See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325773569

Sustainability in software engineering

Conference Paper · December 2017

DOI: 10.23919/SustainIT.2017.8379798

CITATIONS READS
36 3,228

3 authors, including:

Patricia Lago Francesco Osborne

Vrije Universiteit Amsterdam, Amsterdam, Netherlands The Open University
367 PUBLICATIONS 6,975 CITATIONS 124 PUBLICATIONS 1,875 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Francesco Osborne on 08 January 2019.

The user has requested enhancement of the downloaded file.

 Sustainability in Software Engineering

Nina Wolfram Patricia Lago Francesco Osborne

Vrije Universiteit Amsterdam Vrije Universiteit Amsterdam Knowledge Media Institute
The Netherlands The Netherlands The Open University, Milton Keynes, UK
[email protected] [email protected] [email protected]

Abstract—The intersection between software engineering re- state of the art by addressing the following main research
search and the problems related to sustainability and green IT question (RQ):
has been the subject of increasing attention. In spite of that,
we observe that sustainability is still not clearly defined, or How is sustainability currently defined in the context of Soft-
understood, in the field of software engineering. This lack of ware Engineering?
clarity leads to confusion about e.g. what is relevant to measure
or the research implications over time or space. To answer RQ, we analyzed a wide range of primary
studies and employed the systematic mapping study research
This paper provides an overview of how the research so method as defined by Peterson et al. [17]. This type of
far has defined sustainability, and how this definition has been secondary study gives an objective framework for structuring
used to guide which research areas. To this end, we carried and analyzing the research domain and is a suitable means for
out a systematic mapping study for selecting, classifying and
identifying trends in research.
analyzing relevant publications. In this study, we investigate
which knowledge areas and which time scope of sustainability
effects are mostly targeted in scientific research. Our analysis II. M ETHODOLOGY
shows research trends and discusses gaps to be filled.
In order to ensure a traceable and reproducible process,
Keywords—Sustainability, Software Engineering, Systematic the following section will define the protocol used in our
Mapping Study. systematic mapping study.

I. M OTIVATION A. Research Questions

In the past years, climate change and an increasing aware- For the sake of a more structured approach for information
ness for social inequality have made sustainability a growing extraction, our main RQ has been refined into the following
concern in a wide variety of disciplines. The discipline of detailed research questions:
Software Engineering (SE) is no exception. A widely quoted RQ1: Which sub-disciplines of SE, if any, does the definition
definition characterizes sustainability as the ability to meet relate to? SE is a wide field encompassing, according to
“the needs of the present without compromising the ability of the Software Engineering Body Of Knowledge (SWEBOK,
future generations to meet their own needs”[5]. However, when [4]), 15 sub-disciplines. Mapping existing research efforts onto
applied to SE, it is unclear what specific aspects of Software these sub-disciplines will provide an overview of which areas
Engineering and the resulting software systems it refers to. As have witnessed growing research efforts on sustainability, and
of now, there is no agreed-upon definition for sustainability where potential for additional research lies.
in the context of Software Engineering, and existing research
approaches rely upon different notions, ranging from e.g. RQ2: Which time scope is considered in the definition? Sus-
sustainability of the SE process [7] itself to specific software tainability is a concern for the future. However, the perceptions
applications designed to support sustainability goals impacting differ as to how far the future influence of an activity in
other disciplines [8]. sustainable development extends. Sustainability measures have
not only direct primary effects, but also indirect secondary or
Sadly, sustainability is regularly used in research papers as tertiary effects in the long term. Therefore, we are interested to
a vague buzzword, where the authors’ notion of the term is establish how far SE researchers have considered these effects
not stated clearly and has to be extrapolated from the text. in their notion of sustainability.
For a better understanding of existing papers and to give
researchers a motivation to provide a clearer definition of RQ3: How did the definition develop over time? Especially
their approach w.r.t. sustainability, a clear classification of all during the last years, the topic of sustainability has seen a
aspects of sustainability would be desirable. However, before strong increase in research activity in software engineering.
any attempt to establish such a framework for sustainability We are interested in how this may have influenced the way
research, it is necessary to establish an overview of current sustainability is understood and if there has been any shift in
research, in order to achieve a better understanding of which focus of research over time.
factors in software engineering are relevant for sustainability
and how effects propagate throughout the software engineering B. Search Strategy
lifecycle.
The primary studies identified in this mapping study were
To this end, this work provides an overview of the current retrieved by combining the following two search strategies.

