Logo GISdevelopment.net

GISdevelopment > Proceedings > GITA > 2003


GITA 2003 | GITA 2002 | GITA 2001 | GITA 2000 | GITA 1999 | GITA 1998 | GITA 1997
Sessions

Data Management - The Evolution of Data

Disaster Management

E-Biz

Global Solutions

The Human Factor

Innovative Technologies

Mobile

Municipal Perspective

Network Operations Management

System Architecture

System Integration

User Presentations

Work Management


GITA 2003


E-Biz
Printer Friendly Format

Page 1 of 3
| Next |


Providing web services with legacy GIS

Prof. Dr. Dieter Keller
SHH GmbH Fritz-Müller Strasse 107
D-73730 Esslingen
Telephone: +49 711 31508-109
E-mail: keller@shhinfo.de


Abstract
Most of today’s GIS-work has to do with viewing, analyzing especially filtering and updating previously collected data. In other words the data that is already available in most cases, but captured in legacy systems, that are tied to the desktop or to LANs in a classical two tier architecture. But the data has to be made accessible to a broader audience in order to yield its real asset. The Internet especially its WWW service offers new techniques of integrating different systems containing different data in the form of unique web services. The paper describes a modern flexible and inexpensive three-tier system architecture wrapped around an “old” legacy system architecture. Key technologies like CORBA and J2EE are discussed in the context of this modern architecture as well as the usage of XML based SVG graphics and the role of the OpenGIS intermediate format GML. But all those technologies don’t make much sense, if there is no clear business model behind them.

The Utility-It-Environment
NIS [Network Information Systems] are special representations of GIS [Geo- Information Systems]. One popular application of NIS is the support of business processes in utility companies. A utility company runs into serious problems if the support task for kernel business processes is not sufficiently fulfilled anymore. For this, there can be the following reasons:
  • The business processes changed and are not represented correctly in the NIS
  • The technical basis of the data processing system is not supported by the manufacturer anymore.
  • The customers acceptance of the system is too small.
  • New technical evolutions offer considerably more efficient solutions in modeling of the business processes.
The last item includes the wish of a flexible adaptation of the legacy systems in the form of a constant evolution. This becomes more and more important for dynamic markets. After the liberalization the European utility market is a highly dynamic market. So, the legacy systems form a high potential risk and a drawback for rapid development in many enterprises. The following situations demonstrate the problem:
  • The knowledge monopoly of individual employees is a bottleneck; it causes the dependence on people.
  • Changes of the legacy system lead to unexpected side effects and are responsible for the general attitude: Never change a running system.
  • The maintenance costs are high - compared to the corresponding improvements.
  • The system has a high failure rate - the failure rates can be caught only through few employees.
Taking a closer look onto this list, we immediately realize that the NIS-customers and -providers share the same problems. This is because the problems are due to standard-problems of the information technology. Most of the time the provider had those problems 1-3 years earlier.

The NIS Legacy System
In the following chapters our own legacy NIS will be used as an example for the constant evolution that will finally lead to Web Services. Fortunately, our legacy NIS, Paris, has to fight comparatively little with problems of the above mentioned kinds since it is still relatively "young". When the product was shipped for the first time in 1997, it was a typical desktop NIS with a CAD frontend (MicroStation) and a local data container (MS-Excel). Already 2 years later, the product changed and was shipped as a fully equiped client/server NIS using all advantages of this technology in a Local Area Network [LAN] environment. An outstanding feature of the legacy system can be described best by “Modelling instead of Programming". In other words the NIS offers a simple object-oriented modeling (including inheritance) which is based on an integrated GIS databasearchitecture. In the utility sector usually inflexible industry-specific shells are used for approaching business processes. Instead of this, our legacy NIS offers a flexible object modelling that directly adapts to business processes. As a starting point for an object modelling process, object libraries for electricity, gas, water, waste water and long-distance heating are provided, so object modelling never starts from the scratch. When switching to the client/servers architecture a new software layer was established, that allowed almost every relational database system to be connected with the NIS (Oracle, DB2, Interbase etc.).

From the NIS point of view the legacy system supports currently most of the business processes of utility-companies. Office-tools are simply integrated and the connection to other programs can be simply realized by data exchange. But the liberalization of the markets goes hand in hand with company fusions high/dynamic changes in the IT. The future will bring a lot of changes.

Page 1 of 3
| Next |

Applications | Technology | Policy | History | News | Tenders | Events | Interviews | Career | Companies | Country Pages | Books | Publications | Education | Glossary | Tutorials | Downloads | Site Map | Subscribe | GIS@development Magazine | Updates | Guest Book