paper presents novel
concepts and techniques based on an incremental paradigm  of  gesture
recognition and  a  cooperative  communication  between  modules  for
pattern recognition and for diagram parsing.  These  concepts  and
techniques  have been used successfully to build several experimental
gesture-based and syntax-directed diagram editors.


Paper Session: Evolving Design
   Session Chair: Eiji Kuwana, NTT Software Laboratories
   Discussant: Alistair Sutcliffe, City University, London

Integrating Theoreticians' And Practitioners' Perspectives With Design
Rationale
   Victoria Bellotti, Rank Xerox EuroPARC
QOC design rationale represents argumentation about design alternatives and
assessments. It can be used to generate design spaces which capture and
integrate information from design discussions and diverse kinds of
theoretical analyses. Such design spaces highlight how different
theoretical approaches can work together to help solve design problems.
This paper describes an example of the generation of a multi-disciplinary
QOC design space which shows how designers' deliberations can be augmented
with design contributions from a combination of different theoretical HCI
approaches.

Management Of Interface Design In Humanoid
   Ping Luo, University of Southern California
   Pedro Szekely, Information Sciences Institute, University of Southern
California
   Robert Neches, Information Sciences Institute, University of Southern
California
Today's interface design tools either force designers to handle a
tremendous number of design details, or limit their control over design
decisions. Neither of these approaches taps the true strengths of either
human designers or computers in the design process. This paper presents a
model based collaborative system that focuses human designers on decision
making, and utilizes the bookkeeping capabilities of computers for regular
and tedious tasks. We describe a tool that decomposes high-level design
goals into the necessary implementation steps, and helps designers manage
the myriad of details that arise during design.

The Evolution Of An Interface For Choreographers
   Tom W. Calvert, Simon Fraser University
   A. Bruderlin, Simon Fraser University
   S. Mah, Simon Fraser University
   T. Schiphorst, Simon Fraser University
   C. Welman, Simon Fraser University
This paper describes the evolution of the Life Forms interface, a
compositional tool for the creation of dance choreography, and highlights
some of the important lessons we have learned during a six year design and
implementation period.  The lessons learned can be grouped into two
categories: 1) Process, and  2) Architecture of the Interface.  Our goal in
developing a tool for choreography has been to provide computer-based
creative design support for the conception and development of dance.
Design evolution was driven by  feedback from the choreographers and users
who were members of the design team, combined with our knowledge of current
thinking of design and composition.  Although the interface evolved in a
relatively unconstrained way, the resulting system has many of the features
that theoretical discussion in Human Interface Design has projected as
necessary.  The Life Forms interface has evolved incrementally with one
major discontinuity where adoption of a new compositional primitive
required a completely new version.


Paper Session: Structuring Images for Interaction
   Session Chair: Wendy A. Kellogg, IBM T.J. Watson Research Center
   Discussant: Chris Schmandt, MIT Media Lab.

Human-Machine Perceptual Cooperation
   Francis K. H. Quek, University of Michigan
   Michael C. Petro, University of Michigan
The human-machine perceptual cooperation (HMPC) paradigm combines a human
operator's high level reasoning with machine perception to solve
spatio-perceptual intensive problems. HMPC defines two channels of
interaction: Focus of attention (FOA) by which the user focuses machine
perception, and context. As the user moves the FOA via a pointing device, a
smart cursor operates proactively on the data, highlighting objects which
satisfy the current context. The FOA permits foveal emphasis, enabling the
user to vary motor precision with image clutter. HMPC provides for contexts
at four levels of abstraction. This permits the efficiency of the system to
degrade gracefully as data quality worsens. We describe an HMPC application
where a human operator works with a machine to convert scanned raster maps
into vector format.

VideoMap And VideoSpaceIcon: Tools For Anatomizing Video Content
   Yoshinobu Tonomura, NTT Human Interface Laboratories
   Akihito Akutsu, NTT Human Interface Laboratories
   Kiyotaka Otsuji, NTT Human Interface Laboratories
   Toru Sadakata, NTT Human Interface Laboratories
A new approach to interacting with stored video is proposed. The approach
utilizes VideoMAP and VideoSpaceIcon. VideoMAP is the interface that shows
the essential video features in an easy to perceive manner.  VideoSpaceIcon
represnets the temporal and spatial characteristics of a video shot as an
intuitive icon.  A video indexing method supports both tools.
These tools allow the user's creativity to directly interact with the
essential features of each video by offering spatial and temporal clues.
This paper introduces the basic concept and describes prototype versions of
the tools as implemented in a video handling system.   VideoMAP and
VideoSpaceIcon are effective for video handling functions such as video
content analysis, video editing, and various video applications which need
an intuitive visual interface.

Automatic Structure Visualization For Video Editing
   Hirotada Ueda, Hitachi Ltd.
   Takafumi Miyatake, Hitachi Ltd.
   Shigeo Sumino, Hitachi Ltd.
   Akio Nagasaka, Hitachi Ltd.
We developed intelligent functions for the automatic description of the
video structure, and the visualization methods for the temporal-spatial
video structures obtained by these functions as well as the functions.  The
functions offer such descriptions as cut separations, motion of the camera
and filmed objects, tracks and contour lines of objects, existence of
objects, and periods of existence.  Further, identical objects are
automatically linked.  Thus the visualization methods supported by object
links allow users to directly manipulate the structure itself, descriptions
and raw video data without paying too much attention to the levels within
the hierarchy of video information.