978-3-901882-99-9 c 2017 IFIP

 The first strategy used an automatic extraction approach
(see left-hand side in Fig. 1) [10] supported by the Smart
Topic Miner (STM) [12], which is the tool used by Springer
Nature for classifying conference proceedings in the field of
Computer Science1 . STM builds on the Rexplore system2
and characterizes publications according to the research topics
from the Computer Science Ontology (CSO), a large-scale
OWL ontology of research areas. It does so by associating
to a scientific paper all the concepts in CSO whose label
is found in the title, the abstract, or the keyword set, and
all super-areas and synonymous of those. For example, a
publication associated with the term "xADL" would also be
tagged with higher-level topics such as "Software Architecture Fig. 1. Overview and numbers of the search and selection process
Description Languages", "Software Architecture", "Software
Engineering", and "Computer Science". CSO currently in-
cludes 17K concepts and was generated by running the Klink- C. Study Selection
2 ontology learning algorithm [11] on the Rexplore dataset,
which consists of about 16 million publications, mainly in the From the publications returned by the two search strategies,
field of Computer Science. For the purpose of this analysis, after a removal of duplicates, a number of inclusion and
we used only the branch describing the Software Engineering exclusion criteria was applied to the dataset to eliminate those
domain3 , which was reviewed by five domain experts. We titles that were formally not acceptable or not relevant. Figure
extracted the primary studies by querying a Scopus4 dump 1 shows an overview of the different steps performed on the
from the 1980-2013 period, and selecting all publications data sets.
which 1) were tagged by the STM with the CSO concept 1. Elimination according to formal criteria: In order
"Software Architecture", and 2) contain in the title, in the to be admissible for evaluation, publications had to fulfill the
abstract, or in the keyword set at least one of the following following inclusion criteria:
keywords:
I1 - Published in English
Ecologic Ecological Ecology Sustainable Sustain- I2 - Peer-reviewed
ability Biodiversity Ecosystem Ecosystems sustain- I3 - Available as full-text.
able development environmental impact climate Set 1 of publications from the manual SLR had already
change environmental protection green been filtered according to these formal criteria and contained
a total of 115 publications. Set 2 consisted of 920 papers
This automatic extraction strategy has the obvious ad- that were automatically extracted by the automatic extraction
vantage of being able to process a very large dataset of approach, of these 850 publications remained after elimination
publications with no human intervention. However, it presents according to these criteria.
some limitations. First, the underlying Scopus dataset does not
include papers published after 2013, therefore we could not 2. Merging and duplicate removal: Merging the two data
apply this technique on the most recent publications. Secondly, sets and removing the duplicates between both sets left 954
the terms in the papers can sometimes be misleading, and papers to be processed further.
thus a further human intervention is necessary to filter out 3: Elimination according to relevance: The publications
not relevant publications. were then analyzed for their relevance to the topic of sustain-
ability in Software Engineering. For this purpose, the following
The second search strategy used the primary studies from a inclusion and exclusion criteria were defined:
previous literature study (see other input at left-hand side of
Fig. 1), a systematic literature review (SLR) on sustainability E1 - No reference to sustainability or Green IT: Some
in SE. Manually performed on the ACM Digital Library, IEEE of the chosen keywords, although often connected
Xplore and SpringerLink, this SLR did include all publications to the aspect of sustainability, are also used in
that were indexed in the used libraries at the time of the other contexts in SE. For example, “environment"
extraction in 12/2016 and hence complemented the results of may be used in the context of environmental sus-
the automatic extraction strategy. tainability, but frequently refers to the “software
execution environment". This resulted in a large
In particular, it used the following search string, imple- number of false positives that were not relevant
mented in each database’s specific format: to sustainability or green IT.
E2 - Sustainability in the context of software use:
allInTitle: (sustainab* OR green* OR ecolog*) AND Studies referring to software as a tool that is
software used to achieve sustainability in other disciplines,
such as agriculture, education, or supply chain
1 Demo available at rexplore.kmi.open.ac.uk/STM2_demo/ management, were explicitly excluded.
2 http://technologies.kmi.open.ac.uk/rexplore/ I4 - Sustainability in the context of software engi-
3 Available at http://rexplore.kmi.open.ac.uk/data/SE-ontology.owl neering: The focus of our study lies in the cre-
4 https://www.elsevier.com/solutions/scopus ation process of software and its sustainability
 properties and the way it influences sustainability
properties of the software product.
4. Related work: Ten extracted publications pursued a
