Here we provide a short list of the typical document management features. Content management shares all of these but adds on top. In this context document management deals with non-structured (office) documents, while content management is about structured content.
A DMS stores different type of documents, like docx, odf, pdf, images… etc, to bind different tools and processes. It’s similar how an operating system deals with files, where the type (file type) is indicated by the filename extension.
It’s driven by role based user management and the workflow. A user with certain role (or roles) can only have access right to a document at specific workflow stages. This can be read-only or write access.
During editorial work the user has to check out the document which usually puts an exclusive lock on the document, so others will only get read-only access to it. When the user is done, she checks it in, so then the lock gets removed.
Metadata is “data on data”. Data what’s not stored as part of the object content, instead it’s additive information. It can hold generic data, like the author signature, timestamp.. etc. or something very specific for the given content type.
Metadata is usually a list of name - value pairs. Values can be:
- simple type - so literals (number, string, date, ...)
- complex type - complete structures, like an address with several fields
Values can be controlled by a predefined vocabulary, which is a great step forward better consistency. Even better if the value itself is a “thing”, topic, node… whatever terminology we use, but it’ll bring us to the linked data world. More about this later.
Metadata takes a very important role in document and content management. It can be used to implement:
- or drive the publishing
Version handling is essential in document management. Often we want to keep all versions, the whole lifecycle of our documents. Even when the user deletes a document, the system should only flag it as deleted.
A check-out / in loop should create a new version, but versioning does not stop here. Documents have metadata and also linked together. We cannot simply version the content, but need to version also the metadata and the links together with the content.
Metadata values are bound to a certain document version. Also links. If we roll back to a previous version, we don’t want to break the links. We want to get back to the stage as it used to be.
Workflows in CMS control the different editorial and publishing tasks. It’s based on user roles and workflow statuses.
Workflow status examples:
User role examples:
Based on roles the workflow limits which user can move a document from one workflow state to another by doing a certain operation.
Larger teams might need to assign different roles to people with specific access rights and define a more formal workflow. Also mission critical content might require a well defined strict workflow for even smaller teams.
Besides text document, a DMS/CMS also stores media objects: images, videos, audio, slideshow files… etc. These binary files are usually linked to content objects and not directly embedded in. Of course publishing can embed them at a later stage.
The media objects have many metadata fields describing the media properties, author, copyright info.
One single media object can have several variants, what we call renditions. Most asset management system have automated processes to generate the necessary renditions.
Typical use cases:
- a video has different formats, because there is no single format which is supported by all web browsers
- a raster image has lower and higher resolution variants, while a lower resolution image is fine on screen, we need high resolution for paper production