Papers
CHI
99 15-20
MAY
1999
Hyper Mochi Sheet: A Predictive Focusing Interface for
Navigating and Editing Nested Networks through ia
Multi-focus Distortion-Oriented
View
Masashi Toyoda and Etsuya Shibayama
Department of Mathematical and Computing Sciences
Tokyo Institute of Technology
2- 12- 1 Oookayama, Meguro-ku,
Tokyo 152-8552 JAPAN
+8 l-3-5734-3870
{toyoda,etsuya} @is.titech.ac.jp
ABSTRACT
Multi-focus distortion-oriented views are useful in viewing
large information on a small screen, but still have problems
in managing multiple foci during editing. The user may have
to navigate information space by focusing and defocusing
multiple parts to obtain multi-focus layouts that change ac-
cording to various editing situations. As a result, it becomes
haphazard to navigate and edit large nested networks such
as hypertexts. We propose a user interface for quickly ob-
taining desirable layouts. The interface uses two techniques:
focus size prediction and predictive focus selection. These
techniques are based on a user test and experiences in appli-
cations. We also describe two example applications.
Keywords
distortion-oriented view, multi-focus, editing, navigation
INTRODUCTION
Multi-focus distortion-oriented views [ 14,10,1,4] are useful
in viewing large information on a small screen. These views
provide more flexible layouts of focused parts while preserv-
ing the overall context compared to single-focus distortion-
oriented views[8, 13, 12,7,5]. As such they seemingly have
the potential for scalable editing of large networks such as
visual programs and hypertexts.
In reality, however, they
have not yet supported efficient editing interfaces of large
networks due to problems in managing multiple foci. More
flexibility
and freedom of multi-focus layouts often require
more work by the user than in single-focus views, in which a
change of layout involves only focus movement and change
in magnification
factor. Rather, in multi-focus views, the
user may have to perform bo