research goal similar to ours, i.e. to analyze sustainability
research in software engineering. Accordingly, we decided to
treat them separately and discuss them in the related work
section.
5. Extraction of sustainability definitions: Finally, all
publications were analyzed to elicit the notion of sustainability
as meant by the authors. All papers containing a clearly stated
definition of sustainability, either directly or as a quote, were
included. Publications where a definition and focus could be Fig. 2. Distribution of publications over the time scopes of sustainability
derived implicitly from context, application or examples were
also included. All cases that did not allow to derive a clear
definition or merely seemed to be using “sustainability" as a provide reference models for the sustainable development
buzzword were discarded, which resulted in a final count of of software by giving an overview of sustainability aspects,
168 primary studies. The extracted definitions were analyzed providing a sustainable software engineering lifecycle model,
for a number of characteristics that allowed us to map the or defining metrics. Two general research papers reflect on
differences between the various research works in an accurate current definitions of Green IT and sustainability, whereas
way. The overall research focus of each primary study was one manifesto created by a large number of sustainability
mapped to a Knowledge Area, as defined in the SWEBOK [4]. researchers from software engineering aims to further the
Apart from that, the author’ sustainability approach was char- discussion and awareness of sustainability in SE research.
acterized according to its focus on either the process or the One case study reflects on changes and variations in the
product, as well as the considered time scope for sustainability development of sustainability over time by taking a public
within the software engineering lifecycle. We also extracted ICT project in India as an example. And, finally, one paper
the order of effects taken into account by the authors as either establishes a plan for integrating sustainability education into
immediate, enabling or systemic and noted down which of the the curriculum of Software Engineering students.
environmental, economic, social and technical dimensions of
sustainability were considered. Section III-B goes into detail
as to how exactly this classification was performed. B. Classification
Due to space limitations, a complete list of all papers in the The 168 primary studies including a clear definition of
final selection including their extracted classification is made sustainability were categorized according to a number of cri-
available online5 . teria, in order to give an overview about prevailing notions of
sustainability and Green IT in SE research. Before answering
III. R ESULTS the research questions we will now describe the strategy and
rationale for this classification.
The following section will first present an overview of the
related work we identified among the extracted papers. We will 1) SWEBOK Knowledge Areas: Since one of the goals is to
then first go into more detail on the classification we used to identify areas of increased activity of sustainability research,
analyze the single publications’ definition on sustainability and we assigned all publications to knowledge areas as they are
then make use of the data gathered with this classification in defined by the SWEBOK[4]. These were chosen according to
order to address the research questions we defined in section the research area focus of the paper and, as not all publications
II-A. clearly belonged to a single knowledge area, up to 2 knowledge
areas per paper were assigned.
A. Related Works 2) Focus on Process or Product: SE research on
This study is by far not the first endeavour to improve the sustainability has two focus areas: On one hand, it focuses
understanding of sustainability in Software Engineering. As on making the engineering process more sustainable, for
mentioned previously, the literature search also resulted in a example by applying agile tactics or by sustaining an open
total of 10 secondary studies that had similar research goals source development community. On the other hand, there is
with respect to sustainability in SE, i.e. providing a basis for effort to render the output, i.e. the product and its direct and
further research by compiling definitions, defining common indirect effects, more sustainable though measures such as
research goals or giving an overview of the development of energy-efficiency optimization.
the research area. This section gives an overview of these
publications. 3) Time scope: We are especially interested to know, which
Three of them are systematic literature studies aiming to timeline researchers are taking into consideration when target-
give an overview of the state of research on sustainability ing sustainability in their research. We therefore categorize
in SE or on SE for Sustainability. Two publications aim to publications according to the time scope that is considered
for sustainability in research, specifically the phase within
5 https://tinyurl.com/y94x8ygp the software engineering lifecycle during which the intended
Fig. 3. Distribution of the time scope of sustainability according to each Knowledge Area

