A fluid treemap interface for personal digital libraries
Proceedings of the 5th ACMIEEECS joint conference on Digital libraries JCDL 05 (2005)
- ISBN: 1581138768
- DOI: 10.1145/1065385.1065512
Available from portal.acm.org
or
Abstract
The UC system employs hybrid quantum/continuous treemaps for fluidly interacting with documents in a personal digital library. By incorporating a document reader application within the visualization workspace, UC supports multi-document reading tasks that have been traditionally accomplished by laying out documents on a physical desk. One of the overall goals of the system is to eliminate the boundary between acquiring and using documents.
Available from portal.acm.org
Page 1
A fluid treemap interface for personal digital libraries
A Fluid Interface for Personal Digital Libraries
Lance E. Good, Ashok C. Popat, William C. Janssen, and Eric A. Bier
Palo Alto Research Center,
3333 Coyote Hill Road, Palo Alto, California 94304
{good,popat,janssen,bier}@parc.com
Abstract. An advanced interface is presented for fluid interaction in a
personal digital library system. The system employs a zoomable planar
representation of a collection using hybrid continuous/quantum treemap
visualizations to facilitate navigation while minimizing cognitive load.
The system is particularly well suited to user tasks which, in the physical
world, are normally carried out by laying out a set of related documents
on a physical desk — namely, those tasks that require frequent and rapid
transfer of attention from one document in the collection to another.
Discussed are the design and implementation of the system as well as its
relationship to previous work.
1 Introduction
The persistence of paper as a preferred medium for document interaction, in
the face of distinct and growing advantages of electronic representations, has
in recent years been well-studied [1]. Perhaps nowhere are the advantages of
paper over extant electronic alternatives so strong as in the case where the
reader wishes to compare a set of passages appearing in several documents, or
to cycle attention among passages from several documents in rapid succession.
Such patterns of interaction arise naturally during a variety of knowledge tasks,
including the preparation of lectures, the reviewing of papers, the analysis of
intelligence briefings, and the writing of a report, among many others. We shall
refer to such patterns as “reading from multiple sources,” with “reading” to be
taken in its general sense to include such activities as annotation, extraction,
summarization, and the like.
We consider reading from multiple sources in the context of a personal digital
library. By the qualifier “personal” here we mean that the user (1) already has
the right to use all of the data objects in the library, and (2) already has local
possession of those objects. An example of such a personal digital library is the
result of scanning one’s office filing cabinet onto a local hard drive.
In recent years, the importance of effective visualization tools and visual
interfaces to digital library collections has been recognized [2,3,4]. In the case of
a personal digital library, an opportunity exists to blur the traditional separation
between searching for materials and using them. The system presented in this
paper pursues this goal.
In the physical world, reading from multiple sources often involves spreading
out a set of documents on a large desk or table. During the course of the activity
A. Rauber et al. (Eds.): ECDL 2005, LNCS 3652, pp. 162–173, 2005.
c
© Springer-Verlag Berlin Heidelberg 2005
Lance E. Good, Ashok C. Popat, William C. Janssen, and Eric A. Bier
Palo Alto Research Center,
3333 Coyote Hill Road, Palo Alto, California 94304
{good,popat,janssen,bier}@parc.com
Abstract. An advanced interface is presented for fluid interaction in a
personal digital library system. The system employs a zoomable planar
representation of a collection using hybrid continuous/quantum treemap
visualizations to facilitate navigation while minimizing cognitive load.
The system is particularly well suited to user tasks which, in the physical
world, are normally carried out by laying out a set of related documents
on a physical desk — namely, those tasks that require frequent and rapid
transfer of attention from one document in the collection to another.
Discussed are the design and implementation of the system as well as its
relationship to previous work.
1 Introduction
The persistence of paper as a preferred medium for document interaction, in
the face of distinct and growing advantages of electronic representations, has
in recent years been well-studied [1]. Perhaps nowhere are the advantages of
paper over extant electronic alternatives so strong as in the case where the
reader wishes to compare a set of passages appearing in several documents, or
to cycle attention among passages from several documents in rapid succession.
Such patterns of interaction arise naturally during a variety of knowledge tasks,
including the preparation of lectures, the reviewing of papers, the analysis of
intelligence briefings, and the writing of a report, among many others. We shall
refer to such patterns as “reading from multiple sources,” with “reading” to be
taken in its general sense to include such activities as annotation, extraction,
summarization, and the like.
We consider reading from multiple sources in the context of a personal digital
library. By the qualifier “personal” here we mean that the user (1) already has
the right to use all of the data objects in the library, and (2) already has local
possession of those objects. An example of such a personal digital library is the
result of scanning one’s office filing cabinet onto a local hard drive.
In recent years, the importance of effective visualization tools and visual
interfaces to digital library collections has been recognized [2,3,4]. In the case of
a personal digital library, an opportunity exists to blur the traditional separation
between searching for materials and using them. The system presented in this
paper pursues this goal.
