Design of Web Application for Document Generating and Publishing

The goal of the current study is to design a service for automatic generation and deployment of technical documentation. Nowadays, the task of quickly delivering up-to-date technical documentation to users is of current importance. Being the main source of information about the product, technical documentation affects different fields, including software development, interaction with end users and knowledge management. Thus, the field of technical documentation is actively explored, especially the automation of document generation that would make the technical writers’ job easier. The main current problem of technical documentation is missing and outdated information. In most cases, the reason behind it is a multi-stage process of generation and deployment that includes interacting with many different programs. This process requires a lot of time and effort, as well as mastering of many programs. It is error-prone, especially to human errors, and it prevents technical writers from quickly updating the documentation and keeping it up-to-date. The study was designed according to Contextual Design framework. We started with theoretical analysis of the field and common solutions addressed to the problem. Then we interviewed technical writers to understand the current workflow and the main user needs. The latter includes remote document generation and deployment, remote management of documentation base, generating documents in different formats from one source, and assembly report with a list of corrupted topics. To address user needs, we described a new document generation process and design low-fidelity and high-fidelity prototypes. The low-fidelity prototypes are then tested by technical writers with the “first-click testing” method to analyze the understanding and user evaluation of the new features. Test results show that the majority of the users succeeded in solving test tasks and appreciated the new features. Among main limitations of the current study, we see narrow target audience and small number of users interviewed and tested. This is because the service was developed according to needs of a certain technical writers team, their documentation workflow and programs used. Also, we admit that the low-fidelity prototype was not evaluated by usability experts. As for following development of the study, we consider conducting a proper usability evaluation and finishing high-fidelity prototype. After that, we could expand the service functionality for it to be used by other documentation teams and adapt it to other documentation workflows.