=Paper=
{{Paper
|id=Vol-2822/paper5
|storemode=property
|title=A Framework for Integrating Web Accessibility Requirements in Agile Methodology
|pdfUrl=https://ceur-ws.org/Vol-2822/p5.pdf
|volume=Vol-2822
|authors=Darliane Miranda,João Araujo
}}
==A Framework for Integrating Web Accessibility Requirements in Agile Methodology==
https://ceur-ws.org/Vol-2822/p5.pdf
A Framework for Integrating Web Accessibility
Requirements in Agile Methodologies
Darliane G Mirandaa , João Araujoa
a
Faculdade de Ciência e Tecnologia (FCT), Universidade Nova de Lisboa, Portugal
Abstract
In the Software Engineering area, Web Accessibility is gaining more space, establishing itself as an
important quality attribute. However, considering the agile context, accessibility requirements continue
to be neglected. Existing methods and models tend to be overly complex for non-technical stakeholders
to properly express their goals and communicate them to the agile team. We propose the development
and assessment of a framework to support the elicitation and analysis of web accessibility requirements,
in an agile context. The framework consists, primarily, of an accessibility requirements model defined
by extending (systematically) a goal-oriented model (the iStar framework).
Keywords
Accessibility, Agile, Goal-oriented Modeling
1. Introduction
The concerns about web accessibility and the discussion about its importance for a wide variety
of users, including those with disabilities, starts in 1994 when Tim Berners-Lee founded the
World Wide Web Consortium (W3C). A couple of years later, a group of members launched the
Web Accessibility Initiative (WAI), an initiative to guide how to handle accessibility in the face
of the web’s growing complexity [1]. Nowadays, websites and web applications (combining
with assistive technologies) have a higher impact on improving persons with disabilities’ access
to socio-cultural, educational, and economic activities [2].
Among the W3C guidelines, the Web Content Accessibility Guidelines (WCAG) 2.1 covers
a wide range of recommendations for making web content more accessible [3]. WCAG is pri-
marily intended for web content developers (page authors, site designers, etc.), web authoring
tool developers and to others who want or need a standard for web accessibility. There are
other relevant standards similar to WCAG 2.1, such as the Section 508 of the Rehabilitation
Act of 1973 [4]; the Guidance on Software Accessibility (ISO 9241-171:2008) [5]; the Infor-
mation Technology-User Interface Accessibility (ISO/IEC 29138-1:2018) [6]; the International
Telecommunication Union (ITU-T) [7]; the European Mandate 376 (EN 301 549) [8].
However, the issues involving web accessibility specifications, implementation, and evaluation
are still a concern among IT professionals [9].Although these guidelines are internationally
recognized, they are often difficult to interpret and understand by developers or software
SEDES’2020: Software Engineering Doctoral Symposium, September 08, 2020, Online Conference
" dg.miranda@campus.fct.unl.pt (D.G. Miranda); p191@fct.unl.pt (J. Araujo)
0000-0002-9372-8724 (D.G. Miranda); 0000-0001-5914-1631 (J. Araujo)
© 2020 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
engineers who end up neglecting their content [9]. In the context of agile methodologies,
approaching this type of requirement by development teams is also a challenging task, as there
is often a lack of knowledge about the subject and, sometimes, it is only considered in the final
phase of the projects [10].
Based on this scenario, we propose the development and assessment of a framework to
support the elicitation and analysis of web accessibility requirements in an agile context, since
the vast majority of development teams and projects now embrace those methodologies [10].
The framework aims to automate the reuse of these requirements, to support the generation of
artifacts such as user stories and personas, and consequently, improve the visualization and
communication between stakeholders, and support the analysis of the possible impact of these
accessibility requirements on the other system’s features.
The remainder of this paper is organized as follows. Section 2 presents the state-of-the-art
web accessibility approaches in the literature. Section 3 presents the research objectives and
methodological approach used in this PhD thesis. Section 4 presents an overview of the past
work and preliminary results. Section 5 presents the future work and expected results, following
by the conclusion in Section 6.
2. State of the Art
After conducting a systematic mapping study (see section 3.1), it was feasible to identify some
relevant works that addressed the problem of accessibility in web-based systems.