Paper Session: Skill Development
   Session Chair: Yvonne Waern, Linkkping University
   Discussant: Jean McKendree, NYNEX Science and Technology

The Growth of Software Skill:  A Longitudinal Look at Learning & Performance
   Erik Nilsen, Lewis and Clark College
   HeeSen Jong, National University of Singapore
   Judith S. Olson, University of Michigan
   Kevin Biolsi, University of Michigan
   Henry Rueter, University of Michigan
   Sharon Mutter, Catholic University
This research follows a group of users over time (16 months) as they
progress from novice towards expert  in their use of Lotus 1-2-3.
Quantitative and qualitative measures of performance are compared with
expert users having over three years of experience.  The results indicate
that the motoric aspect of performance is relatively stable over time,
while improvement in the cognitive component of the skill is dependent on
details of the interaction dialogue. This work  fills a gap between the
work on skill learning and skilled performance.   Implications for the
Keystroke Level Model of skilled performance are discussed as well as
suggestions for  designing the interaction dialogue to facilitate the
development of expertise

Embedding Computer-Based Critics in the Contexts of Design
   Gerhard Fischer, University of Colorado at Boulder
   Kumiyo Nakakoji, University of Colorado at Boulder
   Jonathan Ostwald, University of Colorado at Boulder
   Gerry Stahl, University of Colorado at Boulder
   Tamara Sumner, University of Colorado at Boulder
Computational critiquing mechanisms provide an effective form of
computer-human interaction supporting the process of design. Critics
embedded in domain-oriented design environments can take advantage of
additional knowledge residing in these environments to provide less
intrusive, more relevant critiques. Three classes of embedded critics have
been designed, implemented, and studied: Generic critics use domain
knowledge to detect problematic situations in the design construction.
Specific critics take advantage of additional knowledge in the partial
specification to detect inconsistencies between the design construction and
the design specification. Interpretive critics are tied to perspective
mechanisms that support designers in examining their artifact from
different viewpoints.

How To Aid Non-Experts
   Mark Neerincx, University of Amsterdam
   Paul de Greef, University of Amsterdam
Knowledge is lacking about whether, or, under which conditions a help
system is effective. This paper presents two experiments. The first
evaluates a commercially available system for statistical analysis, and the
second evaluates a statistical system developed with a methodology for
designing aiding systems. We have identified two crucial conditions for
effective aiding: (i) the system should present the knowledge the user is
lacking, which has the consequence that the help should be based on an
expert-model of the users' task, and (ii) the costs of the communication
with the help facilities should be minimal.


Paper Session: Voices and Faces
   Session Chair: Steven Feiner, Columbia University
   Discussant: S. Joy Mountford, Apple Computer, Inc.

A Design Space For Multimodal Systems: Concurrent Processing and Data Fusion
   Laurence Nigay, Laboratoire de Gtnie Informatique
   JoFlle Coutaz, Laboratoire de Gtnie Informatique
Multimodal interaction enables the user to employ different modalities such
as voice, gesture and typing for communicating with a computer. This paper
presents an analysis of the integration of multiple communication
modalities within an interactive system. To do so, a software engineering
perspective is adopted. First, the notion of "multimodal system" is
clarified. Based on a precise definition of multimodality, we then propose
a design space and a method for classifying multimodal systems. In the last
section, we present a software architecture model that supports the most
salient properties of such systems: concurrent processing and data fusion.
Two multimodal systems developed in our team, VoicePaint and NoteBook, are
used to illustrate the discussion.

VoiceNotes: A Speech Interface for a Hand-held Voice Notetaker
   Lisa J. Stifelman, Massachusetts Institute of Technology
   Barry Arons, Massachusetts Institute of Technology
   Chris Schmandt, Massachusetts Institute of Technology
   Eric A. Hulteen, Apple Computer, Inc.
VoiceNotes is an application for a voice-controlled hand-held computer that
allows the creation, management, and retrieval of user-authored voice
notes---small segments of digitized speech containing ideas, reminders, or
things to do.  Iterative design and user testing helped to refine the
initial user interface design.  VoiceNotes explores the problem of
capturing and retrieving spontaneous ideas, the use of speech as data, and
the use of speech input and output in the user interface for a hand-held
computer.  In addition, VoiceNotes serves as a step toward new uses of
voice technology and interfaces for future portable devices.

Communicative Facial Displays As A New Conversational Modality
   Akikazu Takeuchi, Sony Computer Science Laboratory Inc.
   Katashi Nagao, Sony Computer Science Laboratory Inc.
A face is an independent communication channel that conveys emotional and
conversational signal encoded by some facial displays.  Facial displays can
be viewed as communicative signals coordinating conversation.  We attempt
to bring facial displays into computer human interaction as a new modality
that makes the interaction tighter and more efficient while less cognitive
load.  As the first step, a speech dialogue system was selected for
investigating the power of communicative facial displays.  The conversation
between a user and the speech dialogue system with facial displays was
analyzed, and it is shown that conversation with the system with facial
displays was more successful than that without facial displays.


