5.2.2 Update control in a distributed environment 
The client/server is an advance concept in the database 
field, which is maximizing the degree of data sharing and 
system performance between different users. That concept 
can be used in the proposed model. There are two main 
objectives for that model: to propagate the update from the 
original copy to the imported one, and to verify the update 
from authorised node, and then propagate to other nodes. 
The proposed server has for main functions to support 
different views for different users (i.e. the abstract A1 from 
A for node 2); to control the updates in the net (ie. to 
propagate any update which took place at any primary node 
to the abstracted copies in the net; to include criteria 
[specifications] to evaluate the updates that might take 
place at other [non-specialised] nodes; to make use of the 
updates made at other nodes [after being approved] into the 
master data set A as well as the other copies of A in the 
net). Controlling the updates between different nodes can 
be applied by a set of rules [l. El-Sharaki, 1994]. 
NODE (1) 
A 
     
  
   
an NIDA CA 
SERVER 
TED, 
= 
    
BA 
   
  
NODE (2) NODE (3) 
  
L.DIR : Local directory LTM : Local transaction manager 
G.DIR : Global directory GTM : Global transaction manager 
A.M. : Application Manager 
Figure 4: A model for update control in a distributed 
environment 
  
  
  
5.2.3 Analysis of well established Information Systems 
Development Methodologies 
The development of a Geoinformation Utility shows 
properties that must be carefully analyzed and designed in 
order to meet the requirements of the customers; in terms 
of customers, one should differentiate between Local Users, 
around the information node in the network, and Global 
(Federal) Users requiring integration of information from 
various nodes. The development of such an utility is to be 
guided by System Development Methodologies that can 
best analyze the related complex problems (eg. distribution 
of data and processes, communication aspects and fault 
tolerance), structure them and make them easier to deal 
with. 
Four well established classes of System Development 
Methodologies (Soft, Structured, Socio-Technical, and 
Object Oriented) have been examined in an attempt to 
identify an optimum methodology for the development and 
maintenance of a Geoinformation Utility based on 
Federated Database Systems [P. Addai, 1995]. Critical 
success factors for this development and maintenance have 
been identified, together with the activities to achieve them. 
The assessment of the methodologies has been based on 
how well they support those activities. 
638 
The result is that none of the well established 
methodologies fully support the development and 
maintenance of a Geoinformation Utility; they are too 
much targeted to the development of centralized 
information systems. An optimum mix of methodologies 
is recommended. Considering the four generic phases 
of the development of information systems, ie. System 
Strategy and Planning, System Analysis, System 
Design (global and detailed), and System Realisation 
(including implementation, operation and evaluation), a 
mix of Soft and Structured methodologies is 
recommended for the Strategy and Planning phase, and 
a choice between, or a combination of Structured 
(project phasing approach) and Object-Oriented 
(product approach and horizontal/vertical integration) 
methodologies for the remaining phases of the system 
development. 
5.2.4 Other on-going research projects 
* Quality management: Quality of a product or service 
is referred as meeting the requirements and/or 
expectations of the client, and is of strategic importance 
for an organization. In a distributed environment, 
information is generated by heterogenous databases 
and processes following different quality policies and 
systems. The quality management of the 
Geoinformation Utility is complicated, and its integrity 
can be affected, by those heterogenous quality 
management processes. Total Quality Management 
(TQM) standards and practices in industry are 
analyzed, the need of geoinformation production 
organizations for TQM is assessed, and an attempt is 
made to develop TQM models, as well as 
implementation and maintenance strategies as an 
assurance for Geoinformation Utility integrity [M. 
Sarpoulaki, 1994; D. Musiega, 1996]. 
* Prototype for federating heterogeneous GISs to 
support decision making: The main objective of the 
research is to develop a spatial decision support system 
prototype for federating heterogeneous GISs in order to 
share data and knowledge in a distributed environment, 
in the framework of developing a multi-decision support 
system for environmental decision making [M. Radwan, 
Y. Bishr, E. de Espinoza, T. Mabote, 1996]. 
* Geoinformation infrastructure for data and 
services sharing: The objective of the research is to 
set guidelines and to develop a prototype for accessing 
database contents from different application nodes; the 
system aims at federating heterogeneous systems 
where the heterogeneity could be semantic, syntactic 
and/or schematic. It includes the design of, and access 
to meta data and the evaluation of wether it should be 
decentralised or not; the assessment of the different 
alternatives of networking, and outline criteria for 
selecting the most proper networking protocol that 
supports the geoinformation infrastructure requirements 
(including performance analysis of alternatives); and the 
study of alternatives for the design of the federation's 
database schema (i.e. standard schema vs. tools for 
on-line design of database schema, as required by the 
client) [Y. Bishr, 1996]. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996