V-Modell XT and ITSM for Confluence

Working collaboratively as team creating software-based products requires information flow. Not only need coding software developers to communicate the architecture and various quality aspects of the software artifacts to non-coding team members, but developers also need to know about things like business priorities, contract information, and market opportunities. This communication needs cannot be solely answered with a markup language and a source code management server. While this tool selection is a valid solution for developers or technical writers, it cannot be the whole story, especially for large teams. Software coding is only a small amount of the work to be done for a successful product. And I hate to admit - as a software developer loving to code - coding is not the most important one.

Communication is king. Creating documents nobody is interested in is a waste of resources. While only implemented features of a product are able to support users getting their work done, there is also the need to move information from one stakeholder's brain to another. Information is required to support stakeholders to make the right decisions. There is also the need to store information in a tool more reliable for recalling than the brain. 

Information workers handle different kinds of information. But for software project we often work with the same set. Architecture decisionwork instructioncontractinterface documentationpatternsstakeholder descriptioncodes, test plan, build plan, persona, user story, use case, meeting minutes, and so on and so on. Well, this set is quite large. Not every artifact is be a document in a wiki. It may be a bug report in a issue management system or an executable test case run on a continuous delivery server. In case a document is actual to be created to provide a given information, there is no need to reinvent the wheel for each document instance. Templates leverage the organization of the actual information. Team members should fetch the most appropriate template for the communication task at hand, add the information in a form their readers expect and push it to a location where interested parties can find it easily. This is about concentrating on the readers' demands and their expectations.

At smartics we work with Confluence as a team collaboration platform. Each member writes a journal which is readable by every member of the team. This is because records are easier to handle than documents since there is no need to update records. Albeit some information is better organized as documents than a flow of events organized by time and some tags. This is when Doctypes for V-Modell®XTDoctypes for Service Management (ITSM, Information Technology Service Management) and other doctype add-ons enter the stage. If we need to compile information, we choose from the experience of other people to organize it. With space and page blueprints organizing information with projectdoc Toolbox in Confluence is quite easy.

In the end there is (still) often the need to export information from the wiki in form of static documents, mostly PDFs, to be sent to external stakeholders. Still this transformation takes quite a large amount of time (and we wished that everyone was working directly with Confluence), but we are working on it to get it done more quickly by increasing the amount of automation. Fortunately there is still room for improvement!