effects on sustainability manifest. Depending on their focus, papers are only focusing on sustaining software on a technical
we assigned the following time scope categories: level without explicitly taking other dimensions into account.
• Development process: papers that attempt to make 5) Effects: Software Engineering not only directly impacts
the development part from the first designs to the sustainability in its various dimensions, but also indirectly. In
release of a software product more sustainable by order to determine to what extent authors take into account
guaranteeing development speed and quality. these indirect effects, all publications were also categorized
• Testing process: papers that exclusively deal with the into immediate (direct), enabling (indirect) and systemic im-
sustainability of the testing phase as a subprocess of pact, according to the order of effects targeted by the authors.
the development phase.
• Maintenance process: papers that aim to make the C. Time Scope of Sustainability
maintenance of a finished product more sustainable A major interest in this mapping study was the timeline
by, for example, reducing the personal, monetary and that authors of papers on sustainability in software engineering
environmental cost. took into consideration during their research. Figure 2 shows
• Runtime: papers that aim to achieve a positive effect how the papers are distributed over the different categories,
of software on sustainability during the time of its with Runtime being the most prominent, mostly due to a
execution. large number of papers focusing on software energy efficiency.
• Lifecycle: papers that do not specifically focus on any Apart from this, research considering sustainability during the
of the aforementioned categories, but look at sustain- complete software lifecycle was most common.
ability effects during the whole software lifecycle.
D. Sustainability by Knowledge Area
As an example, energy-efficiency optimization would man-
ifest during software runtime, while optimization in the soft- When looking at how the selected papers are distributed
ware architecture would make the maintenance phase of soft- over the SWEBOK knowledge areas, as presented in figure 3,
ware more sustainable. Of course these different time scopes a number of hotspots become immediately obvious. Software
overlap as well: development and testing happen during the Requirements, Software Design and the Software Engineering
maintenance process too, albeit in a limited extent. Papers were Process feature the biggest number of papers. Computing
assigned to a single category, in which the major focus lay. Foundations, which covers SE areas such as networks and
algorithms, is another prominent knowledge area.
4) Sustainability Dimensions: Sustainability is commonly
divided into different dimensions, depending on the type of Upon further inspection of how papers in these Knowledge
effects a product or process has. Normally considered are the Areas are distributed in terms of time scope, SE Process shows
environmental, economic and social dimensions of sustainabil- a stronger focus on the Software Lifecycle compared to other
ity. Apart from this, we also add technical sustainability as areas, whereas Computing Foundations has a strong focus on
a fourth dimension to this list, considering that a number of sustainability during Software Runtime.
Fig. 4. Distribution of papers over the different sustainability dimensions

Fig. 6. Number of publications by year. Note that there are less publications
for the years after 2014 due to limitations in the search strategy as detailed
in section II-B

Fig. 5. Mapping of papers according to their focus and the order of effects
the authors are targeting

