ViViUse

Estabilishing a shared project vision is a central goal of the requirements engineering process. The effective koordination and communication of all relevant stakeholders' needs is necessary to enable a software development team to implement a suitable solution. This process of the requirements communication supports the development and negotiation of a shared understanding of a project's goals, plans, its status and its context.

Based on such a shared understanding, a shared project vision can arise which describes the size of the future system. This allows for an evaluation of the relevance of individual requirements and their meaning for the system. A shared project vision therefore supports the discovery of common ground for all stakeholders' activitie within the project. Therefore, the shared project vision significantly supports a project's success.

The achievement of a shared understanding can be supported by a proactive exchange of information between stakeholders and the development team. Current requirements engineering practices achieve this exchange through a textual presentation of the project vision in a specification. However, the drawbacks of natural language like multiple meanings of words or abstractions can result in undiscovered misunderstandings.

The medium of video is known to be one of the most efficient documentation options for the communication between people. Vision videos look to enable the use of these advantages for the communication of project visions among stakeholders by clearly presenting project visions in a simulated use case. The videos can be presented to stakeholders to create a base for discussions which themselves can lead to the achievement of a shared understanding.

The creation of vision videos was explored in the research program "ViViReq" between 2017 and 2020. Over the course of the project, the research group Software Engineering estabilished a quality model for vision videos as well as a production- and usage process. The research efforts culminated in an extensive guideline of 52 recommendations for the creation of vision videos. A summary of the creation process can be found in the following visualization:

Overall, the developed creation process consists primarily of the three phases of preproduction, shooting and postproduction. The individual steps of the preproduction in particular can already be perceived as a supplement to existing requirements analysis processes, since essential elements of the project vision must already be defined during the planning of a vision video. THis way, the creation of a vision video already supports the achievement of a common understanding between the stakeholders involved.

The following paragraphs present different types of vision videos that support the requirements engineering process.

Affordable Vision Videos

A central result of the research project "ViViReq" was the Affordable Video Approach. According to this approach, an elaborate production of a high quality video is not necessary for the creation of a shared understanding on a project vision. The video production can be done by project members without relying on the support of professional video producers. In videos created following the Affordable Video Approach, individual parts of the project vision that have not yet been implemented at the time of video production are often replaced in a simple manner. For example, in the following video clip, a parcel carrier points an ordinary TV remote at the barcode of a parcel to open a recipient's trunk. This simulates the reading of the code by a special device that has not yet been implemented.

Despite such a substitution, the project vision is still clearly recognizable and can be discussed by stakeholders. For example, such a simple video could be used to discuss whether the letter carrier's access should only apply to the trunk or whether the passenger door could also be opened if the trunk is already full.

Animated Vision Videos

Another way to create vision videos is to use animation software. Such software can be used to create vision videos that illustrate a project vision using simple presentation tools. In addition, no actual shooting with actors is necessary. Thus, there is no threat to the privacy of the participants and no props or recording devices such as cameras and microphones are needed. Furthermore, there is no risk of audio quality issues during outdoor recordings, which could be caused by wind or background noise.

PreVision Videos

Lastly, the Software Engineering research group also developed the idea of PreVision Videos. Such a vision video can be created based on the planning artifacts created during preproduction. These planning artifacts include a storyboard, which combines simple pencil drawings of the planned scenes with a description and dialog from a script.

For PreVision Videos, pencil drawings can be combined with dialogue recorded by stakeholders to provide a preliminary version of a later vision video. PreVision videos can provide a first glimpse of the project vision to be portrayed and are also suitable as a preparatory step for an Affordable Vision Video by presenting them to the performers. This gives the performers an initial sense of the context of each scene and the emotions portrayed in the video. The following video clip presents an example PreVision video.

Goal

The research project ViViUse deals with the application and use of previously created vision videos. A primary goal is the investigation of different usage scenarios in order to derive recommendations for their use in requirements engineering. Content-related aspects such as the viewing mode and context are evaluated. Furthermore, ViViUse includes the investigation of the acquisition and generation of feedback on the content of vision videos.

Usage of Emotions in Vision Videos

The thematic area of emotions will explore how emotions can be used in vision videos to support the elicitation of requirements.

One aspect to be investigated is what influence emotions included in the vision video have on the perception of the video content and thus on the requirements elicitation.

The focus of this research pillar is on subjective emotions perceived by humans. Automatic recognition of emotions in videos is not the focus of ViViUse.

Feedback

Vision videos are primarily suitable for creating a shared understanding of the vision between all relevant stakeholders. Disagreements between the project visions of different stakeholders can be found by discussing the contents of vision videos. Since these discussions consist of feedback from stakeholders, generating as much and as rich feedback as possible is desirable.

The Feedback research area explores various ways to gather feedback from stakeholders and to motivate stakeholders to actively provide feedback on the seen content.

For example, we will compare whether viewing vision videos within groups of people generates quantitatively more feedback than individual viewing. Secondly, we look to determine which viewing method, online or offline, results in more and higher quality feedback.

Interactivity in Vision Videos

One way to motivate stakeholders to provide larger and more information-rich amounts of feedback is to enrich vision videos with various interaction options.

For example, a vision video could consist of several paths that stakeholders can explore and evaluate. Surveys could then be used to select the paths that will make up the final vision. This would allow requirements engineers to gather both new and adjusted requirements in real time through the interaction process. A vision agreed on in this way provides a high level of shared understanding through prior discussions and consultation between stakeholders.

In the research area of interactivity, possible types of interaction in the context of vision videos will be investigated and evaluated for this purpose. We try to identify interaction techniques, or combinations of several techniques, that are particularly suitable for collecting feedback from stakeholders.

Current Publications

• Busch M., Shi J., Nagel L., Sell J., Schneider K. (2022) Vision Video Making with Novices: A Research Preview. In: Gervasi V., Vogelsang A. (eds) Requirements Engineering: Foundation for Software Quality. REFSQ 2022. Lecture Notes in Computer Science, vol 13216. Springer, Cham. Pages 251-258. doi.org/10.1007/978-3-030-98464-9_19
• L. Nagel and O. Karras, "Keep Your Stakeholders Engaged: Interactive Vision Videos in Requirements Engineering," 2021 IEEE 29th International Requirements Engineering Conference Workshops (REW), 2021, pp. 51-57, doi.org/10.1109/REW53955.2021.00014
• L. Nagel, J. Shi and M. Busch, "Viewing Vision Videos Online: Opportunities for Distributed Stakeholders," 2021 IEEE 29th International Requirements Engineering Conference Workshops (REW), 2021, pp. 306-312, doi.org/10.1109/REW53955.2021.00054