The knowledge and information contained in documents can have a significant impact on the efficiency and productivity of your company.
The Knowledge Provider (KPro) is a general and comprehensive infrastructure that provides cross-application and cross-media document management technologies. The Knowledge Provider is a component of SAP Web Application Server and can be integrated into any SAP application.
When dealing with documents, the Knowledge Provider differentiates between the following information entities:
· Content
· Content index
· Administrative data
· Model data
A text file exists for a section of documentation. This text file contains the content. The file exists independently of whether or not other information about this section exists. Administrative data could be the person responsible, the creation date, and the name of the documentation to which the section in question belongs. The model data defines, among other things, what attributes the document has. The existence of the attribute “author” is therefore part of the model data. The content index is a secondary representation of the content. The content index makes it possible to have content searches (for example, full-text search). You do not always have to define the content index separately. Sometimes it may be more efficient to duplicate parts of the administrative data (for example, author) in the content index or in the content (self-describing documents).
In the context of the Knowledge Provider, the term “document” is used in a broad sense. It can refer to text, video, audio, software programs, websites, and controls. In other words, the Knowledge Provider definition of the term “document” is much wider than the classical sense of the word. In the Knowledge Provider, a document can be any self-contained document-like object.
In the context of the Knowledge Provider, there is no fixed definition of the term “document”. The Knowledge Provider allows its client applications the freedom to define and create their own specific definitions of document models. An application-specific content model contains in particular the following elements:
· Classes for relationships
· Classes for logical documents
· Classes for physical documents
Content models function as the structure and the architecture of document-like objects within the application-specific context. This means that every client application can use its own specific metadata descriptions, while still being able to exchange information with other applications.
Application-specific content model for a magazine article
In this example, there are two classes, one for a logical document, and one for a physical document with specific properties. There are also a number of classes for relationships. The instances of the relationship classes CONTENT VERSION, FORMAT VERSION, and LANGUAGE VERSION make it possible to administrate any number of different versions according to different criteria (modified content, translation, format conversions).
The Knowledge
Provider provides both simplification and automation functions in several
areas of document management. A central part of this is a two-level approach
to document management, consisting of intension and extension. The logical
properties of a document, including meaning and purpose, are referred to as
intension. The actual physical
representation of the document itself is the extension. Intension and
extension are represented in the Knowledge Provider by two administration
objects: logical document and physical document.
The intension
of this section is the conceptual background of the Knowledge Provider.
(However, this statement does not make any assumptions about how this will be
achieved.) This section is available both in German and in English. It is also
available in electronic form, for example, in the form of HTML Help or plain
HTML, and can also be printed. Therefore, the physical representation of the
document exists in various forms, that is, extensions. Documents in
isolation are worthless. Every document-like object has relationships to other
objects. These relationships can be either implicit or explicit. The Knowledge
Provider administrates not only document-like objects themselves, but also the
relationships between objects. The Knowledge Provider also allows relationship
classes to be defined, which means you can identify and classify the types of
relationships between documents. Relationships and
references between documents are dynamic. If you make a change to a document
in the Knowledge Provider, this does not mean that a costly chain reaction of
manual changes to other documents has to be made.
If a note is
created which relates to a specific document, the link between the note and
the document can be made explicit by means of a relationship. In the same way,
relationships can be used to display hyperlink paths in documents, thus making
explicit the interrelationship between the documents linked by
hyperlinks. If various
documents are interconnected by several relationships (for example,
hyperlinks), these documents are no longer self-contained. This problem
applies to all types of relationships, not just hyperlinks, and can occur
between different documents, and between documents and business objects. The
Knowledge Provider provides a solution to this problem by referencing logical
documents (for example, logical hyperlink) and by implementing context
resolution at runtime. This process is called late binding. This means that
the physical document version that best matches the specified context is not
selected from the set of physical documents belonging to the specified logical
document until runtime. The version is selected on the basis of context
information (user context or business context). Late binding, then, is the
interface between intension and extension. Both information managed by the
Knowledge Provider and other information from the Knowledge Provider client
application can be involved in the resolution process. The network of
relationships remains intact, even if documents are translated, if document
content is edited, or if a document is converted to a different format. This
all helps to avoid expensive and time-consuming extra work. Logical
Hyperlink: Context Resolution The Knowledge
Provider provides a powerful versioning concept. Knowledge Provider versioning
supports multiple versioning based on a number of different criteria.
Versioning can control language versions and format versions, among others.
For example, a document can exist in German, English, and French, and in both
PDF and HTML format. The Knowledge
Provider allows equal access to all document-like objects via HTTP, regardless
of the underlying medium and the application-specific design of the website.
The check-in and check-out mechanism ensures that authors cannot make
conflicting changes to the same object at the same time. Text Retrieval & Information Extraction
offers a range of retrieval strategies, including integration of search
engines to allow full-text searches. The Knowledge
Provider provides document storage services. This means that SAP-based
document management functions can be combined with a wide range of storage
media. See also Content Management Service.
Intension and Extension
Relationships
Late Binding and Context
Resolution
Versioning
Equal Access and
Retrieval
Content Storage Media