Second IFCIS International Conference on Cooperative Information Systems
(CoopIS'97)
Tutorials
TUTORIAL 1: FULL-DAY
Object and Agent Standards for Information Integration
and Interoperation
Sumi Helal, MCC, helal@mcc.com
Darrell Woelk, Rosette Webworks Inc., D.Woelk@rosette.com
ABSTRACT
New technological advances such as the World Wide Web are responsible for
the explosive growth of the available sources of information. They have
also created unprecedented demands on accessing information by a diverse
population of users on heterogeneous platforms. We are, therefore, faced
with grand challenges with respect to the massive scale and the
definitive heterogeneity of these systems.
Information standard developments have recently gained a critical mass and
are anticipated to play a major role in enabling information integration
in open and growing environments. These standards offer high degrees of
portability of a given information source, and interoperability among
different heterogeneous sources.
In this tutorial we focus on two categories of standards that reflect two
orthogonal approaches to information integration: object- and agent-based
standards. Both approaches provide support for mediated interoperation
(or brokering) of data and services in dynamic and open environments.
The tutorial covers the major object and agent standards and describe
their current development and acceptability status. The tutorial also
compares the two approaches and speculates about their prospects in the
near future.
=======================================================================
TUTORIAL 2: HALF-DAY
Design of Workflow Management Applications and Systems
Prof. Dr. Stefan Jablonski
University of Erlangen-Nuernberg
jablonski@informatik.uni-erlangen.de
ABSTRACT
Statistics says that about 80 % of all projects in the area of workflow
management fail (in Germany, 1995-1996). Why is there such a high
failure rate? Not only inadequate workflow models (workflow languages)
are the reason but also inadequate methodologies to develop workflow
management systems and applications which both have to be regardes as
software systems.
There are decent methodologies for software development. Nevertheless,
there are no methodologies out there yet which show how to design and
implement either workflow management applications and systems. Most
publications about the development of workflow management systems do not
reflect the experiences and results gained over two decades of software
engineering. For instance, almost no approach clearly distincts between
logical design (implementation model) and its subsequent enactment
(implementation architecture). Thus, most architectures for workflow
management systems are very specific and 'handmade' and do not expose
general features which can be leveraged by similar approaches.
Workflow management systems are a specific kind of software systems
which also needs methodological support during design and
implementation. In the tutorial a development methodology for workflow
management systems will be presented which reflects experiences gained
either throught the development of a concrete workflow management
product in an industrial company and through the studies of software
engineering and software architecture.
Also development methodologies for the design and implementation of
workflow management applications are still missing. There are well
developed workflow models (e.g. workflow languages); however, software
engineers do not know who to deploy such a workflow model in order to
get to a decent workflow management application. The tutorial will
present a development methodology for workflow management systems which
supports the design and implementation of workflow management
applications.
=======================================================================
TUTORIAL #3: HALF DAY
Current Issues in Information Integration
Daniela Florescu, AT&T Laboratories
Alon Levy, AT&T Laboratories, levy@research.att.com
ABSTRACT
In the last few years there has been considerable interest in the
problem of providing access to large collections of distributed
heterogeneous information sources (e.g., sources on the World-Wide
Web, company-wide databases). This interest has spawned a significant
amount of research in Database Systems and related fields (e.g.,
Artificial Intelligence, Operating Systems, Human Computer
Interaction). This has led to the development of several research
prototypes for information integration and recently, we are seeing the
beginnings of an industry addressing this problem.
This tutorial will survey the work on information integration, to
illustrate the common principles underlying this body of work, to
assess the state of the art, and identify the open research problems
in this area. The tutorial will illustrate the issues involved in
information integration through several implemented systems. Among
others, the tutorial will cover the topics of information integration
architectures, modeling information sources, query processing and
optimization algorithms, handling semistructured data and wrapper
generation, and issues in commercializing information integration
concepts.
Biographies:
Daniela Florescu is a senior member of the technical staff at AT&T
Research Laboratories. She received his B.Sc. in Computer Science and
Mathematics from the University of Bucharest in 1991, Masters of
Computer Science from University of Paris VI in 1992, and a Ph.D. in
Computer Science from INRIA and the University of Paris VI in
1996. Her current research interests are information integration,
query optimization in object-oriented database systems, query
execution models for parallel databases, query reformulation in
multidatabase systems, semi-structured data, and web-site management
systems,
Alon Levy is a principal member of the technical staff at AT&T
Research Laboratories. He received his B.Sc. in Computer Science and
Mathematics from the Hebrew University in Jerusalem in 1988, and his
Ph.D in Computer Science from Stanford University in 1993. His current
research interests are information integration, semi-structured data,
materialized views, web-site management systems, knowledge
representation and connections between database systems and Artificial
Intelligence. In the past 3 years, he has led the Information
Manifold at AT&T Laboratories.