Document overview

Documents are instances of document classes

Document classes contain essential properties and methods that comprise documents. Individual documents are called instances of their class or subclass. For information about creating document classes, see Creating a document subclass.

All document classes are assigned system-created properties, such as Document Title and Creation date. Custom properties are created and assigned to specific subclasses by system designers or advanced users, and are used to further identify the documents for later retrieval.

Documents are versionable

A user, or a process like a workflow, can check out a document to edit it. When you check in the updated document, both the new version and the previous version are preserved. Each time that you check in the document, another version is created for that document. See About versioning for more information.

Documents have content

The content of a document is composed of pieces of information such as word-processing files, spreadsheets, graphics, and images. See Document content for more information.

Network-wide search and retrieval

You can assign titles to your documents by using simple, easy-to-read phrases instead of with file names or ID numbers. Users and automated processes can search for documents based both on their descriptive properties and their content. Regardless of the user or the document location, complex path commands with coded document names are not required. Client applications are designed to make it easy for users to find and use important documents and other objects.

Documents can be filed in folders

You can file documents in folders for easy display and organization, just like you are using Microsoft Windows Explorer. However, while Microsoft Windows enables only one instance of a document, FileNet® P8 enables documents to be filed in many different folders. An action that is taken on any of these document instances is made on all the others. For example, DocumentA could appear in FolderA and FolderB. If DocumentA is checked out from FolderA, it is also marked as checked out in FolderB. See About folders for more information.

Documents can be associated with lifecycle policies

You can pass a document through a predetermined series of states and transitions that are called a lifecycle. See Document lifecycles for more information.

Documents can be workflow definitions

A workflow definition is a document that describes the instructions, paths, and queues available to a workflow. Workflow definition documents are versionable, containable, and can have a lifecycle that is associated with them.

Documents can be attachments in a workflow step

You can attach documents and folders to a step in a workflow. Depending on your access rights, you can

• View the document as read-only.
• Edit the document.
• Save the document as a new version as part of the execution of the workflow step.

Documents can subscribe to server-side events

You can trigger actions such as delete or check-out by using the event subsystem. See Subscribing to events for more information.

Documents are securable

All documents are designed for easy application of security templates, as the document version goes through predefined states. See IBM® FileNet P8 security for full security information. Also, see About versioning for more information.

Documents can be grouped into compound documents

A compound document consists of a single parent document, containing zero or more child documents. Any child document can be the parent of a separate compound document structure.

Compound documents are referenced by component relationship objects or "links" that

• Represent a parent to child document relationship.
• Define the structure of a compound document.
• Specify rules for binding the parent to a child component.

For more information, see About compound documents.

Path-based retrieval of folders and documents

Content Platform Engine supports path-based retrieval of folders and documents in order to support web-based access or content through Universal Resource Locators (URLs) as well as to allow relative URLs embedded in documents. Retrieving a document by using a URL returns only the first content element even if the document version has multiple content elements.

Documents are retrieved by the name assigned to the reference object and not by the value of the document title property. The reference object (formally called the ReferentialContainmentRelationship object) is what actually causes a document object to appear in a folder.