Developing CM Strategies in an IT-centric World
New Synthesis or Clash of Cultures
While change is constant, certain periods, like the one we are living in, are remarkable even by that standard for their impact on the way the world works, moves, and informs itself. One of these major areas of change is the evolution of the Internet and World Wide Web from an entertainment and research medium to a primary transaction interface between organizations of all types and their intended user/clients.
Looking back, we have seen the Internet grow from novelty to ancillary medium and recently toward core transaction vehicle at all levels of government and industry, often with a remarkable lack of the rigorous design techniques that characterize well-crafted IT systems. Driven by well-publicized failures and an ever greater reliance on the Internet for core business and governmental functions, the content world is turning to its IT cousins for the rigor needed to manage its growth. The results of this turning, a remerging of two worlds long separated by custom, technology and, too often, turf, have been decidedly mixed.
Content and Data, a Deep Divide
One might ask, “what has been separated and why?” The answer comes first from the fact that the world we know as “content” grew from the text and computer publishing worlds with their roots in the 1960s. Back then, the computer publishing and business data processing communities were parts of the same family, each drawing from the strength of the other. Then in the early seventies, these two communities took different paths; building different techniques, growing different vendor communities, and addressing problems in very different ways. The business computing world developed database technologies and used them to build systems that became the foundation of business automation. The text world focused on computer typesetting, word processing, desktop publishing, and full text search, creating the “markup language” with tags embedded in the text stream to give it meaning.
Along with this technological divide came differing world views. Data Processing—IT as it came to be called—emphasized rigor, structured designs, and a “rectangular” data universe bounded by tables and rows in the database. Text, driven largely by the growth of word processing, emphasized flexibility and user experience, mapping data structures to the often unpredictable flow of text. Through the 70s and 80s, these two communities matured, communicating only when absolutely necessary and developing a not always unjustified disdain for one another. This situation persists today: two important segments of the information world following different paths, often talking with each other only when forced to. In an exaggerated but not unique example, a large public sector organization recently found itself facing the need to dramatically improve its content management and publishing effectiveness. Retaining a consultant to review its condition and help develop an enhancement plan, the organization found that its internal IT group had virtually disconnected from the content and publishing side, even though the involved information products were the core value-add of the organizations’ work. Interviews with both groups revealed that all previous attempts at collaboration had ended in misunderstanding and contention, driving management on both sides to retreat. On the publishing side, groups quietly began hiring staff with sufficient technical ability to create a “shadow” IT capability while IT followed a path that ignored the needs and challenges of the publishing side, whose work it was supposed to support. The result was an organization largely incapable of confronting any significant technological or organizational challenge, even though it had nearly 100 staff members with some connection to technology.
Growth of the Internet has Changed the Landscape
When the culture at large discovered it could complete transactions using PCs and web browsers, and as commerce and government realized how much easier their world could be if their clients dealt with a computer rather than a person, the rules began to change. Seemingly overnight, all manner of public and private institutions have rushed to put their important transactions “on the web.” Neither public nor private organizations were prepared for this onslaught. The Internet and web, powerful as it is, had not developed the rigorous design and development techniques needed to ensure that transactions attempted electronically by clients with widely varying computer abilities would be successful or, if unsuccessful, wouldn’t create havoc by their failures. Clearly, something new was needed and that something, in the view of many content and IT professionals, was to apply to the Internet world the rigorous techniques already present in traditional IT. The worlds of text and data were about to re-engage.
Content Management planners responsible for most web-based activity, facing growing demand and stung by their own and their counterparts’ failures, began to realize that their success depends increasingly on the rigorous techniques and resources present in the world of traditional Information Technology. Indeed, as content-related projects grow ever larger and more inclusive, CM strategists have often looked to the IT System Development Cycle as their savior. IT professionals, with nearly 40 years of cumulative experience in the planning and development of large systems for mission critical business functions, were ready to step in where the need has surfaced. Both groups have looked for an increased spirit of cooperation in the hope that a new, more powerful model would emerge by which information of all types may be managed and delivered.
The example organization, after review of its technological focus, found that the software tools used by its content management organizations were largely unknown to its IT group, making even general discussions of problems and solutions difficult. Statements by IT staffers like, “can’t we just do that with word processing?” added little to the conversation and generated increasing mistrust on both sides.
