Introduction
The already existing need of friendly environments for effective
information access has been further enforced by the growth of the Internet,
which is causing a dramatic change in both the kind of people who access
the information (in a near future everybody will interact with
computers in her/his everyday life) and the
types of information itself (ranging from unstructured multimedia
data to traditional record-oriented data). To cope with these new
demands, the interaction techniques traditionally offered to DBMS users
have to evolve and eventually integrate in a powerful interface to the
global information infrastructure. The new interaction mechanisms
must be especially friendly and easy-to-use, since, given the enormous quantity
of information sources available on the Internet, most of the users
remain permanent novices with respect to each one of the
sources they have access to.
In particular, there is the need of both effective visualizations of multimedia information and retrieval tools able to overcome the existing dichotomy between browsing and querying. Indeed, in current systems the user has the duty of locating the information of interest by browsing multiple sources, which could be complex and disorganized, in a labyrinth of billions of links. This often results in both user's disorientation and cognitive overhead. Moreover, database information is traditionally structured, and retrieved by issuing queries in some ad-hoc languages. On the contrary, information residing on the Internet is retrieved by following hypertext links. In order to join such different environments, a first approach (already yielding commercial products) has been to design software tools permitting the integration of relational databases into the World Wide Web. The main idea of such tools is to allow the users to query the database through forms that are automatically created on the fly by the tool itself (e.g. [2, 3, 4]). The tool then assembles the inputs to a form into a SQL query that a DBMS can process. These tools are still very unsatisfactory from the user's point of view, not only because s/he has to know in advance where data are located, but also for the primitive interaction style, which does not support the whole set of common GUI widgets.
Why Visual
Several visual mechanisms have been traditionally adopted in so-called visual query systems (see [5, 6]). Visual representations of data schemata and instances, as well as direct manipulation query mechanisms and sophisticated browsing facilities have made the interaction with traditional databases simpler and more accessible to people without a specific preparation. Thus, they can be successfully used also in moving along the information highways. Furthermore, being "visual" is essential (even if it is not enough) when dealing with multimedia data, as it often happens on the Internet.
One of the main advantages of visual representations relies on their capacity to allow perceptual inferences, replacing arduous cognitive comparisons and computations. The most used visual representations are forms, diagrams, and icons, each one being particularly suited depending on what is to be represented. Some systems adopting 3D diagrams have recently been proposed [7, 8]. They are based on the idea that the introduction of a third dimension can increase the representational power, as shown in the precursor work on general interactive 3D environments developed at Xerox [9]. The third dimension, together with a suitable layering of the nodes, can be effectively used in visualizing the links of complex networks, such as the Web [10], when traditional overview diagrams are useless. The databases themselves can be made more accessible by transforming them into networks (unfortunately, there is no automatic translation mechanism), as in the HIFI Project [11], where a navigational interface is proposed that presents data in external databases as hypermedia networks of nodes and links. To access the database information through the interface, the user can browse across the network possibly intermixing querying (to filter relevant nodes on the network) and navigation.
Finally, it is worth noting that visual techniques fit in well with new approaches to information managing, such as the information foraging theory [12]. Information foraging refers to activities associated with assessing, seeking, and handling information sources, in analogy with the optimal foraging theory found in biology and anthropology. As animals adapt their behavior and their structure through evolution in order to survive to the changes of the environment, so humans adapt for gaining and making sense out of information. Optimality models developed for studying the animals behavior can be extended to model the information foragers in order to provide them with better interfaces and data distribution strategies. This theory also stresses the importance of using powerful user models to facilitate the seeking of information online. Suitable user models constitute the basis for equipping the interfaces with the flexibility to offering the most adequate visualizations and interaction styles to suit the different users' preferences. In other words, the interface should not be a monolithical package, instead it should dynamically change and adapt depending on the disparate needs of the various users [13].
- 1
-
C.L. Borgman.
are Online Catalogs Hard to Use? Lessons learned from
Information-Retrieval Studies.
Journal of the American Society for Information Science.
37:6, pages 387-400, 1986.
- 2
-
B.F.Rasmussen.
- A Web to Database Interface.
http://arch-http.hq.eso.org/bfrasmus/wdb, 1995. - 3
-
J.Ng.
- a Mosaic-SQL Gateway.
http://base.ncsa.uiuc.edu:1234/gsqlsrc/gsql.tar, 1993. - 4
-
R.L.Hudson.
Information Navigator.
http://info.oit.umass.edu/navigator.html, 1995. - 5
-
T.Catarci, M.F.Costabile, S.Levialdi, C.Batini.
Query Systems for Databases: A Survey.
Technical Report SI/RR - 95/17, , Università degli Studi di Roma,
Sapienza, Dipartimento di Scienze Dell'Informazione, 1995.
- 6
-
T.Catarci, M.F.Costabile (Eds.).
Issue on Visual Query Systems.
Journal of Visual Languages and Computing, 6:1, 1995.
- 7
-
J. Boyle, P.M.D. Gray.
Design of 3D metaphors for Database Visualisation
Proc. of the IFIP 2.6 Third Working Conference on Visual
Database Systems (VDB-3), pages 157-175, 1995.
- 8
-
M.H. Rapley, J.B. Kennedy.
Dimensional Interface for an Object Oriented Database
In: Interfaces to Database Systems, Series Workshop in Computing,
Springer-Verlag, pages 143-167, 1994.
- 9
-
G.G. Robertson, S.K. Card, J.D. Mackinlay.
Visualization Using 3D Interactive Animation.
Communication of the ACM, 36:4, pages 57-71, 1993.
- 10
-
S.Mukherjea, J.D.Foley, S.Hudson.
Complex Hypermedia Networks through Multiple
Hierarchical Views.
Proc. of the CHI'95 Conference: Human Factors in Computing Systems,
pages 331-337, 1995.
- 11
-
U.Cavallaro, F.Garzotto, P.Paolini, D.Totaro.
: Hypertext Interface for Information Systems.
IEEE Software, November 1993, pages 48-51, 1993.
- 12
-
P.Pirolli, S.Card.
Foraging in Information Access Environments.
Proc. of the CHI'95 Conference: Human Factors in Computing Systems,
pages 51-58, 1995.
- 13
-
T.Catarci, S.K.Chang, M.F.Costabile, S.Levialdi, G.Santucci
Visual
Interface for Multiparadigmatic Access to Databases.
IEEE Transactions on
Knowledge and Data Engineering, to appear in 1996.
- The concept of permanent novice has been introduced by Borgman in the context of online catalogues [1].
About this document ...
Databases and the Web: New Requirements for an Easy Access
This document was generated using the LaTeX2HTML translator Version 96.1 (Feb 5, 1996) Copyright © 1993, 1994, 1995, 1996, Nikos Drakos, Computer Based Learning Unit, University of Leeds.
The command line arguments were:
latex2html catarci.tex.
The translation was initiated by Isabel Cruz on Sun May 12 19:50:25 EDT 1996