Paper Session: Usability Assessment Methods
   Session Chair: Brian Shackel, Loughborough University of Technology
   Discussant: Clare-Marie Karat, IBM

Iterative Methodology and Designer Training in Human-Computer Interface
Design
   George Bailey, The Church of Jesus Christ of Latter-day Saints
A promising method for user interface design is the iterative design
methodology. Little empirical support for this method has appeared in the
literature.  One difficulty with user interface design is knowing the set
of skills the user interface designer must possess.  Two common groups
involved in user interface design are human factors specialists and
programmers.
The results of this study indicate that iterative design can improve the
usability of a product.  The amount of the improvement may be constrained
by the original design. This study supports the use of human factors
specialists.  A significant difference between designs produced by human
factors specialist and programmers was found.

A Mathematical Model of the Finding of Usability Problems
   Jakob Nielsen, Bellcore
   Thomas K. Landauer, Bellcore
For 11 studies, we find that the detection of usability problems as a
function of number of users tested or heuristic evaluators employed is well
modeled as a Poisson process. The model can be used to plan the amount of
evaluation required to achieve desired levels of thoroughness or benefits.
Results of early tests can provide estimates of the number of problems left
to be found and the number of additional evaluations needed to find a given
fraction. With quantitative evaluation costs and detection values, the
model can estimate the numbers of evaluations at which optimal cost/benefit
ratios are obtained and at which marginal utility vanishes. For a tmediumh
example, we estimate that 16 evaluations would be worth their cost, with
maximum benefit/cost ratio at four.

Estimating the Relative Usability of Two Interfaces: Heuristic, Formal, and
Empirical Methods Compared
   Jakob Nielsen, Bellcore
   Victoria L. Phillips, Bellcore
Two alternative user interface designs were subjected to user testing to
measure user performance in a database query task. User performance was
also estimated heuristically in three different ways and by use of formal
GOMS modelling. The estimated values for absolute user performance had very
high variability, but estimates of the relative advantage of the fastest
interface were less variable. Choosing the fastest of the two designs would
have a net present value more than 1,000 times the cost of getting the
estimates. A software manager would be expected to make the correct choice
100% of the time if decisions were based on at least three independent
estimates. User testing was 4.9 times as expensive as the cheapest
heuristic method but provided better performance estimates.


Paper Session: Auditory Interfaces
   Session Chair: Alex Rudnicky, Carnegie Mellon University
   Discussant: Floris L. van Nes, Phillips Research/IPO

An Evaluation of Earcons for Use in Auditory Human-Computer Interfaces
   Stephen A. Brewster, University of York
   Peter C. Wright, University of York
   Alistair D. N. Edwards, University of York
An evaluation of earcons was carried out to see whether they are an
effective means of communicating information in sound. An initial
experiment showed that earcons were better than unstructured bursts of
sound and that musical timbres were more effective than simple tones. A
second experiment was then carried out which improved upon some of the
weaknesses shown up in Experiment 1 to give a significant improvement in
recognition. From the results of these experiments some guidelines were
drawn up for use in the creation of earcons. Earcons have been shown to be
an effective method for communicating information in a human-computer
interface.

Synthesizing Auditory Icons
   William W. Gaver, Rank Xerox Cambridge EuroPARC
Auditory icons add valuable functionality to computer interfaces,
particularly when they are parameterized to convey dimensional information.
 They are difficult to create and manipulate, however, because they usually
rely on digital sampling techniques.  This paper suggests that new
synthesis algorithms, controlled along dimensions of events rath er than
those of the sounds themselves, may solve this problem.  Several
algorithms, developed from research on auditory event perception, are
described in enough detail here to permit their implementation.  They
produce a variety of impact, bouncing, breaking, scraping, and machine
sounds.  By controlling them with attributes of relevant computer events, a
wide range of parameterized auditory icons may be created.

Computer Aided Conversation For Severely Physically Impaired Non-Speaking
People
   Norman Alm, University of Dundee
   John Todman, University of Dundee
   Leona Elder, Dundee Institute of Technology
   A. F. Newell, University of Dundee
This paper reports the development of a computer-aided conversation
prosthesis which is designed for severely physically impaired non-speaking
people.  The research methodology was to model aspects of conversational
structure derived from the field of conversation analysis within a
prototype conversational prosthesis. The prototype was evaluated in
empirical investigations which also suggested successful strategies for
carrying out satisfying conversation using such a system. Two versions have
been built and tested, one using an able-bodied operator to test the
feasibility of creating conversation from prestored material, the second
being used by a physically impaired non-speaking operator. The prototype
demonstrated the advantages of this interface design in helping the user to
carry out natural sounding and satisfying conversations.


Paper Session: Conceptual Analysis of Users and Activity
   Session Chair:  Susanne Bodker, Aarhus University
   Discussant: Victor Kaptelinin, Institute of General and Educational
Psychology

Exploding the Interface: Experiences of a CSCW Network
   John Bowers, Manchester University
   Tom Rodden, Lancaster University
The development of human computer interaction has been dominated by the
interface both as a design concept and as an artifact of computer systems.
However, recently researchers have been re-examining the role of the
interface in the user's interaction with the computer. This paper further
examines the notion of the interface in light of the experiences of the
authors in establishing a network to support cooperative work. The authors
argue that the concept of the single interface which provides a focus for
interaction with the computer system is no longer tenable and that richer
conceptions of the inter-relationships between users and computer systems
are needed.