Snider et al. [11], presents a study conducted by IBM researchers, which carried out an
analysis of the main sources of data on accessibility used in the company to build an ontology
(EACO), which was used to improve the performance of a question-answering accessibility
conformance chatbot and others systems used in the company to help IT professionals to get
more information about accessibility requirements.
Oliveira et al. [12] proposed a method to elicit web accessibility requirements, followed
by a correlation catalogue that summarizes the identified conflicts among its Non-functional
requirements (NFRs), and a checklist to help engineers during the design and programming. To
accomplish this goal, the authors provided the OmnesWeb tool to support the elicitation activity.
A case study revealed that teams using OmnesWeb gathered more requirements, produced all
the required artefacts, and spent a quarter of the time less than the time spent by teams that did
not use the tool.
Moreno et al. [13] proposed a domain-specific metamodel for web accessibility to support
engineers in modeling aspects of accessibility, using the Model Driven Architecture standards,
such as meta-object facility and Model-Driven Development. The authors developed the Acces-
sibility for Web Applications (AWA), a methodological framework for the development process
of accessible web applications.
The Reichling et al. [14] approach used the User-Centered Design (UCD), which relies on an
interactive process of three main phases: analysis, design, and evaluation. The authors proposed
to integrate an agile approach to UCD by intertwining the two and applying the UCD principles
during the pre-sprint phases to enhance a user-centered approach.
Martín and Yelmo [15] presented a systematic approach to the development of an accessibility
evaluation tool, which was oriented by the Unified Software Development Process (USDP). The
authors proposed artifact templates to deal with the different disciplines managed by USDP: a
business model and a domain model to kick off the requirements elicitation, a use-case model
to represent functional requirements, an analysis model to reflect an intermediate view of the
architecture, and a deployment model to describe the physical bindings.
Younas et al. [16] proposed an elicitation guideline for NFR, for agile software development.
Since in agile methodologies, story cards are used for the elicitation of functional requirements,
the authors proposed another user story card for eliciting NFRs, including accessibility require-
ments. The process involves the requirements elicitation through interviews, identification of
the software type, the identification of requirement type through a glossary, the selection of
experts in the area related to NFR type, list of candidate NFR, validation with specialists and
validation with the end-user.
Scott et al. [17] presented the VERITAS, a framework that simulates and systematically
analyses how users with different impairments interact with the use of ICT products and
services. The framework is supported by three core tools: (i) VerGen, to specify the nature of the
impairments to be simulated in terms of a virtual user; (ii) VerSEd-GUI, to define and configure
a series of actions to test on the user interface of the product or service; and (iii) VerSim-GUI,
where the various impairments are simulated to reproduce the experience of an impaired user.
Although these studies present solutions to improve the elicitation, development, and evalua-
tion of accessibility requirements, they do not consider the development context adopted by
companies, the artifacts used by them, and the type of users involved in the software project.
Our framework aims to fill this gap and also including the use of goal-oriented models, making
this process simple and effective.
3. Research Objectives
One of the main objectives of this thesis is to develop a framework to support professionals
(with or without expertise in the web accessibility domain) in the process of elicitation, analysis,
development, and validation of accessibility requirements following the agile routine. Another
objective is related to the engagement of stakeholders in the requirements engineering process,
improving communication and involvement between them. The framework aims to improve the
specification and use of accessibility requirements through a goal-oriented model in accordance
with the accessibility guidelines proposed by W3C [3]. Also, the framework aims to facilitate
the visualization of requirements according to each project and with each users’ special needs ,
as well as allowing the analysis of the impact of these requirements on the other features of the
system. To achieve these objectives, the following phases need to be addressed.
3.1. Systematic Mapping Study
This phase is concerned with the delimitation of the scope of this thesis. For this purpose,
a systematic mapping study was conducted to identify the existing approaches (including
methods, processes, techniques, frameworks, and tools) in the literature that could be used by
IT professionals to elicit, develop and evaluate accessibility requirements in agile projects to
offer a comprehensive overview of these methods, their characteristics, applicability, possible
limitations and a research roadmap with the open issues requiring future research.
