Figure 1 below depicts an evolutionary approach to the creation of a Knowledge Management environment, which is comprised of explicit and tacit data. The key to Knowledge Management is the successful use, understanding, and availability of data. Explicit data, as shown in the figure, has traditionally been assigned to the purview of Information Technology. Explicit data is broken down into two categories: structured and semi-structured data. Structured data is table data or relational database data; semi-structured data is that which is contextually presented in analyses, reports, technical manuals, or business data. A natural outcome of structured data is the need to configuration manage it. CM of semi-structured data is not so straightforward or common. Tacit data is defined as experience, recipes, or lessons learned that stimulate improved methods or better outcomes due to new skills or knowledge. Knowledge Management is, then, the combination of these two kinds of data - and the cornerstone of the collaborative framework and environment. It may be that the goals of Knowledge Management are the stimulus for content management, and that content management finally understood to be the "missing link" for success. For a KM framework to successfully exist, significant and sufficient Data Management work must have had to be done - work which is currently outside the areas of the traditional IT, CM, and DM task areas. DM must step up to this task and successfully target areas such as KM where its contributions are badly needed - thus fulfilling the goals of "content data management".

We have known for some time that the use and benefit to which data will be put is a function of how it's been acquired, organized, characterized, managed, and stored. The key to making data a business asset is attaching to the data a clearly recognizable indicator that demonstrates ROI (return on investment) of the effort to collect and manage that data - creating processes that ensure economies of scale in investments made in the data: acquisition, use, value, and re-use. But a different set of functions and services must emerge that will augment and address the world of Information Technology data management - none of which are modeled in either current IT or current DM processes and approaches.

At the core of the business organization, there are data and the processes that evolve this data to business events and decisions. However, in order to move to true Knowledge Management, the data must be well-integrated, managed, and available in real-time to all systems, programs, and knowledge workers who can convert the data to business knowledge. Business knowledge is the intangible asset that can distinguish a corporation from a market leadership position. But only if the data asset that business knowledge represents is well-managed and can be distributed in a timely and beneficial way. Hence, the motto of Data Management has always been, "the right data to the right people at the right time". It is only after information technology began to be the pipeline for data delivery (rather than the "library" function that the traditional Data Manager represented, in the past) that issues regarding data collection, manipulation, and storage began to create sizable (and difficult) problems for IT to resolve. So what may be key for the success of Content Management is the successful implementation and automation of the library functions, taxonomy development, location, relevancy, and context assignment for data that the Data Manager has contributed in the past. While data collection, manipulation, and storage create a significant IT problem, a solution which only takes into consideration the "IT approach" and definition to these problems hardly attacks the problem where it exists most fundamentally. Business data are of two disparate types, both of which must be managed: structured information (found in the left side of the figure above) stored in repositories or databases, and unstructured information - which is stored in documents.

In a very fundamental way, it is this last class of information which is actually the more valuable of the two classifications of data - since it is here (in the unstructured realm) that knowledge has actually begun to form. Even when the thinking is preliminary, inconclusive, or not directly related to the current knowledge worker's need (e.g. access), there is some "information" which has begun to convert "data".

Since the data which we have termed "unstructured" are in many different formats (HTML, XML, web content, document images, documents, reports, manuals, drawings and lessons learned. Some of these artifacts lend themselves to formats like XML models - in fact all of them are "useable" in XML if they are properly characterized and the appropriate metadata are created for them - but such data products as reports, manuals, and documents do not easily translate - as lessons learned, or useful information -to XML models without considerable manipulation. It is here that the practice of Data Management probably holds the most promise in the future, and the most use in the IT community.

"Content Data Management" (CDM) is more than just servers, transactions, frameworks, and storage devices - it is attaching the meaning and relevance of the data so that they may be more useful and provide value to users. Ironically, CDM also means "configuration data management" - which is universally missing from most small or large-scale PDM or other similar implementations.

