It is estimated that a minimum of 80% of corporate information resides in documents. The need for greater efficiencies in handling such documents has fuelled the rapid development of Document Management Systems (DMS) through the 90's.

Traditionally, there have been 2 types of document management:

1) fixed images of document pages
2) editable documents, such as word processing files and spreadsheets

These two classes differ largely in the fact that images are static, while editable documents are dynamic and changing. Systems supporting images focus on access, with input, indexing and retrieval as important functions, while systems supporting editable documents focus on creation, with multiple authoring, workflow, and revision control at the centre.

The barriers between these two classes need to be dissolved to provide integrated document management systems that incorporate a full range of document management functions. Why? Because business rely on both and because images can now easily and accurately be searched using Optical Character Recognition (OCR) techniques.

The major challenge is the integration of several layers of software systems; those for image storage and retrieval, workflow management, compound document management, and document presentation.

In the near future, comprehensive document management will be incorporated into the basic operating systems, much like current file access mechanisms. Even so, much configuration will be required.

In general, electronic documents fall on a continuum with static documents at one end and complex, compound documents at the other. A further type now needs to be specified, an 'intelligent' document carrying information about their content, structure and renewable sources. In this way, documents are reflecting the trend toward object-oriented architectures, where information is contained in objects (units of information of a finer granularity than traditional documents) which also contain information about themselves and their originating applications.

Documents in this new object-oriented conceptualisation are considered to be containers of a wide variety of information, rather than single flat files or blobs. Instead, they are simply a collection of pointers to external elements that are dynamically assembled as they are retrieved.

The resulting compound documents can be delivered as one of two types: 1) pre-specified documents that are constructed in the same way each time; or 2) virtual documents that are assembled for a specific purpose, that exist temporarily, and which may change from one viewing to another.

Now here we start to see a link with Knowledge Management, objects created from varying sources with a limited lifetime and a selected audience.

Additionally there are integration matters to deal with, normally financial, HR and logistical systems (e.g. Peoplesoft, SAP). Whilst these are important considerations, the ease of integration, or otherwise, should not override the business needs of the document system.

Colin Herbert has experience of designing and implementing Lotus Notes based systems (including Achiever Plus and developed variants) as well as a number of other systems, e.g. ZyImage and Documentum.

Based in part on the presentation of the International Federation of Library Associations and Institutions, originally in 1995.