3.2. Planning and design
This phase is concerned with the diagnosis of current accessibility requirements approaches
adopted by agile teams, with the definition of the features to support the design and development
of the framework, and with the design of the experiment to be conducted with stakeholders.
First, it is necessary to analyze selected companies that adopt the agile methodologies to identify,
through questionnaires and interviews, how web accessibility requirements are approached
and what is the impact of using these approaches on the final product – we will follow the
principles of a Straussian Grounded Theory (GT) [18]. To analyze this impact, accessibility tests
will be carried out on the products developed through the Achecker evaluation tool [19]. This
first step is necessary to (i) identify the main problems in the approaches used by companies to
elicit accessibility requirements and find solutions to fix them through our proposed method;
(ii) identify and translate these problems to a set of features to be used in the design and
implementation of the framework. After the identification of the features, a catalog of web
accessibility requirements will be defined and represented by extending the iStar framework,
which supports goal-oriented modeling of socio-technical systems and organizations. The
extension will be carried out systematically adapting a process for extending iStar, called
PRISE [20]. Some studies show that goal-oriented models have been used as an effective means
to capture interactions and trade-offs between requirements, helping stakeholders to have a
graphical view of the team’s goals and how to achieve them [21, 22, 23, 24, 25]. The next step is
providing a set of artifacts to be used by the agile team and other stakeholders in the process
of elicitation of accessibility requirements. Also, it will be carried the analysis of the impact
of accessibility requirements on other system’s features and perform the necessary trade-offs.
To accomplish this, we will use multi-criteria decision methods (MCDM) that can be used to
identify possible conflicts (semi-)automatically and help solve them [26].
Regarding the experiment with stakeholders, we need to plan it systematically, starting with
the definition of a hypothesis (e.g., it is possible to use the framework to support IT professionals
and stakeholders in the elicitation of accessibility requirements). The expected deliverables for
this phase are: a) a report with the results of the GT; b) a report with the framework features,
the impact analysis and design planning for the experiment with stakeholders.
3.3. Implementation and evaluation
The implementation of the framework allows not only the elicitation process of the accessibility
requirements, but also the (semi)automated generation of artifacts that can be used in an agile
context, such as user stories and personas, according to the WCAG guidelines proposed by W3C.
The framework will be developed as a web platform that will provide a catalog, an accessibility
requirement model based on the iStar framework (in conformance with the notation of the
goal-oriented model and the artifacts produced by agile teams), a checklist with the correlations
between these requirements with other system’s feature and the impact of the requirements in
the final product. The expected deliverable is the framework including editor, catalog, and the
generator of user stories, personas and other features, which can be consulted by the entire
agile team, at any stage of development.
For the evaluation of the proposed framework, we will carry out case studies in companies
that adhere to the agile methodologies, that already considered the accessibility requirements
in their projects and that work with web-based systems. Based on this evaluation, we will be
able to analyze how satisfactory or not our method was in the process of eliciting requirements
and improving communication between stakeholders. The framework will be evaluated by IT
professionals and accessibility specialists in terms of accessibility, usability, and applicability to
support the elicitation process. Also, it will be carried out an evaluation with a group of people
with special needs, to identify whether the framework complies with WCAG parameters.
Also, one needs to evaluate the usability of the notation of the models approaches from the
perspective of ordinary users, by using biometric equipment like EEG scanners, eye-trackers,
and cognitive processes. These techniques will also be useful to assess the effort spent by users
in order to create, change, and understand different models. We will follow the process described
by Jedlitschka et al. [27].
4. Past work and preliminary results
As mentioned in Section 3.1, a systematic mapping study was conducted to find in the literature
approaches that focused on solving web accessibility issues. The mapping study has shown
that many of these approaches are not widely disseminated in industry, and other identified
techniques have not been empirically validated, which may increase the chances of discontinuity.
Another point to be highlighted is that some of these studies only focused on one type of
disability, which continues to be an accessibility concern. Given that the complexity and
qualities of the web content are constantly changing, frequent validations of the techniques
and methods adopted are crucial to guarantee the minimum of accessibility. Finally, we also
found that many papers focus on accessibility evaluation processes, i.e., to verify that websites
meet the accessibility requirements defined by guidelines such as WCAG. Validation methods
were also identified during the development phase, which is critical to ensuring that the system
is patched and improved before it is made available to the end-user.
