Lost on Earth—the mere data in your head of places you have been to in your life and the pure information you receive upon seeing a nice waterfall are not built together to form knowledge that you could use to decide where to go and what to do.
Sitting at your computer looking into your daily work, you may often encounter situations such as this one: you have prior information on the customers, you get an email which informs you about recent changes in his credit position, but you have not catch the full picture, simply, because the environment is not knowledge-oriented.
We have realized that internal data, document and application management should be built together and form a knowledge management suit which directs your staff during their everyday routine work and provides capabilities for creative users to enhance their possibilities in information extraction and knowledge building.
Supposing your organization shall successfully finish the introduction of a knowledge management system,
The basis of knowledge management is a well-built and maintained information store, which is guaranteed by an introduced and used document management system.
Further functionality can be expected from the integraton of knowledge management with data mining and OLAP tools, which are able to collect and summarize information from large amounts of loosely structured data.
It will be a dramatic change in your company that the users shall work in an event-driven way, instead of the former customer driven or scheduled way. These events shall not only come from a customer or from a deadline task, but also from information collected from the ‘net, events from applications, escalation of workflows of subordinates or colleagues etc. The most appealing feature that at the same time as these events occur, the context necessary for the following step shall be put on your screen: at the right place at the right time and for the right person.
Most of the work can be done for knowledge management if the document management system is created in one step with the introduction of the knowledge management approach.
The first step is the specification of goals, which is similar to the document management goals. However, an additional issue is the definition of knowledge owners.
Your key-users shall be interviewed, the daily work of your regular users shall be monitored. Their information production and use shall be tracked, as well as the sources and users of the information when is produced or used by external entities.
The static and dynamic classification of documents shall be analyzed. The former is used to categorize backlog, the latter for enabling the tracking of data during modifications.
When there is a public or trivial classification system which well describes a certain aspect of the knowledge of the company or a certain department, we suggest the use of these categories. Such apply those categories. Samples:
Besides the common classifications, the interviews must provide sufficient information to create the company’s taxonomy.
Regarding the dynamic classification, our approach is future-oriented: expecting a certain result we just back track to see in what way will it lead to the fulfilment of expectations. Once a typical workflow point is detected where a certain category of documents is needed, we shall create an information node to which the documents can then be classified. This classification thus becomes status-oriented.
Not everything is written into a document, because not everything is allowed to be written down and some things ‘are known’ without writing them down. The information that is important but is not contained by the document is called ‘metadata’.
Metadata helps us e.g. finding proper position in a workflow, or adding some extra comments or notices, keywords, dates etc.
The use of metadata is highly application-dependent, however even in a manually-driven system it may serve a lot of needs.
The document analysis interviews should collect what metadata is used explicitly or tacitly at workflow decision points.
Typical queries and stored queries shall be collected which can be added to certain decision points. These are similar to SQL queries, which serve data from a database, but now instead the document base or certain portions of the Internet could be queried.
These queries may be customized and added to personal sites in an ad-hoc manner, but it may be worth to collect what are the most important external information sources for the company that provide searchable data.