Content silos are an unfortunate byproduct of the move toward electronic documents. In paper form, a document could pass from one business process to the next, albeit very slowly. As organizations implemented new technologies to manage these documents, they used various solutions to address them, which in turn created rigid boundaries to contain documents across the organization. Soon, content became duplicated and stored in various locations, making it difficult to know which documents were accurate. Today, some enterprise content management (ECM) and content services platforms (CSPs) are working to address this problem through the use of content federation.
Content federations look to tie content, or documents, within an organization in a way that makes it accessible to all business processes. As content federations evolve, three different approaches are beginning to emerge. In a recent report titled “The Five Key Trends For 2018 That Shape How We Manage Enterprise Content,” Forrester Principal Analyst Cheryl McKinnon identifies two approaches: interoperability and metadata-centric. There is also a third approach employed by some vendors called the “syndication” approach.
The Interoperability ApproachThis pull approach to content federation searches target repositories for documents to include in the federating repository. This can be a full repository federation or rules can be used to select only portions of the source repository. These subsets may be one specific type of document, such as contracts, or a point in a document's life cycle, such as when it becomes a record.
At this point, a new object is created in the federating repository from the original source. The object can include all or a subset of the original metadata as well as any new additional metadata that is added. The content itself is not transferred. Instead, a pointer is created in the federated repository to the object in the source system.
CSPs, like Alfresco and Nuxeo, are taking this approach to content federation.
The Metadata-Centric ApproachThis is a crawl-based approach to content federation. With this method, it's a user’s responsibility to identify the content that will be federated from a target repository (usually by navigating the source repository through the user interface). Like the interoperability approach, a new object is then created in the federating repository, along with its metadata, and a content pointer is directed at the source repository.
This is the approach taken by CSP M-Files.
The Syndication ApproachThis approach looks at integrating the systems that use content rather than systems that store content. These repository platforms do not address content inside other vendor repositories. It really shouldn't be called a federation approach but rather a new service—content syndication services. These platforms make content accessible in line-of-business applications, like SAP, and allow for both read and write capabilities.
This is the federation approach adopted by OpenText.
Enabling FederationsWhen looking at content federations, there are two components to the integration. While traditional ECM and CSP vendors are offering federation capabilities, the underlying technology is often being delivered by third parties through either OEM or partner arrangements. Repository-to-repository integrations require either an open approach to partners, like we see with Microsoft SharePoint, or a neutral third party. Vendors like Simflofy, SkySync, and Xillio have emerged to fill that gap.
Bridging content silos is something that organizations have been looking to do for over 20 years. In the past, this meant developing and maintaining custom integrations. As CSP vendors address this requirement, more organizations will be able to bridge these content silos and create new process and platform efficiencies.