Searching For Unity Among Diversity: Exploring The Interface Concept
   Kari Kuutti, University of Oulu
   Liam J. Bannon, University of Limerick
HCI is seen by many as focusing on the design of interfaces to computer
systems, yet exactly  what is implied by this focus on "interfaces" is
unclear. In this paper we show how a better understanding of what is meant
by the interface is possible via the concept of  abstraction  levels.  We
show how this levels approach can clarify some ambiguities, and also how it
can be related to different phases in the evolution of the human-computer
interaction field itself. In this context, we are able to account for the
recent interest in activity theory as a possible alternative framework for
HCI work, while stressing the need for HCI research and design to consider
each of the separate, but related, levels.

The Cost Structure of Sensemaking
   Daniel M. Russell, Xerox PARC
   Mark J. Stefik, Xerox PARC
   Peter Pirolli, Xerox PARC
   Stuart K. Card, Xerox PARC
Making sense of a body of data is a common activity in any kind of
analysis.  Sensemaking is the process of searching for a representation and
encoding data in that representation to answer task-specific questions.
Different operations during sensemaking require different cognitive and
external resources.  Representations are chosen and changed to reduce the
cost of operations in an information processing task.  The power of these
representational shifts is generally under-appreciated as is the relation
between sensemaking and information retrieval.
We analyze sensemaking tasks and develop a model of the cost structure of
sensemaking.  We discuss implications for the integrated design of user
interfaces, representational tools, and information retrieval systems.

Paper Session: Demonstration Based Systems
   Session Chair: Allen Cypher, Apple Computer, Inc.
   Discussant: Dan R. Olsen, Brigham Young University

Prototyping an Intelligent Agent through Wizard of Oz
   David Maulsby, University of Calgary
   Saul Greenberg, University of Calgary
   Richard Mander, Apple Computer, Inc. and Stanford University
Turvy is a simulated prototype of an instructible agent. The user teaches
it by demonstrating actions and pointing at or talking about relevant data.
We formalized our assumptions about what could be implemented, then used
the Wizard of Oz to flesh out a design and observe users' reactions as they
taught several editing tasks. We found: a) all users invent a similar set
of commands to teach the agent; b) users learn the agent's language by
copying its speech; c) users teach simple tasks with ease and complex ones
with reasonable effort; and d) agents cannot expect users to point to or
identify critical features without prompting.
In conducting this rather complex simulation, we learned some lessons about
using the Wizard of Oz to prototype intelligent agents: a) design of the
simulation benefits greatly from prior implementation experience; b) the
agent's behavior and dialog capabilities must be based on formal models; c)
studies of verbal discourse lead directly to an implementable system; d)
the designer benefits greatly by becoming the Wizard; e) qualitative data
is more valuable for answering global concerns, while quantitative data
validates accounts and answers fine-grained questions.

A Synergistic Approach to Specifying Simple Number Independent Layouts by
Example
   Scott E. Hudson, Georgia Institute of Technology
   Chen-Ning Hsi, Georgia Institute of Technology
A grid-based technique to specify simple number independent layouts by
example is described.  This technique was originally developed to support
layout specification for a parallel program visualization system but can be
applied to aid other simple graphical layout tasks as well.  The technique
works by allowing the user to construct an example layout using a
grid-based interaction technique.  This example can then be generalized
into a layout algorithm which can be applied to create layouts of any size.
 However, rather than simply choosing the "best" generalization, the system
decribed here takes a synergistic approach.  New examples from a set of
alternative generalizations are presented to the user so that they can
guide and control the generalization process.  This provides more
understanding and control of the generalization process and typically
allows a correct generalization to be constructed from only one small
example.

Marquise: Creating Complete User Interfaces by Demonstration
   Brad A. Myers, Carnegie Mellon University
   Richard G. McDaniel, Carnegie Mellon University
   David S. Kosbie, Carnegie Mellon University
Marquise is a new interactive tool that allows virtually all of the user
interfaces of graphical editors to be created by demonstration without
programming. A graphical editor is a program where the user creates and
manipulates graphical objects with a mouse. This is a very large class of
programs and includes drawing programs like MacDraw, graph layout editors
like MacProject, visual language editors, and many CAD/CAM programs. The
primary innovation in Marquise is that it allows the designer to
demonstrate the overall behavior of the interface. To implement this, the
Marquise framework contains built-in knowledge about palettes for creating
and specifying properties of objects, selecting, moving, deleting, and
other options. This allows most of the end user's actions to be
demonstrated without programming, which means that Marquise can be used by
non-Programmers. Furthermore, a graphical editor can be demonstrated with
Marquise in about 1/80th the time it would take to implement it by
programming using a conventional toolkit.


