software.”
- @GartnerTedF, Ted Friedman, Gartner Analyst for Data Management and Integration
I was excited and inspired upon reading this tweet. I think Ted is pointing out that the real key to any enterprise information solution, rather than any innovative client presentation or visualization layer, is about getting and managing data as usable information. Rather than a fancy GUI of the application/software, the right data is the the ultimate goal, and the path to that is a metadata-driven approach. Of course I may going off in another tangent than what Ted was intending. I do not believe he has yet provided more detail, but I look forward to his elaboration.
At any rate, I will continue with my interpretation and ideas. My view is that metadata is key to this, but it’s only part of the story. That’s because mapping concept X in System 1 to concept Y in System 2, using metadata is only a “dumb connection.” And the mission is more than about data; the true goal is information. So really, we need more than even just metadata. The mission is about CONNECTING systems and CONTEXTUALLY retrieving ONLY appropriate content. That's quite a challenge today since you need to map, in a point-to-point fashion nearly every permutation of metadata connections in order to answer any ad hoc question that might arise, or to deliver any contextual content. That’s a lot of spaghetti integration to manually code and certainly it's impossible to maintain with any realistic budgets today.
Unless an ontology is used for the specific business domain(s) at hand.
For those unaware, ontology in the information sciences domain, according to Wikipedia “is a formal representation of the knowledge by a set of concepts within a domain and the relationships between those concepts. It is used to reason about the properties of that domain, and may be used to describe the domain.” Therefore, an ontology goes far beyond the concepts simple metadata, or even a dictionary or taxonomy.With an ontology, metadata, business rules, processes, behaviors, relationships are embedded together. And the ontology acts as an overlay for existing systems without being invasive to those systems. Therefore, the ontology now becomes the heart of powerful new applications, in the spirit of Ted's quote above. Of course a client layer GUI is required to interact with the ontology which then interacts with the source systems (tools and applications). But most of the logic, business rules, and metadata are indeed within the ontology.
All sounds great theoretically, but can a solution really exist? Actually it already does. zAgile Teamwork illustrates just such a solution. With an ontology, using semantic web standard technologies, based on an industry standard metamodel for software engineering, zAgile Teamwork integrates not just the metadata but the concepts of Projects, Requirements, Tasks, Isses, Test Cases, Check-ins, Customer Support Cases, and more. Therefore, ad hoc queries, faceted searches, and contextual content can all be served in a consistent fashion.
With zAgile Teamwork, powerful new metadata driven applications are possible thanks to the software engineering ontology and framework under the covers. The real magic is indeed this middleware of the solution and the GUI can be any new or existing application or tool. For example using a template within a wiki, now Requirements can be captured in a structured way inside the wiki and tied to Projects, Issues, and Tasks in an issue managment system. Likewise Test Cases can be created in the wiki and similarly and automatically tied to appropriate concepts. And for example, do you want to know what Customer Cases in Salesforce.com are connected to a Task, or Issue in JIRA or a Requirement in the Confluence wiki? Yes, there’s a Connector for that too. In fact, all you have to do is add a Connector for whatever tool or application you have, and you can INSTANTLY integrate all knowledge of all connected systems. Too good to be true? Learn more here and give zAgile Teamwork a spin today and see for yourself.
Comments