A data management system comprises a database system that stores information about all the documents and drawings, their storage locations, revision history, approvals, issues and inter-relationships. It makes it easy to find documents by various search criteria, and to also find all related documents. It may also serve to manage the archiving or back-up of data, and the gathering of data for issuing or transmittal.

Data management systems used to be either very powerful, costly systems that could handle immense volumes of data and involved a daunting implementation process, or else were suited to small offices but had definite upper limits on their ability to handle the data of a growing business. As well as the data volume handling capability, data management systems are also expected to provide access control to ensure that only those who are authorised can view, print or alter data.

Access security

There are two main methods of implementing access security. One relies on the file access control mechanisms of the computer network operating system. The other imposes an additional layer of protection by hiding all the data it controls in a secret place, called a 'vault'. The files stored in the vault are encrypted or otherwise concealed, and can only be accessed via the data management system.

A vault-based system offers the ultimate in access security, but at the cost of considerable complexity and generates concerns about the continued accessibility of the data if the management software malfunctions.

Straightforward task

Practical Programs of Australia has developed DataViewer, a CAD-oriented data management system suitable both for quite small networks and data volumes, as well as for very large systems. Its initial implementation is quite a straightforward task that does not take long or cause major disruption to the office while in progress, as is the case with vault-based systems. It runs on Windows NT or 2000 networks and uses the access controls provided by the network system.

Database systems

The main factor affecting a system's ability to handle very large data volumes is the type of database system it uses for holding the records. There are basically two types of database systems.

One uses an application program that is run on each user's computer whenever that user needs to access the database. The actual database files are stored in one location, such as on the network file server with shared access. These systems need to copy sections of the database into a local work area whenever users have to alter or extract data. If there are a large number of simultaneous users and very large database files, the network traffic soon becomes so great that the system becomes sluggish.

The other type is called a client-server system. Here the database server program runs only on the network server and runs continuously. Users run the database 'client' application, which sends queries to the central data manager and receives copies of just the selections of data required, or sends update requests to the server. Common systems of this type are Oracle, Microsoft SQL/Server, Interbase, Sybase.

Usually, the smaller data management systems have a database system built-in of the shared-data type. The large systems are usually designed to work with one particular client-server database system.

DataViewer is unusual in being able to work with a variety of database systems of both types. This makes it adaptable to a wide range of office sizes. It also means that it can make use of an office's existing database system, such as that used for inventory control or ordering. The database systems it supports are Microsoft Access, Paradox, Interbase, Microsoft SQL/Server and Oracle. Interbase is a client-server system produced by the makers of the programming system used for DataViewer, Delphi, and will often prove to be the most cost-effective option when an office does not already have a database system.

Typical Windows application

Data management systems, by their nature, usually need to be customised to each office situation. With DataViewer, the system can be set up much like a typical Windows application, and configured to the office situation interactively via a 'wizard' that asks questions in sequence. The system provides default on-screen and printer form layouts, which are often adequate as delivered, but can at any time be customised. Hence the system can be implemented quickly and easily and brought into effective operation, and then its configuration and the design of its forms and screens can be refined in the light of experience as work proceeds. It also avoids the daunting implementation hurdle that had obstructed the adoption of some data management systems.

DataViewer can be configured to perform repetitive tasks automatically. It can be used to circulate information by sending people links to the central data rather than sending copies. This avoids filling up disks with multiple copies of files and also avoids the risk of having superseded information held locally.

DataViewer does not require data to be moved to its own special storage areas, and allows one to organise data storage as desired, possibly split over several servers. It provides several levels of access permission, applied to individual documents or groups of files. In addition, files can be protected with 128-bit encryption without having to be moved or renamed.

Projects can be set up and configured to display data files organised in a hierarchic system appropriate to that project, even though their actual storage locations are organised quite differently. This is called a Project Tree. A web client module allows access to the system with proper control, from anywhere on the intranet or Internet. An unusual facility is digital signatures

Further, DataViewer includes a system for collecting selected files and their related files and bundling them together in a zip file together with transmittal details. It can also organise the emailing of such transmittals. The standardised zip process can considerably compress many types of data, including line-based CAD data, although raster image data is harder to compress if scanned to a compressed format.

DataViewer also includes adata compression system for colour raster images that has an unusual capability called 'colour streaming' when the compressed data is used on a network or web server. In those situations, it serves up just enough resolution and data extent to satisfy the display needs at the time, interactively. For example if one has to access a large map and the entire extents of the map fill the viewing area, it sends no more resolution than can be usefully displayed - enormously less data volume than the full extent, full resolution of the data on the server.

Kinetic Distribution

(031) 266 7027