Role of data in Geographic Information Systemsby Sathya Prasad Nanjundaih, GIS Technologist, GIS Section, Land Information Center (LIC) |
|
| Contents Page | Al-Khabar Home Page | CGIS Home Page |
Recently, there was a discussion in one of the GIS list servers on a very basic topic "What is GIS?" There were hundreds of definitions from people belonging to various disciplines. Each one had his/her own viewpoint and each one was correct from his/her perspective. People from every discipline have their own definition for GIS. Inspite of GIS being defined through various disciplines, the components of GIS can be categorized into five key components: hardware, software, data, people and methods.
Assuming the readers of this magazine are aware of the definition of GIS and its components, let us look at the another question - What is the Most Vital Component of GIS? Hardware and software can be purchased off the shelf from vendors. Of course, a few components of software such as graphical user interfaces (GUI) and tools for input and manipulation of data need to be developed in-house. Data is the most vital component, as most of it has to be built in-house even though some amount of it may be acquired from other established GIS databases. Studies in GIS implementation have proved that collecting data and encoding it as intelligent information requires about 80% of the total cost of GIS implementation. Specialized people and efficient methods are other key components that can make GIS all the more successful.
Head of Centre for Geographic Information Systems (CGIS), Qassim Mohammed Al-Ghanim, fully agrees that data is the foundation for any GIS to prosper. GIS in Qatar has proved its point by amassing a wealth of data at the CGIS and its various agencies. Here, GIS is being implemented as an "Operational System" as compared to an "Informational System".
An
"Operational System" is a system that helps a department or
enterprise to operate on a day-to-day basis. On the other
hand, an "Informational System" has to do with analyzing
data and making decisions regarding the present and future
based upon past accumulated data. GIS cannot be implemented
as an informational system without gathering sufficient data
over a period of time. And data can only be gathered over a
period of time efficiently through an operational system.
The reason being that GIS data keeps changing with time -
going through a metamorphosis - in other words GIS data is
dynamic and not static. Hence, GIS must first be implemented
as an operational system so that the system feeds itself
over a period time. Later on, it can be expanded to serve as
an informational system. In order to implement GIS to
perform as both operational and informational systems, the
suitable data models must be designed.
The Arc/Info GIS software has changed over the past few decades as the inherent data models are becoming more robust in handling various kinds of features like regions and route systems. More amendments are being incorporated in the new GIS software: Spatial Database Engine (SDE) in the directions of Relational Database Management Systems (RDBMS) and Object Oriented Programming (OOP). Issues that are popular in the RDBMS and OOP will definitely serve in designing the data models for the next generation of people implementing the GIS technology. One of the concepts being implemented in the RDBMS world in terms of handling huge amount of data in Operational and Institutional Systems is "Data Warehousing".
Data Warehousing provides organizations with flexible, effective and efficient means of storing, retrieving and archiving sets of data. In essence, data warehouses are large databases organising operational data in a repository for easy query and analysis. Since Qatar's GIS is still an operational System, the "Data Warehousing" concept plays a significant role as data sets grow with the archiving of features. Proper strategies and methods must be considered in implementing a GIS Data Warehouse, otherwise soon most of the GIS databases will be turned into Data junkyards containing inaccessible and obsolete data.