Paper Session: Collecting User-Information for System Design
   Session Chair: Rolf Molich, Baltica Insurance
   Discussant: Alison Lee, NYNEX

Exploring the Applications of User-Expertise Assessment for Intelligent
Interfaces
   Michel C. Desmarais, Centre de researche informatique de Montrtal
   Jiming Liu, Centre de researche informatique de Montrtal
We describe a technique for automatically constructing a fine-grain model
of a user's knowledge-state based on a small number of observations.  The
user knowledge state is specified as a set of weights on each knowledge
unit that indicates a likelihood of mastery.  These weights are updated
every time a knowledge unit is reassigned a new weight (by a question and
answer process).  The updating method is based the Dempster-Shafer
algorithm with an empiridally derived network of knowledge units.  A User
Knowledge Assessment Tool (UKAT) has been implemented that employs this
model.  Through simulations, we explore an entropy-based method of choosing
the questions and compare the results with a random sampling method.
Results show that the proposed technique is useful and efficient in
inferring a detailed model of an individual's knowledge, but the
entropy-based method can induce a bias in some conditions.

Planning For Multiple Task Work - An Analysis Of A Medical Reception
Worksystem
   Becky Hill, University College London
   John Long, University College London
   Walter Smith, University College London
   Andy Whitefield, University College London
This paper presents an investigation of interactive worksystem planning in
the multiple task work domain of medical reception. In an observational
study of a medical reception worksystem, three different types of plan were
identified: the task plan, the procedure plan and the activity plan. These
three types of plan were required for effective working in the domain of
medical reception, because of the nature of the multiple task work. It is
proposed, therefore, that to design effective interactive human-computer
worksystems for the domain of medical reception (and possibly other work
domains of a similar nature), the designer must specify the three different
types of plan and the relationships between them.

The Diary Study: A Workplace-Oriented Research Tool to Guide Laboratory
Efforts
   John Rieman, University of Colorado at Boulder
Methods for studying user behavior in HCI can be informally divided into
two approaches: experimental psychology in the laboratory and observations
in the workplace. The first approach often yields results that have little
effect on system usability, while the second may deliver essentially
anecdotal data that do not support general conclusions. This paper
describes how two similar approaches in another field, animal behavior,
produce complementary results. In support of a similar research
interactions in HCI, the paper suggests the diary study technique, a tool
for objective data collection in the workplace that can help define
laboratory experiments. A diary study is described that focuses on
exploratory learning.


Paper Session: Video Support for Workplace Collaboration
   Session Chair: Hiroshi Ishii, NTT Human Interface Laboratories
   Discussant: Mike Robinson, Aarhus University

Turning Away from Talking Heads: The Use of Video-as-Data in Neurosurgery
   Bonnie A. Nardi, Hewlett-Packard Laboratories
   Heinrich Schwarz, Hewlett-Packard Laboratories
   Allan Kuchinsky, Hewlett-Packard Laboratories
   Robert Leichner, Hewlett-Packard Laboratories
   Steve Whittaker, Hewlett-Packard Laboratories
   Robert Sclabassi, University of Pittsburgh
Studies of video as a support for collaborative work have provided little
hard evidence of its utility for either task performance or fostering
telepresence, i.e. the conveyance of a face-to-face like social presence
for remotely located participants.  To date, most research on the value of
video has concentrated on "talking heads" video in which the video images
are of remote paticipants conferring or performing some task together.  In
contrast to talking heads video, we studied video-as-data in which video
images of the workspace and work objects are the focus of interest, and
convey critical information about the work.  The use of video-as-data is
intended to enhance task performance, rather than to provide telepresence.
We studied the use of videio during neurosurgery within the operating room
and at remote locations away from the operating room.  The workspace shown
in the video is the surgical field (brain or spine) that the surgeon is
operatifntg on.  We discuss our findings on the use of live and recorded
video, and suggest extensions to video-as-data including its integration
with computerized time-based information sources to educate and co-ordinate
complex actions among distributed workgroups.

One is Not Enough: Multiple Views in a Media Space
   William Gaver, Rank Xerox Cambridge EuroPARC
   Abigail Sellen, Rank Xerox Cambridge EuroPARC
   Christian Heath, Rank Xerox Cambridge EuroPARC
   Paul Luff, Rank Xerox Cambridge EuroPARC
Media spaces support collaboration, but the limited access they provide to
remote colleagues' activities can undermine their utility.  To address this
limitation, we built an experimental system in which four switchable
cameras were deployed in each of two remote offices, and observed
participants using the system to collaborate on two tasks.  The new views
allowed increased access to task-related artifacts; indeed, users preferred
these views to more typical "face-to-face" ones. However, problems of
establishing a joint frame of reference were exacerbated by the additional
complexity, leading us to speculate about more effective ways to expand
access to remote sites.


Paper Session: Perspectives and Illusions
   Session Chair: Gillian Crampton Smith, Royal College of Art

How Fluent is Your Interface? Designing for International Users
   Patricia Russo, Sunsoft
   Stephen Boor, The Boston Company
To successfully build bridges between worlds, software developers must
increase their awareness of cross-cultural differences, and make changes to
the traditional software development process. Creating fluent interfaces
for international markets goes beyond translating text and date, time, and
number formats. This paper presents and explains a cross-cultural checklist
of issues including text, local formats, images, symbols, colors, flow, and
advertising and marketing. Suggestions for an effective international
product development cycle are provided. The suggested development cycle
incorporates international design feedback and usability testing before the
initial product is released.