The value or success metric for Content Data Management should not be how fast it serves up the data, or how large a server it features. These are IT metrics. And that's the problem - the wrong metric is being advanced for the wrong function. The metric for CDM has to be not only the evolving (and somewhat traditional) DM functions (planning, identification, metadata development, "card catalog" characterization, relationally-organized, etc) but also well the data are versioned and change-managed - which are typical configuration management functions - and the measurement of how often the correct data got to the user, saving him or her time in the process and enabling a more valuable outcome through quality data use.

Another area which is complicating the realm of and for Content Data Management is the dissimilar views about which data are the most critical for the organization: "management" views different data (budget, manufacturing, quality, schedule, contract, financial) as critical than does "engineering". "Engineering" views analysis, technical, scientific, CM, and quality data as the most critical. Neither is exclusively correct - the organization runs on all of that data. Until and unless PDM tools allow all types of data to be incorporated effectively in the tool - and "content" is assigned to the data - there will continue to be deficits and disconnects for everyone who is a user of the data. And so long as PDM systems require users and customers to "use our framework" and require them to do significant and costly customization as an add-on cost, PDM systems will continue to be unsuccessful tools for a collaborative use environment. In the end, if a system is not useful or flexible it's not a good solution and it's a black eye for its proponents and purchasers.

The biggest misconception there is about resolving "content data management" issues is that just throwing money at the problem will not mitigate the issues or resolve the dilemma. Only characterization will provide the kind of effort needed to translate the data into relationships that create information from data - which will then assist users to create knowledge.

The real void in the technology and tools world today is that of Content Management Systems (document and rich media information) - to complement Database Management Systems (structured data). Unfortunately, there is no magic bullet - organizations cannot purchase good decisions and fundamental management of data out of a shrink-wrapped box.

Portals present opportunities for addressing the voids, but even then, they connect related sites and subjects which can unite users in learning and sharing environments. Despite the vehicle that a portal offers, it still does not feature rudimentary characterization and relationship management for information - and it does not address the "content" (meaning) that enables the road to knowledge management - because it does not contribute to information management.

The good news is that many emerging and existing IT capabilities allow an infinite array of "data" to be stored and shared. The bad news is that, no matter how many servers your organization has, you eventually run out of space and must "housekeep". Presently, there are not best practice methods for organizing data to keep it useful. The set of capabilities needed to create effective Content Data Management are not elusive nor must we wait on technology to afford to us - they are basic library skills that Data Managers possess which must be harnessed and brought to bear on electronic data.

Such an approach to Data Management must incorporate the following facets and areas (note that most of these attributes are not featured in "tools" - which is a big part of the problem):

• 1 Pragmatic and capable of catering for different needs;
• 2 Incremental - since funding is usually limited and is only available over a period of years;
• 3 Phased - because solving World Hunger usually fails, and a build on success strategy is almost always successful;
• 4 Non-technical - as users must be an integral part of the process, as well as being the owners, user, and stewards of the data;
• 5 DM education strategy which is formulated, implemented, and funded. Nobody said DM was free.
• 6 Coordinated from a centrally based oversight function which will ensure that corporate standards exist and are applied, and equally sponsored and composed of IT, DM, and CM;
• 7 Data is captured only once - to reduce the possibility of variance and redundancy
• 8 Definition of data must include textual descriptions and the structure (format, characteristics) of the data;
• 9 Data must have an identified owner - who alone is authorized to change or amend the data;
• 10 Good change management and version control to prevent two versions of data being present at the same time.

In summary, Content Data Management isn't high tech, it isn't automatic, and it isn't cheap - but it yields significant and impressive return on investment - which is the objective. It's only through demonstrated return on investment that the value equation is established and acceptance of the value-add of CM and DM becomes a realistic outcome.

Trackback(0)

TrackBack URI for this entry

Comments (0)

Subscribe to this comment's feed

Write comment