After analyzing the results, we also identified that the agile approach, adopted by most
companies in the world, was not considered in most of the studies. Those methodologies are
broadly used by improving project quality and reducing development costs. Although the
adoption of those methodologies intending to deliver high-quality products, we identified that
many organizations still adopt a waterfall approach to accessibility testing, leaving it to the
end of the project lifecycle. This information was crucial to delimit the scope of our proposed
framework.
5. Future work and expected results
The thesis is currently on the initial steps mentioned in the section 3.2, regarding the planning
and design phase (see Appendix A). The results of this thesis will contribute to software
development in general, and requirements engineering in particular, with a support framework
for improving the accessibility requirements’ process with early detection of accessibility
problems in agile requirements models. The expected contributions are: (i) a framework to
facilitate the process of identifying, analyzing and implementing accessibility requirements and
to support the generation of artifacts as user stories and personas; (ii) a goal-oriented accessibility
requirements model suitable for agile routine; (iii) an accessibility catalog; (iv) an identification of
issues on the currents methods adopted by companies to incorporate accessibility requirements
in agile projects; (v) an evaluation of the framework from the perspective of common users and
requirements engineers, by using biometric equipment like EEG sensors and eye-trackers [28],
combined with process metrics such as effort, in the production, understanding, and modification
of the accessibility requirements model (measured in terms of time to complete tasks and of
achieved correctness), in order to classify engineers’ and user’s experience and progress during
modelling tasks.
6. Conclusions
This paper presents a broad overview of the doctoral thesis work plan that is being carried out
in the context of accessible web systems. After conducting a systematic mapping study, it was
possible to identify a gap to be filled: the need to develop a framework that can improve the
process of eliciting accessibility requirements considering the agile development. The paper
also presented the past and future work, including the conduction of a Grounded Theory to
identify and understand the current elicitation, development, and evaluation process of this type
of requirement in agile teams, followed by the development and validation of the framework
using well-established techniques adopted in the literature.
From the several reports created in each one of the phases, the most relevant contributions will
be submitted to major scientific conferences, seminars, and journals. The produced framework
will be published in an open software repository so that requirements engineers can access it
and use it in their projects.
References
[1] R. Rutter, P. H. Lauke, C. Waddell, J. Thatcher, S. L. Henry, B. Lawson, A. Kirkpatrick,
C. Heilmann, M. R. Burks, B. Regan, et al., Web accessibility: Web standards and regulatory
compliance, Apress, 2007.
[2] ITU, Global consultation report on icts and disability launched at the un high level meeting
on disability and development, 2013. URL: https://www.itu.int/en/action/accessibility/
Pages/hlmdd2013.aspx, [Online; accessed 21-May-2020].
[3] L. Henry, Web content accessibility guidelines (wcag) overview, WWW Consortium
(W3C) (2018).
[4] G. S. Agency, Section 508.gov: Gsa government-wide it accessibility program, 2020. URL:
https://www.section508.gov/manage/laws-and-policies, [Online; accessed 24-July-2020].
[5] ISO 9241-171:2008(E), Ergonomics of human-system interaction. Part 171: Guidance on
software accessibility, Standard, International Organization for Standardization, Geneva,
CH, 2008.
[6] ISO/IEC 29138-1:2018(E), Information technology- User interface accessibility- Part 1: User
accessibility needs, Standard, International Organization for Standardization, Geneva, CH,
2018.
[7] Supplement 17 to ITU-T H-series Recommendations Guide for addressing accessibility in
standards, Technical Report, 2014. URL: http://www.itu.int/rec/T-RECH.Sup17-201411-I/en,
[Online; accessed 24-July-2020].
[8] L. Martínez, M. Pluke, Mandate m 376: New software accessibility requirements, Procedia
Computer Science 27 (2014) 271–280.
[9] S. Trewin, B. Cragun, C. Swart, J. Brezin, J. Richards, Accessibility challenges and tool
features: an ibm web developer perspective, in: Proceedings of the 2010 international
cross disciplinary conference on web accessibility (W4A), 2010, pp. 1–10.