Representation in Virtual Space: Visual Convention in the Graphical User
Interface
   Loretta Staples, Loretta Staples Design
The graphical user interface (GUI) typically provides a multi-windowed
environment within a flat workspace or "desktop."  Simultaneously, however,
controls for executing commands within this interface are increasingly
being rendered to produce three-dimensional effects.  This paper explores
ways in which the space of the GUI desktop might be literally and
figuratively deepened through the incorporation of visual devices that have
emerged during the history of art-specifically perspective and light
effects.  By enriching the visual vocabulary of the GUI, greater semantic
complexity becomes sustainable.

Principles, Techniques, and Ethics of Stage Magic and Their Potential
Application to Human Interface Design
   Bruce Tognazzini, Sunsoft
Magicians have been designing and presenting illusions for 5000 years. They
have developed principles, techniques and ethical positions for their craft
that this paper argues are potentially applicable to the design of
human/computer interfaces.  The author presents a number of specific
examples from magic and discusses their counterparts in human interface
design, in hopes that human interface practitioners and researchers will,
having recognized the potential applicability of magic, go further on their
own to explore its domain.


Paper Session: Model-Based UI Development Systems
   Session Chair: Charles van der Mast, Delft University of Technology
   Discussant: Roger Took, University of York

Separations of Concerns in the Chiron-1 User Interface Development and
Management System
   Richard N. Taylor, University of California, Irvine
   Gregory F. Johnson, Northrop Corporation
The development of user interfaces for large applications is subject to a
series of well-known problems including cost, maintainability, and
sensitivity to changes.  The Chiron UIMS has been built to address these
software engineering concerns.  Chiron introduces a series of layers that
insulate components of an application from changes in other components.  To
separate application code from user interface code, artists are attached to
application abstract data types (ADTs). Operations on ADTs within the
application implicitly trigger artist activities.  Chiron also provides
insulation between the user interface layer and the underlying windowing
systems and toolkits.  Concurrency is pervasive in the Chiron architecture;
there is no requirement for a user interface listening/dispatching routine
to have exclusive control.

A Second Generation User Interface Design Environment:  The Model And The
Runtime Architecture
   Piyawadee "Noi" Sukaviriya, Georgia Institute of Technology
   James D. Foley, Georgia Institute of Technology
   Todd Griffith, Georgia Institute of Technology
Several obstacles exist in the user interface design process which distract
a developer from designing a good user interface.  One of the problems is
the lack of an application model to keep the designer in perspective with
the application.  The other problem is having to deal with massive user
interface programming to achieve a desired interface and to provide users
with correct help information on the interface.  In this paper, we discuss
an application model which captures information about an application at a
high level, and maintains mappings from the application to a desired
interface.  The application model is then used to control the dialogues at
runtime and can be used by a help component to automatically generate
animated and textual help.  Specification changes in the application model
will automatically result in behavioral changes in the interface.

Beyond Interface Builders: Model-based Interface Tools
   Pedro Szekely, Information Sciences Institute, University of Southern
California
   Ping Luo, Information Sciences Institute, University of Southern
California
   Robert Neches, Information Sciences Institute, University of Southern
California
Interface builders only support the construction of the menus and dialogue
boxes of an application.  They do not support the construction of
interfaces of many application classes (visualization, simulation, command
and control, domain-specific editors) because of the dynamic and complex
information that these applications process.  HUMANOID is a model-based
interface design and construction tool where interfaces are specified by
building a declarative description (model) of their presentation and
behavior.  HUMANOID's modeling language provides simple abstraction,
iteration and conditional constructs to model the interface features of
these application classes.  HUMANOID provides an easy-to- use designer's
interface that lets designers build complex interfaces without programming.


Paper Session: Meetings and Collaborative Writing
   Session Chair: Gary M. Olson, University of Michigan
   Discussant: Norbert A. Streitz, GMD Integrated Publication and
Information Systems Institute

Tivoli: An Electronic Whiteboard for Informal Workgroup Meetings
   Elin Ronby Pedersen, Roskilde University
   Kim McCall, Xerox PARC
   Thomas P. Moran, Xerox PARC
   Frank G. Halasz, Xerox PARC
This paper describes Tivoli, an electronic whiteboard application designed
to support informal workgroup meetings and targeted for the Xerox
Liveboard. Tivoli provides its users with the simplicity, facile use, and
obvious functionality of conventional whiteboards, while taking advantage
of its computational power to support the users' meeting practices. The
paper presents the motivations for Tivoli's design and describes the
current version in operation. It then reflects on several issues
encountered in designing Tivoli, including the need to reconsider the
desktop GUI, the use of strokes as fundamental objects, the generalized
wipe interface technique, and the use of meta-strokes as commands.

The User-centred Iterative Design Of Collaborative Writing Software
   Ronald M. Baecker, University of Toronto
   Dimitrios Nastos, University of Toronto
   Ilona R. Posner, University of Toronto
   Kelly L. Mawby, IBM Canada Ltd.
This paper presents the user-centred iterative design of software that
supports collaborative writing. The design grew out of a study of how
people write together that included a survey of writers and an experiment
studying writing teams linked by a variety of communications media. The
resulting taxonomy of collaborative writing is summar- ized in the paper,
followed by a list of design requirements for collaborative writing
software suggested by the work. The paper describes two designs of the
software. The first prototype supports synchronous writing and editing from
workstations linked over local area and wide area networks. The second
prototype also supports brainstorming, outlining, and document review, as
well as asynchronous work. Lessons learned from the user testing and actual
usage of the two systems are also presented.