In the physical world, reading from multiple sources often involves spreading
out a set of documents on a large desk or table. During the course of the activity
A. Rauber et al. (Eds.): ECDL 2005, LNCS 3652, pp. 162–173, 2005.
c
© Springer-Verlag Berlin Heidelberg 2005
Page 2
A Fluid Interface for Personal Digital Libraries 163
the reader is at various times absorbed in the close study of one document
or another, while at other times, be it in transition or in respite, the reader
broadens his or her focus momentarily to regard the workspace as a whole. Fluid
transitioning among these modes of attention is achieved, and the continuity of
orientation maintained, by virtue of the persistence of the layout of the physical
documents in a single visual field, and through the ability of the reader to adjust
his or her center and field of attention rapidly and at will.
The system described in this paper is an attempt to simulate and, at least
with respect to navigation, surpass in the electronic realm the affordances of
reading from multiple paper sources. A uniform and consistent interaction inter-
face is provided to every document in a zoomable virtual workspace, irrespective
of its native electronic “format” or “type.” This universality is enabled by a fast,
transparent, and automatic means of converting arbitrary documents of interest
into a common intermediate working representation. To support the naviga-
tional cues that get built up in spatial memory during the course of the activity;
advanced visualizations known as treemaps are used in the system’s interface.
Provision is made for navigation by interactive incremental textual search.
1.1 Universality of Representation and Consistency of Interface
Phelps and Wilensky [5] note that “picking a format is often tantamount to
choosing a browser/editor/viewer with its packaged bundle of features and lim-
itations.” The diversity of applications required to interact with different doc-
ument types leads to a proliferation of user interfaces, placing the additional
burden on the user of having to learn and remember their proper operation.
Personal digital library systems can mitigate these problems by allowing
documents to be converted into a common intermediate format for purposes of
indexing and viewing. Reading applications need then support only this single
common format. Because the addition of new information sources is a frequent
and integral part of the reading activity, conversion into this common format —
including the extraction and processing of any requisite metadata, must be as
fast, transparent, and automatic as possible.
1.2 UC: A System for Fluid, Seamless Interaction
This paper describes a system for reading from multiple documents which ad-
dresses the above mentioned shortcomings of current digital reading systems.
The system, called UC (a name deriving from its initial but no longer used in-
ternal name “UpLib Client”) is built on the UpLib personal digital library plat-
form [6], which provides an extensible, format-agnostic, and secure document
repository layer. UpLib achieves a degree of universality in accommodating mul-
tiple document formats through heavy reliance on two principal transductions
of the original document, one into the image domain and the other into the text
domain. From the perspective of UC, the image projection facilitates simulation
of the visual experience of reading from multiple paper documents, while the text
projection enables search-based navigation. By virtue of its leverage of UpLib,
the reader is at various times absorbed in the close study of one document
or another, while at other times, be it in transition or in respite, the reader
broadens his or her focus momentarily to regard the workspace as a whole. Fluid
transitioning among these modes of attention is achieved, and the continuity of
orientation maintained, by virtue of the persistence of the layout of the physical
documents in a single visual field, and through the ability of the reader to adjust
his or her center and field of attention rapidly and at will.
The system described in this paper is an attempt to simulate and, at least
with respect to navigation, surpass in the electronic realm the affordances of
reading from multiple paper sources. A uniform and consistent interaction inter-
face is provided to every document in a zoomable virtual workspace, irrespective
of its native electronic “format” or “type.” This universality is enabled by a fast,
transparent, and automatic means of converting arbitrary documents of interest
into a common intermediate working representation. To support the naviga-
tional cues that get built up in spatial memory during the course of the activity;
advanced visualizations known as treemaps are used in the system’s interface.
Provision is made for navigation by interactive incremental textual search.
1.1 Universality of Representation and Consistency of Interface
Phelps and Wilensky [5] note that “picking a format is often tantamount to
choosing a browser/editor/viewer with its packaged bundle of features and lim-
itations.” The diversity of applications required to interact with different doc-
ument types leads to a proliferation of user interfaces, placing the additional
burden on the user of having to learn and remember their proper operation.
Personal digital library systems can mitigate these problems by allowing
documents to be converted into a common intermediate format for purposes of
indexing and viewing. Reading applications need then support only this single
common format. Because the addition of new information sources is a frequent
and integral part of the reading activity, conversion into this common format —
including the extraction and processing of any requisite metadata, must be as
fast, transparent, and automatic as possible.
1.2 UC: A System for Fluid, Seamless Interaction
This paper describes a system for reading from multiple documents which ad-
dresses the above mentioned shortcomings of current digital reading systems.
The system, called UC (a name deriving from its initial but no longer used in-
ternal name “UpLib Client”) is built on the UpLib personal digital library plat-
form [6], which provides an extensible, format-agnostic, and secure document
repository layer. UpLib achieves a degree of universality in accommodating mul-
tiple document formats through heavy reliance on two principal transductions
of the original document, one into the image domain and the other into the text
domain. From the perspective of UC, the image projection facilitates simulation
of the visual experience of reading from multiple paper documents, while the text
projection enables search-based navigation. By virtue of its leverage of UpLib,
Sign up today - FREE
Mendeley saves you time finding and organizing research. Learn more
- All your research in one place
- Add and import papers easily
- Access it anywhere, anytime
Start using Mendeley in seconds!
Readership Statistics
5 Readers on Mendeley
by Discipline
20% Medicine
by Academic Status
40% Ph.D. Student
20% Student (Bachelor)
20% Researcher (at an Academic Institution)
by Country
20% South Korea
20% China
20% United Kingdom