[10] I. Choudhury, Agile methods for engineering, in: Management for Scientists, Emerald
Publishing Limited, 2019.
[11] S. Snider, W. L. Scott II, S. Trewin, Accessibility information needs in the enterprise, ACM
Transactions on Accessible Computing (TACCESS) 12 (2020) 1–23.
[12] R. Oliveira, L. Silva, J. C. S. P. Leite, A. Moreira, Eliciting accessibility requirements an
approach based on the nfr framework, in: Proceedings of the 31st Annual ACM Symposium
on Applied Computing, 2016, pp. 1276–1281.
[13] L. Moreno, P. Martínez, B. Ruiz, A mdd approach for modelling web accessibility., in:
IWWOST@ ICWE, Citeseer, 2008.
[14] M. Reichling, S. S.-s. Cherfi, Integrating accessibility as a quality property in web devel-
opments, in: IEEE 7th International Conference on Research Challenges in Information
Science (RCIS), IEEE, 2013, pp. 1–12.
[15] Y.-S. Martín, J. C. Yelmo, Guidance for the development of accessibility evaluation tools
following the unified software development process, Procedia Computer Science 27 (2014)
302–311.
[16] M. Younas, D. Jawawi, I. Ghani, R. Kazmi, Non-functional requirements elicitation guideline
for agile methods, Journal of Telecommunication, Electronic and Computer Engineering
(JTEC) 9 (2017) 137–142.
[17] M. J. Scott, F. Spyridonis, G. Ghinea, Designing for designers: Towards the development
of accessible ict products and services using the veritas framework, Computer Standards
& Interfaces 42 (2015) 113–124.
[18] J. Corbin, A. Strauss, Basics of qualitative research: Techniques and procedures for devel-
oping grounded theory, Sage publications, 2014.
[19] G. Gay, C. Q. Li, Achecker: open, interactive, customizable, web accessibility checking, in:
Proceedings of the 2010 International Cross Disciplinary Conference on Web Accessibility
(W4A), 2010, pp. 1–2.
[20] E. J. T. Gonçalves, T. Heineck, L. de Oliveira, J. Araújo, J. Castro, The prise tool to support
istar extensions., in: WER, 2019.
[21] F. Wanderley, J. Araujo, Generating goal-oriented models from creative requirements
using model driven engineering, in: 2013 3rd International Workshop on Model-Driven
Requirements Engineering (MoDRE), IEEE, 2013, pp. 1–9.
[22] H. Haidar, M. Kolp, Y. Wautelet, Goal-oriented requirement engineering for agile software
product lines: an overview, Louvain School of Management Research Institute Working
Paper Series, Louvain, Belgium (2017) 1–36.
[23] R. Mesquita, R. Nascimento, M. Lucena, L. Souza, Using the istar framework for planning
and monitoring sprints in scrum projects: An experience report., in: iStar@ ER, 2019.
[24] A. Sen, S. Jain, An agile technique for agent based goal refinement to elicit soft goals in
goal oriented requirements engineering, in: 15th International Conference on Advanced
Computing and Communications (ADCOM 2007), IEEE, 2007, pp. 41–47.
[25] J. Lin, H. Yu, Z. Shen, C. Miao, Using goal net to model user stories in agile software
development, in: 15th IEEE/ACIS International Conference on Software Engineering,
Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), IEEE, 2014,
pp. 1–6.
[26] R. A. Ribeiro, A. M. Moreira, P. Van den Broek, A. Pimentel, Hybrid assessment method
for software engineering decisions, Decision Support Systems 51 (2011) 208–219.
[27] A. Jedlitschka, M. Ciolkowski, D. Pfahl, Reporting experiments in software engineering,
in: Guide to advanced empirical software engineering, Springer, 2008, pp. 201–228.
[28] S. C. Müller, T. Fritz, Stuck and frustrated or in flow and happy: Sensing developers’
emotions and progress, in: 2015 IEEE/ACM 37th IEEE International Conference on
Software Engineering, volume 1, IEEE, 2015, pp. 688–699.
A. Work Plan
Figure 1: Timeline of the research.
B. Poster
Figure 2: The poster with an overview of this research.