Exploiting Version Support In Cooperative Systems
   Anja Haake, GMD-IPSI
   Jkrg M. Haake, GMD-IPSI
Current CSCW applications support one or more modes of cooperative work.
The selection of and transition between these modes is usually placed on
the users. At IPSI we built the SEPIA cooperative hypermedia authoring
environment supporting a complete set of situations arising during
collaborative work and the smooth transitions between them. While early use
of the system shows the benefits of supporting smooth transitions between
different collaborative modes, it also reveals some deficits regarding
parallel work, management of alternative documents, or reuse of document
parts. We propose to integrate version support to overcome these
limitations. This leads to a versioned data management and an extended
user-interface enabling concurrent users to select a certain state of their
work, to be aware of related changes, and to cooperate with others either
asynchronously or synchronously.


Paper Session: Automated UI Generation
   Session Chair: Ernest Edmonds, Loughborough University of Technology
   Discussant: JoFlle Coutaz, Laboratoire de Gtnie Informatique-IMAG

Generating User Interfaces From Data Models And Dialogue Net Specifications
   Christian Janssen, Fraunhofer-Institut f r Arbeitswirtschaft und
Organisation
   Anette Weisbecker, Fraunhofer-Institut f r Arbeitswirtschaft und
Organisation
   J rgen Ziegler, Fraunhofer-Institut f r Arbeitswirtschaft und
Organisation
A method and a set of supporting tools have been developed for an improved
integration of user interface design with software engineering methods and
tools. Animated user interfaces for database-oriented applications are
generated from an extended data model and a new graphical technique for
specifying dialogues. Based on views defined for the data model, an expert
system uses explicit design rules derived from existing guidelines for
producing the static layout of the user interface. A petri net based
technique called dialogue nets is used for specifying the dynamic
behaviour. Output is generated for an existing user interface management
system. The approach supports rapid prototyping while using the advantages
of standard software engineering methods.

Encapsulating Knowledge For Intelligent Automatic Interaction Objects
Selection
   Jean M. Vanderdonckt, Facultts Universitaires Notre-Dame de la Paix,
Institut d'Informatique
   Franuois Bodart, Facultts Universitaires Notre-Dame de la Paix, Institut
d'Informatique
TRIDENT is a set of interactive tools that automatically generates user
interface for highly-interactive business-oriented applications. It
encompasses a component for intelligent interaction objects selection based
on three differents concepts. First, an object oriented typology classifies
abstract interaction objects to allow a presentation independent selection.
Second, guidelines are translated to automatic rules to select abstract
interaction objects from both an application data model and a dialog model.
Third, these guidelines are encapsulated in a decision tree technique to
make the reasoning obvious to the user. This approach guarantees a target
environment independent user interface. Once the target environment
specified, abstract interaction objects are mapped into concrete
interaction objects to produce the observable interface.

Providing High-level Control and Expert Assistance in the User Interface
Presentation Design
   Won Chul Kim, The George Washington University
   James D. Foley, Georgia Institute of Technology
The capability of the user interface design tools can be enhanced by
providing high-level design control throughout the user interface
presentation design process. The interface builders that provide parts to
put together user interfaces are too low-level. We have developed a
framework that unites the essential knowledge components needed for
effective user interface presentation design. To demonstrate its viability
we have constructed a tool based on encapsulated design knowledge that
establishes high-level style preferences and provides expert assistance for
the dialog box presentation design and menu structuring.


Paper Session: Searching: Tools and Strategies
   Session Chair: George W. Furnas, Bellcore
   Discussant: Gabriele Rohr, IBM Entwicklungslabor Bkblingen

Orienteering in an Information Landscape: How Information Seekers Get From
Here to There
   Vicki L. O'Day, Hewlett-Packard Laboratories
   Robin Jeffries, Hewlett-Packard Laboratories
We studied the uses of information search results by regular clients of
professional intermediaries.  The clients in our study engaged in three
different types of searches: (1) monitoring a well-known topic or set of
variables over time, (2) following an information-gathering plan suggested
by a typical approach to the task at hand, and (3) exploring a topic in an
undirected fashion.  In most cases, a single search evolved into a series
of interconnected searches, usually beginning with a high-level overview.
We identified a set of common triggers and stop conditions for further
search steps.  We also observed a set of common operations that clients
used to analyze search results.  In some settings, the number of search
iterations was reduced by restructuring the work done by intermediaries. We
discuss the implications of the interconnected search pattern, triggers and
stop conditions, common analysis techniques, and intermediary roles for the
design of information access systems.

Using Icons to Find Documents: Simplicity Is Critical
   Michael D. Byrne, Georgia Institute of Technology
A common task at almost any computer interface is that of searching for
documents, which GUIs typically represent with icons.  Oddly, little
research has been done on the processes underlying icon search.  This paper
outlines the factors involved in icon search and proposes a model of the
process. An experiment was conducted which suggests that the proposed model
is sound, and that the most important factor in searching for files is the
type of icons used.  In general, simple icons (those discriminable based on
a few features) seem to help users, while complex icons are no better than
simple rectangles.

