Professionals in requirements engineering (RE) use methodical frameworks and guidelines in a variety of ways to structure and support their tasks. Which frameworks are particularly important? And what specific expectations and needs do RE professionals have of them?
These questions were the subject of a discussion at the 2024 GI FG RE annual meeting, a conference of the “Requirements Engineering” (FG RE) special interest group in the German Computer Society (Gesellschaft für Informatik, GI). The discussion was organized by the working group “Software Requirements Frameworks” (AK REFrame) of the FG RE. This article documents important content and results.
These questions were the subject of a discussion at the 2024 GI FG RE annual meeting, a conference of the “Requirements Engineering” (FG RE) group in the German Informatics Society (GI e.V.). The discussion was organized by the working group “Software Requirements Frameworks” (AK REFrame) of the FG RE. This article documents important content and results.
Discussion of methodical frameworks in RE
Methodical frameworks and guidelines play an important role in requirements engineering (RE): They are the subject of training and further education (e.g. IREB CPRE); they guide the activities of RE practice in the form of process models, methods and techniques (relevant international standards, process models in companies); they structure RE as an object of research (definitions, categorizations and comparisons of RE techniques), and much more.
Twenty-five people took part in the discussion, 14 of whom were currently working in the industrial practice and 11 in research. Five of them had been familiar with both areas from their careers. The one-hour session was moderated by Andreas Birk. Two other members of the AK REframe, Andrea Wohlgemuth and Andrea Herrmann, provided support from the plenum.
A brainstorming session opened the discussion with the question: What are important requirements frameworks from the point of view of the participants? This question sparked a lively discussion. The moderator collected the answers on moderation cards. The following table summarizes the results.
| Brainstorming Results (English Translation) | Brainstorming Results (Original Results in German) |
|
Concrete methods and techniques: User Stories & everything around them… Personas Vision Modeling, especially: conceptual models Modeling as documentation, e.g. in interviews Modeling as a working technique (“spontaneous modeling”) Different model types; in the sense of conceptual areas of modeling such as data, processes, interaction Language Templates “Everything beyond language that gives shape to requirements” Concrete examples; also in addition to models; principle: abstract vs. concrete Prototypes Videos require clear process definitions, e.g. information flow Maps with documents; including temporal and other dependencies Review techniques in general Peer Review (external view) |
Konkrete Methoden und Techniken: User Stories & alles darum herum … Personas Vision Modellierung, speziell: Konzeptuelle Modelle Modellieren als Dokumentation, z.B. in Interviews Modellierung als Arbeitstechnik (“spontan Modellieren“) Verschiedene Modelltypen; i.S.v. konzeptuellen Bereichen der Modellierung wie Daten, Prozesse, Interaktion Language Templates “Alles über Sprache hinaus, was Requirements Gestalt verleiht“ Konkrete Beispiele; auch ergänzend zu Modellen; Prinzip: abstrakt vs. konkret Prototypen Videos brauchen klare Prozessdefinitionen, z.B. Informationsfluss Landkarten mit Dokumenten; inklusive zeitliche u.a. Abhängigkeiten Review-Techniken allgemein Peer Review (Außensicht) |
|
Procedural principles: Iterative approach using methods such as PDCA, agile sprints or recurring reflection Fail Early, i.e. the right balance between insufficiently mature and too much effort Process: “mini V’s” in sprints |
Vorgehensprinzipien: Iteratives Vorgehen nach Methoden wie PDCA, agilen Sprints oder mit wiederkehrender Reflexion Fail Early, also die richtige Balance zwischen zu wenig ausgereift und zu viel Aufwand Prozess: „Mini-V’s“ in Sprints |
The following were mentioned at the very beginning:
- Agile User Stories
- Personas
- Models and modeling
Each point represents a variety of methods and techniques. User stories, for example, understood as a framework of associated methods and techniques, include the role-feature-reason template, INVEST checklist, backlog management and much more. Modeling is a basic attitude (“model often and spontaneously”, “a picture is worth a thousand words”) and also the collection of different model types (data, processes, interaction, etc.).
During the session, further frameworks were added, in particular:
- Vision / vision statement
- Language templates
- Use of concrete examples, also in addition to models
- Use of concrete artifacts such as prototypes and videos
- Maps of documents and their dependency relationships
- Review techniques, especially the frequent use of uncomplicated peer reviews
Various principles of procedure were also mentioned:
- Iterative approach (including methods such as PDCA, agile sprints, frequent reflection)
- Fail Early (moving forward quickly and lightly, and correcting as needed)
- Agile sprints with all aspects of the V-model compactly embedded
Overall, techniques for presentation and documentation before requirements were clearly predominant compared to pure procedural and process aspects.
Further discussion and next steps
The brainstorming session led to a lively discussion with an exchange of experiences. Key points and statements were continuously noted on flipcharts. They concerned the content of methods and techniques and their application, as well as expectations regarding the form of frameworks.
The results of the brainstorming session are a good basis for further activities of the AK REFrame: They show what practitioners intuitively consider a framework (user stories, personas, modeling, etc.). It also became clear which aspects are considered particularly important, for example:
- Artefact-related techniques for the presentation and handling of requirements are more central than pure processes and procedures.
- It must be possible to apply frameworks in a lightweight and flexible way.
- Frameworks must be understandable and applicable, especially for non-RE experts.
The AK REFrame will continue to build on these results. It will announce and document further activities on its community website here.
More Blogs
Workshop Overview: REFraming Elicitation, REFrame Workshop 2/2023
A major activity of the REFRame working group and community was the REFrame 2/2023 workshop at the International IEEE RE 2023 conference on September 4, 2023, in Hannover. The workshop's title and key topic area was "REFraming Elicitation", which refers to frameworks...