Naming the Pain in Requirements Engineering: Comparing Practices in Brazil and Germany

📝 Abstract

As part of the Naming the Pain in Requirements Engineering (NaPiRE) initiative, researchers compared problems that companies in Brazil and Germany encountered during requirements engineering (RE). The key takeaway was that in RE, human interaction is necessary for eliciting and specifying high-quality requirements, regardless of country, project type, or company size.

💡 Analysis

As part of the Naming the Pain in Requirements Engineering (NaPiRE) initiative, researchers compared problems that companies in Brazil and Germany encountered during requirements engineering (RE). The key takeaway was that in RE, human interaction is necessary for eliciting and specifying high-quality requirements, regardless of country, project type, or company size.

📄 Content

Naming the Pain in Requirements Engineering:
Comparing Practices in Brazil and Germany Daniel Méndez Fernández1, Stefan Wagner2, Marcos Kalinowski3, André Schekelmann4, Ahmet Tuzcu1, Tayana Conte5, Rodrigo Spinola6, Rafael Prikladnicki7 1Technische Universität München, Germany daniel.mendez@tum.de / ahmet.tuzcu@mytum.de 2University of Stuttgart, Germany stefan.wagner@informatik.uni-stuttgart.de 3Universidade Federal Fluminense (UFF) kalinowski@ic.uff.br 4Niederrhein University of Applied Sciences, Germany andre.schekelmann@hs-niederrhein.de 5Universidade Federal do Amazonas (UFAM) tayana@icomp.ufam.edu.br 6Universidade Salvador (UNIFACS) rodrigo.spinola@pro.unifacs.br 7Pontifícia Universidade Católica do Rio Grande do Sul (PUC-RS) rafael.prikladnicki@pucrs.br

Requirements Engineering (RE) constitutes an important success factor for software development projects, since unambiguous and stakeholder-appropriate requirements are critical determinants of quality and productivity [1] [2]. One of the problems we still face, however, is that it is difficult to find proper empirical figures that could demonstrate particular success factors in RE [3]. Empirical studies are inherently difficult for RE with its high variability and long-term feedback cycles. Yet, this makes it difficult to justify the choice and adoption of particular RE practices, as we lack sufficient knowledge about which problems we face in RE, what relevance they have, and what their causes could be. Unfortunately, the lack of evidence in RE leads to the problem that decisions on RE practices are mostly driven and justified by conventional wisdom only.
In response to this problem, we initiated the Naming the Pain in Requirements Engineering (NaPiRE) initiative1 hosted under the umbrella of the International Software Engineering Research Network2. NaPiRE constitutes a globally distributed family of surveys on the state of RE practices including problems practitioners experience in RE as well as their causes and effects [4]. The resulting knowledge base shall allow us to get a better understanding about the relevance of particular problems and to establish means to mitigate problems in the long run. Here, we show selected results from the survey edition conducted in 2014/15 by comparing two data sets obtained from two different countries: Brazil and Germany. We intentionally select those countries as their industrial practices show several different characteristics. The data results from responses from 118 Brazilian companies of which we select 74 as they where sufficiently complete, and from 54 German companies of which we select 41 respectively. We use the data to illustrate commonalities and differences in the problems practitioners experience and discuss what we can learn from them. In the following, we directly discuss key results from the survey while a richer description can be taken from our complementary material provided on the project website1 I. WHAT IS THE STATUS QUO IN REQUIREMENTS ENGINEERING PRACTICES? Before discussing the revealed problems in detail, we take a look at selected facets that characterise the practical environments of our respondents (with rounded numbers). Table 1 shows the size distribution of the participating companies for Brazil and Germany. It is possible to observe that our sample includes German

1 www.re-survey.org 2 isern.iese.de companies of larger sizes. 44% of the responding German companies employ more than 2,000 people in contrast to 21% in Brazil. Brazilian responses show with 61% a tendency towards small and medium-sized companies (250 employees and less), which contrasts to Germany with only 29%. In both countries, respondents have at least 3 years of experience in their role (BRA: 75%, GER: 81%). Finally, the organisational role of our responding companies shows a tendency to product development (BRA: 79%, GER: 56%) while the role of a contractor is underrepresented in Brazil (5%) in contrast to our German responses (37%).

Brazil Germany 1–10 employees 11 3 11–50 employees 15 6 51–250 employees 17 3 251–500 employees 5 2 501–1,000 employees 3 3 1,001–2,000 employees 5 6 More than 2,000 employees 18 18 Invalid (missing) answers 3 0 Table 1. Size distribution of participating companies.

Some characteristics we see in the data indicate further differences on how projects are conducted and people work together in the countries. The first difference constitutes the chosen process model. In Brazil, we can see a tendency to agile methods such as Scrum (66%) or Extreme Programming (13%), which is slightly higher than in Germany (58% and 2%). This tendency becomes, however, more evident if we consider that German respondents stated to use in 44% of the cases a waterfall model (BRA: 30%), which we can explain by the contracting role where multi-staged bidding p

This content is AI-processed based on ArXiv data.