Queries-R-Links: Graphical Markup for Text Navigation
   Gene Golovchinsky, University of Toronto
   Mark Chignell, University of Toronto
In this paper we introduce a style of interaction (interactive querying)
that combines features of hypertext with Boolean querying, using direct
markup of text to launch queries. We describe two experiments that compare
the relative ease of expressing Boolean queries as text versus a graphical
equivalent. The results of these experiments show that the expression of
queries in the graphical format is significantly easier than the textual
equivalent. We then describe the Queries-R-Links system that we have
developed at the University of Toronto. Queries-R-Links uses the graphical
markup method to launch Boolean queries interactively using direct markup
of text. This work represents significant progress towards information
exploration systems that combine the useful features of information
retrieval querying and hypertext browsing.


Paper Session: Hands, Menus and Dr. Fitts
   Session Chair: Bonnie E. John, Carnegie Mellon University
   Discussant: Robert J.K. Jacob, Naval Research Laboratory

Human Performance Using Computer Input Devices in the Preferred and
Non-Preferred Hands
   Paul Kabbash, University of Toronto
   I. Scott MacKenzie, University of Guelph
   William Buxton, University of Toronto
Subjects' performance was compared in pointing and dragging tasks using the
preferred and non-preferred hands.  Tasks were tested using three different
input devices:  a mouse, a trackball and a tablet-with stylus .  The
trackball had the least degradation across hands in performing the tasks,
however it remained inferior to both the mouse and the stylus.  For small
distances and small targets, the preferred hand was superior.  However, for
larger targets and larger distances, both hands performed about the same.
The experiment shows that the non-preferred is more than a poor
approximation of the preferred hand.  The hands are complimentary, each
having its own strength and weakness.  One design implication is that the
non-preferred hand is well suited for tasks that do not require precise
action, such as scrolling.

The Limits of Expert Performance Using Hierarchic Marking Menus
   Gordon Kurtenbach, University of Toronto
   William Buxton, University of Toronto
A marking menu allows a user to perform a menu selection by either
popping-up a radial (or pie) menu, or by making a straight mark in the
direction of desired menu item without popping-up the menu.  A hierarchic
marking menu uses hierarchic radial menus and "zig-zag" marks to select
from the hierarchy.  This paper experimentally investigates the  bounds on
how many items can be in each level and how deep the hierarchy can be
before using a marking to select an item becomes too slow or prone to
errors.

Lag As A Determinant Of Human Performance In Interactive Systems
   I. Scott MacKenzie, University of Guelph
   Colin Ware, University of New Brunswick
The sources of lag (the delay between input action and output response) and
its effects on human performance are discussed.  We measured the effects in
a study of target acquisition using the classic Fitts' law paradigm with
the addition of four lag conditions.  At the highest lag tested (225 ms),
movement times and error rates increased by 64% and 214% respectively,
compared to the zero lag condition.  We propose a model, according to which
lag should have a multiplicative effect on Fitts' index of difficulty.  The
model accounts for 94% of the variance and is better than alternative
models which propose only an additive effect for lag.  The implications for
the design of virtual reality systems are discussed.


Paper Session: Finding and Keeping Stuff
   Session Chair: Nick Hammond, University of York
   Discussant: Andrew Dillon, HUSAT Research Centre

Computer Image Retrieval by Features:  Suspect Identification
   Eric S. Lee, Saint Mary's University
   Thom Whalen, Communications Research Center, Ottawa
Correct suspect identification of known offenders by witnesses deteriorates
rapidly as more are examined in mugshot albums.  Feature approaches, where
mugshots are displayed in order of similaity to witnesses' descriptions,
attempt to increase identification success by reducing this number.  b
methodology is proposed for system design and evaluation based on
experiments, computer simulations, and four classes of system performance
measures:  identification performance, retrieval rank, tolerance
performance, and feature quality.  This was used to develop a system for
640  mugshots of known offenders.  In three empirical tests, over 90% of
witness searches resulted in suspects retrieved in the first eight mugshots.

Empirically-based Re-design Of A Hypertext Encyclopedia
   Keith Instone, Bowling Green State University
   Barbee Mynatt Teasley, Bowling Green State University
   Laura Marie Leventhal, Bowling Green State University
This paper reports on the processes used and guidelines discovered in
re-designing the user interface of the hypertext encyclopedia, HyperHolmes.
 The re-design was based on the outcomes of a previous experiment and was
evaluated experimentally. Results showed that the new system resulted in
superior performance and somewhat different styles of navigation compared
to the old system and to paper. The study provides empirical support for
design guidelines relating to tiled windows, navigation tools, graphics and
hierarchical navigation.

Bridging the Paper and Electronic Worlds: The Paper User Interface
   Walter Johnson, Xerox PARC
   Herbert D. Jellinek, Xerox PARC
   Leigh Klotz, Xerox PARC
   Ramana Rao, Xerox PARC
   Stu Card, Xerox PARC
Since its invention millenia ago, paper has served as our primary medium of
communication, excepting oral language.  Its inherent physical properties
make it easy to use, transport, store, and cheap to manufacture.  Despite
these advantages, paper remains a second class citizen in the electronic
world.  In this paper, we present a new technology for bridging the paper
and the electronic worlds.  In the new technology, the user interface move
beyond the workstation and onto paper itself.  We describe paper user
interface technology and its implementation in a particular system called
XAX.