Unfortunately, the CM-IT model has produced mixed results, sometimes failing altogether. The reasons are complex, but the problems are not inevitable. The remainder of this article looks at some of the reasons why CM and IT have struggled to achieve synthesis and offers some thoughts on how these two critical parts of the information world can work together effectively.
Similar Needs but Very Different Perspectives
It would seem natural that two communities like CM and IT, each built on the need to create and manage large amounts of data, would work well together. That this has not universally been the case is often ascribed by the participants to some failing or incompetence on the “other side.” However, such dismissive responses usually mask the real causes and make addressing them much more difficult. In fact, CM and IT often fail to work well together because each has a very different view of the information world. Interviews with CM and IT groups in the example organization revealed highly competent people who completely failed to understand the others’ needs, limitations, and perspective. Some of these key points of difference are described below:
1. Growth of the IT world has been based on a rigorous “system development” model that fuses analysis and design in a tightly organized process created to support mission-critical business systems. Content, primarily web based, has grown as a creative response to demand always ahead of its level of organization.
2. The definition of “user,” the key to effective design, has changed dramatically from the traditional business model of “staff as primary user” to today’s growing “end user as primary participant” model.
3. Content providers in text-based efforts behave very differently from input staff in business systems. In business systems, staff is trained to work within the system’s boundaries and expected to perform at threshold accuracy defined by the systems developers. In text-based efforts, providers often come from outside the organization and behave in ways that can confound even the most circumspect system designer. To make matters worse, text providers often reject all or part of the system’s needs, focusing instead on their own.
4. The database, long a staple of business system design, assumes a certain “rectangular” nature in data that text content does not meet. Indeed, text tends to break many of the database model’s foundation precepts.
Bring together a content management group and an IT organization, often using identical terms like “requirements analysis,” “user,” “systems,” “data,” “document” and “provider” but with dramatically different though not always fully understood definitions, and you have the recipe for misunderstanding and distrust. In many efforts, including the example organization, this misunderstanding ultimately spells failure as each group spends time and effort trying to convert the other to its way of thinking. Factor in personalities and the effort grinds to a halt or hews to the desires of the side with the most organizational power, ignoring the other’s needs and expertise. After a couple of these controversies, many organizations retreat to a situation in which content and IT avoid each other whenever possible. In the example organization, both content and IT groups, unable to find common ground, descended on management, lobbying for their positions and facing top managers with impossible choices between opposed factions.
So what can be done?
A New Content-IT Synthesis is Needed
If the worlds of text content and IT are to find ways to work in a mutually supportive manner, several things will be necessary, understood, and supported by both sides.
A New Definition of “User” and “Provider”
Users— IT implicitly views its users as internal staff, hired and trained to work with the systems in place and cognizant of their performance requirements. Many if not most business systems involve only these internal users, providing the system designer with a predictable resource around which to design the transactions needed to make systems work. The new user, however, is the end-consumer with no training in the transactions he or she is expected to complete, widely varying computer savvy, and a propensity to react to failure by complaining or going elsewhere. Moreover, customer/user service staffs, long the buffer between system and end-user, are shrinking under cost pressures, often leaving system and consumers to fend for themselves based solely on “help” links on the screen.
This redefinition of user has been one of the most difficult aspects for both content and IT designers. Many IT system development assumptions map poorly to a largely uncontrolled and untrained user population, and content management planners have often been caught unprepared for the dramatically higher need for careful planning required by their attempts to support core transactions. The example organization faced a situation in which meetings between IT and user groups were heavily laden with tension and mistrust, most meetings accomplishing little.
The new definition must take all of this into account without asking either side to give up the core of its beliefs: IT system development models still apply, and their rigor is no less appropriate to the new landscape. Likewise content managers must continue to work in a world of untrained users and content that can confound the rigor of database models. The key is not which rules are applied but when each is appropriate. Before content and web-based system functions can be designed with the rigor of the IT world, each organization, each transaction set, and each user population must be carefully described:
- Who and where are they and how do they view themselves?
- Specifically what behavior is needed from them in order for the system to work?
- What may be expected from them and how consistent is that behavior likely to be?
- What do they want from the system and what information do they need to successfully use the system?
