Modern KM tools and techniques for collaboration
Search and regulatory compliance
Another key facet of incorporating a data marketplace in data catalogs that reinforces collaboration is the ability to search through it. Oftentimes, when organizations upload taxonomies, business glossaries, and other semantic information, they can attach certain types of catalogued assets that become searchable according to the particular business concept or term with which they’re associated. According to Laine, “If there’s a particular document or data contact that’s aligned within your business glossary, you can search for the larger category or term associated with it, and it would bring up those results.” This feature can inform a number of KM use cases, particularly regulatory compliance ones in which officers are attempting to see documents related to a particular dimension of a specific compliance concept.
The ability to centralize this functionality within a catalog in which users can comment on and score the usability of cataloged assets effectively democratizes this process throughout the organization. “From a compliance perspective, you can go in and start your research based off a policy, or a procedure that I need to do off a policy, or based off business rules,” Laine specified. “You can say, ‘Show me all the data around this privacy policy.’ Do I have things encrypted? Do I have business rules that tell me I can or cannot pass off this information?” Other cataloging functionality relies on regular expres- sions and additional metadata models to identify assets according to terms for specific compliance mandates, sensitive data, and more. “You can apply RegEx to metadata, meaning looking at table or column names, and you can have RegEx looking at the content of columns, meaning the values of the columns,” Zwicker stipulated. “As you build these patterns, you’re basically building a library to say what you’re looking for.”
Data fabric architecture
The data fabric architecture has become particularly renowned for its propensity to foster enterprise collaboration while centralizing data, assets, content, and organizational knowledge. There are two forms of this type of architecture: physical and logical. A physical data fabric places all these enterprise resources within a single framework—usually a cloud-based platform with processes for managing all aspects of data and enterprise knowledge. A logical data fabric, in contrast, “leaves data in its systems of record but virtualizes it and brings it together,” Glaser revealed. “It presents that single pane of glass, that irrefutable version of the truth of what enterprise data is at any point of time, in real time.” This architecture has substantial consequences for KM, particularly in relation to conceptual data modeling and collaborations among business units, sources, and use cases for business concepts and definitions.
A data fabric enables organizations to leave sources where they are, including “existing schema, existing ports, and protocols for getting at it, and brings it together where a low-code developer has designed a conceptual data model and abstracts that data,” Glaser indicated. When defining concepts like customer, for example, that may be pertinent for a case management application, organizations can easily cull from a number of respective sources—without moving their contents—to inform that definition. Thus, whether parts of that data pertain to Salesforce, business glossaries, customer documentation, a document repository, or a digital asset management (DAM) hub, “you would design the customer data object as a single entity, sourced with its constituent parts from all these different systems brought together, so you can have one view of your customer without having to mutate the source of the record,” Glaser said.