By Don Reid
Senior Associate, Zasio Enterprises, Inc.
President, Toronto Chapter, Association of Records Managers and Administrators

Document management is clearly in vogue these days. Everyone is talking about and thinking about document management, and trying to manage their information resources using this new and exciting concept.

Records, files, documents call them what you will are most often created and frequently stored using computers. The power of computers has enhanced the creation and access to information, and allows us to handle more information than ever before. But it also provides more access to information than ever before.

Clearly something must be done to manage that information. Equally clearly most of the electronic information - up to 8% - is created in the form of documents as distinct from database records.[1] It seems as if each new development in technology occurs, the ease of access to information is enhanced yet ease of finding suffers in inverse proportion.

The rise of documents also has a dark side: the information glut. The explosion of desktop documents has spilled over onto servers. Many people are hooking up to the Internet and other on line services, where millions more documents reside, ready for the taking and misplacing. The average user has enough trouble creating directories or structuring files into folders, much less remembering later where he or she has put the files.

One of my favourite experiences is of conducting an inventory of electronic records in an office where it took them an average of 10 minutes to retrieve a file loaded on a local hard drive, but it took me about 30 seconds to determine the problem: one DOCS subdirectory and everything named after the author; Jane1, Jane2, etc.

There are software packages that allow up to 254 characters in the name identifier field but is this the panacea? This may work if you are using one software package. With "software suites" like Microsoft Office Pro and Perfect Office, it may be possible to tie documents from various applications together. However, for those users that have Lotus and WordPerfect and Inmagic or such products that are not related to a suite, and those of us who have not moved to Windows NT or Windows 95 or OS/2, we still rely on those old DOS conventions of eight character names and three character extensions to find our files in subdirectories.

There are broader implications for both the librarian and the records manager from both the explosion of information and the need to manage it. Traditionally, records managers tended to organize, classify and maintain unpublished documents and records created internally or received into a business as a function. Librarians tended to classify, organize and maintain information either published by the business or obtained from published sources. These broad definitions may be open to challenge but within this context, the roles were relatively clear.

The advent of desktop publishing, Internet, and the exchange and access to electronic information, have undercut the tried and true methods of identifying authors, copyright, or maintaining control of versions of published material.

The result can be literally a mountain of information of various formats available to users. But how to access that information? The solutions garnering much interest today among information management professionals are the document management software packages that promise to manage you out of your information quagmire. Most of them can full text index all documents and thereby make everything retrievable. But the index adds to your hard disk requirements and, although there is an opportunity to classify records by document type, there generally is little opportunity to create a logical and hierarchical structure by which records can be stored and sorted by subject type. You generally must rely on title, document abstract, document type and the full text index to sort records and make determinations on retention requirements.

If your environment is distributed electronic storage with true shared access to information, is this enough? I think not. What document management requires is the adoption of some of the sound principles of library and records management: classify records for retrieval and retain records to meet legal and business requirements. This sounds simple, but it requires implementation of decisions and some forethought and much ground work before you begin to solve your problems by throwing software at them.

It may be cost prohibitive to look at document management software for your environment. As an analyst with a large organization, I worked on a document management pilot project to encompass 50 users on a distributed computing environment. The project, conducted within the premise of open database design using a SQL database product with database and document management software alone, would cost in the neighbourhood of $50,000. Although the payback in terms of access to information (generally a "soft dollar" saving and by nature is hard to measure), the expense is warranted. The reality is that neither records management nor library programmes are likely to have this kind of money at their ready disposal.

There are two alternatives to spending the money for a document management system: firstly, have Information Technology take ownership and introduce the programme, or secondly develop naming convention standards, directory maintenance and creation standards, a policy for accessing and storing information etc. I will dismiss the IT group and risk raising their ire. Most IT groups do not have the expertise in terms of classification requirements, naming conventions, vocabulary control, retention scheduling requirements and copyright/information distribution to adequately impose the controls required around a document management system.

But to manage documents without using document management software? Using Novell's Netware and DOSSHELL or Windows, it is possible to create menu items that will store documents in predetermined directories and subdirectories based on the selection of the appropriate subject identifier from the menu. The main requirements for systems like this are a sound knowledge of the network software and the operating system in use. More importantly, the group setting the policy must have the ability to train users to follow the system as devised and the mandate to enforce the policy.

Although we have all heard this more times than we can remember, it is essential to have management support and a champion for any document management system. Without the ability to enforce policy and carry the programme to the users, any system will fail. You can have the best software or organization and classification control system on the planet, but it will fail without the authority to support the system.

Even if you have a sure winning argument to convince users how much better off they will be if they access your system to store and retrieve the information, the first stumbling block will be the requirement to share information.

Within a true document management system, all data is available to be shared, provided user access and logon and security environments gives them access to the data. All you need to do is tell your users that their information will be accessible across the organization and they will immediately balk. They do not see the information as a corporate resource but rather as their personal collection of data.

Without the mandate and the champion, this attitude will not be overcome, and the programme will fail as business units opt out of the control system. So the initial requirements for a document management system should be the mandate and the authority to act and the ability to provide some organization and classification controls to the information you are trying to manage.

But at some point, even the most organized, strongest group will be unable to adequately manage the information without using software. If your system is large enough to require you to keep archive documents, search for items with full text and profile searches, track multiple versions of the same document, retrieve documents from a multiple server LAN or a WAN environment, or manage thousands of documents, then you need some software help.

I have already mentioned that the lack of capacity to classify the material other than by document type as a major shortcoming. Another is that your users must complete a profile of each document they enter as the index into the retrieval system. As this is very intrusive to the user, it is essential that the software does not make matters worse by being difficult to use. Another thing I would recommend avoiding: any software that uses a document storage path for any level of retrieval. Users were never able to keep the paths straight before, why would you expect them to do so with a document management system?

Document software creates many opportunities for better management of information resources but it requires management, discipline and planning as does all management. So, if you thought you could get off easy, think again. "If you are poorly organized, buying a computer simply makes you a poorly organized person with a computer. By the same token, a document management system alone can't undo your company's information snarl."[2] [1] DataPro Document Imaging Systems, p. 1, McGraw Hill, November 1994. [2] PC Magazine, June 29, 1993, v12, n12, pp. 171 190.