- How likely are they to know about the system, the transactions, and the desired outcomes, and how much help will they need to use the system effectively?
- What are the consequences, for them and the organization, of their failure or refusal to use the system?
- What are the likely results and costs of their failures: complaints, switch to competitors, incomplete data, lawsuits, political activity, etc?
Most of these answers will present as ranges, usually from the most to least favorable situation. Accordingly, our use of this information must often be expressed in designs that help the majority to successfully use systems without human help and to accurately predict the level of human support needed to help the others. This level of functional ambiguity is often foreign to IT groups, including those within the example organization, used to working with largely knowable challenges. Their responses are often dismissive, adding to the tension between them and the content groups they support.
While these are not trivial challenges, they can be developed by a careful and thorough analysis effort… and they should be completed before traditional IT systems techniques are applied. Indeed this information forms the prelude to rigorous design, and it is the responsibility of the content management planner to conduct the necessary analysis to develop and organize it. In the example organization, content groups faced with the pressures of tight schedules and rigid deadlines (having failed to look closely at their needs and constraints) found themselves unprepared to clearly explain them to IT, reinforcing IT’s perception that user communities are undisciplined and fuzzy in their thinking.
Providers—Likewise, the definition of “provider” has changed dramatically from the traditional business model in which system input is created or controlled by internal staff to a more variable model in which complex text content is often contributed by authors with little incentive to follow stringent format rules and where even transactional data is generated by untrained end-users. Indeed, today’s web-based transaction world often presents requirements in which the provider and user are the same person.
In any case, the new information world must be built on content provided by less controlled and less predictable providers. This new reality requires a rethinking of the system development model to provide controlled content where it is needed, often substituting an additional validation process in place of predictable content providers.
Many of the same questions listed above for users must be answered for providers and the appropriate design adjustments made for the answers. Again, it is the responsibility of the content management planner to conduct and organize this information as a prelude to the application of rigorous design techniques.
A New Understanding of Data and Content
A major difference between text content and IT views of the world is the content, or data, itself. The business IT world has developed highly effective techniques to design and manage large amounts of critical data. However, these techniques assume that data is typically designed by the same planners who design and build the systems themselves, making it relatively straightforward to ensure that data and functions integrate well. Content managers, on the other hand, usually face text content that is designed for the needs of the provider and consumer and with which the system must deal. This antagonism between business and text content has always made collaboration between IT and CM difficult, no
less so today.
The answer lies in acceptance, by both sides, that the other’s view of content has merit and must be accommodated by system design, procedures, and software. IT designers must realize, for example, that not all text content can be successfully mapped to a relational model, and content management planners must realize that text in HTML or proprietary formats does not contain sufficient logic to be managed effectively. The availability of XML and standard markup languages based on it provide the basis to manage text-based content at the level of rigor necessary for effective system design and operation, but it takes open-minded thinking on all sides to make it happen in any particular effort. Not all XML need be as complex as some document-centric forms, and not all XML is a simple remapping of relational data to the new notation. When both sides can accept these realities, they can get down to designing and using XML as it was intended and avoiding it where it doesn’t apply.
Conclusion: A New Mutual Respect
When the IT and content management communities come to understand that each has a major and legitimate role in the future, that each knows things the other doesn’t, and that each offers tools and techniques appropriate to portions of that information future, amazing things are possible. The approaches described here will help, but the most important factor in this new relationship is a conscious decision on each side to respect and work with the other. Only when that mutual respect makes itself felt in an open, collaborative approach to the growing electronic world will we see the Internet begin to approach its true potential.
About the Author
Barry Schaeffer is President of X.Systems Inc., a consulting and system development firm specializing in the conception and design of text-based information systems, with industrial, legal/judicial, and publishing clients among the Fortune 500, non-profit organizations and government agencies. During his more than forty-year career, Mr. Schaeffer has held management and technical positions with The Bell System, Xerox, Planning Research Corporation, US News & World Report, and Grumman. As a consultant, he has supported over 40 clients, including The Bureau of National Affairs, The Congressional Research Service and OTA, The U. S. OMB, McGraw Hill, Boeing, Rockwell, United Technologies, Matthew Bender, IBM, and U. S. Defense agencies. He is a frequent contributor on subjects related to information management.