E. Sustainability Dimensions
In figure 4, we are presenting an overview of how the
definitions in all included publications are distributed over
the various dimensions of sustainability. The vast majority of Fig. 7. Time Focus of SE research on sustainability plotted over time
publications are considering environmental sustainability if not
as a main concern, then at least as a partial concern. In contrast
to this, technical or social sustainability are only part of the One goal of this study is to examine to what extent the
considerations in a minority of papers. definition of sustainability in research has changed over time.
Figure 7 shows how publications over the years relate to the
time scope within the SE lifecycle that they are addressing.
F. Sustainability Focus and Effect Of course given the low number of papers before 2008 is
hard to give a comprehensive interpretation of the development
In figure 5, the focus of a publication on sustainability of the time scope between 2008 and 2016. Still, papers
of product, process, or both is displayed in context with the treating sustainability effects during software runtime have
order of effects targeted by the author. Generally, the bulk of only become prominent after 2009, while other scopes had
papers is solely focusing on sustainability of the product and already been addressed already earlier.
the immediate effects it has on sustainability. In relation to
that, publications that also consider the SE process, have a
tendency to focus as well on secondary and tertiary effects. IV. D ISCUSSION
In the following, the previously presented results of this
G. Sustainability research over the years mapping study will be discussed and used to answer the our
research questions. We will then discuss the extent to which
Figure 6 illustrates how the number of publications on the results are subject to threats to validity.
sustainability in SE has strongly increased over the years. The
numbers for the years 2014-2016 are only considered prelim- A. Research Questions
inary, due to the limitations in the used data set mentioned in
II-B. However, a trend can be seen and it becomes obvious RQ1: Which sub-disciplines of SE, if any, does the defi-
how interest in sustainability has increased. nition relate to? By analyzing the distribution of publications
over the SWEBOK knowledge areas (see Figure 3) we observe towards environmental sustainability. Mitigation
a clear trend: the four top-most researched areas are also those of this bias, however, is difficult: Green IT is
that have the most pervasive/broadest influence on the software a discipline that is clearly defined as addressing
lifecycle. These are (1) the SE Process (which includes the environmental sustainability concerns. Analogous
largest cluster of studies in sustainability research encom- keywords for the other dimensions of sustainabil-
passing the complete lifecycle, and hence providing both a ity (like social or technical) are harder to define
framework and a basis for research on sustainability); the areas or would be too vague, hence failing to filter
of (2) Software Requirements and (3) Software Design (which publications in a reasonable manner.
are both activities early in the SE lifecycle and therefore have Limitation of automatic extraction: The data
a high impact on the resulting product); and (c) the area of dump on which the extraction was performed
Computing Foundations (which provides a basis for further was restricted to publications until 2013 included,
research on Green IT and sustainable data centers). which becomes evident in figure 6. This weak-
ness has been balanced by including the primary
The remaining areas witness a significantly lower number
studies of a previous SLR as another source of
of studies. Among them, we think that potential for further
publications after 2013.
research surely exists in the area of Software Quality, es-
Classification of papers: The classification de-
pecially since at least some partial aspects of sustainability,
scribed in section III-B was executed by the first
such as resource efficiency, are already established as an
author alone. As such, it may contain a personal
aspect of software quality. Similarly, Software Testing, SE
bias. To mitigate this risk, the other authors have
Management, Models & Methods and Professional Practice
been consulted in case of doubt and in randomized
are areas where already a certain amount of research exists,
checks.
but in practice practitioners in these areas are largely unaware
of sustainability.
V. R ELATED WORK
RQ2: Which time scope is considered in the definition?
The majority of papers either focus on sustainability during Since this literature study also resulted in a number of re-
the complete Software Lifecycle or at Software Runtime only. lated research publications, these will now briefly be reviewed
In comparison to this, other time periods during the Software here to complement the results of this paper.
Lifecycle are less represented in research, this might be caused This is by far not the first literature study with a focus
by the fact that sustainability during the development or on sustainability in software engineering. Penzenstadler et al.
maintenance process is often not explicitly defined: it comes [14] performed a systematic literature review on the state of
under the name of other software quality aspects such as main- sustainability research in SE and divide the focus of research in
tainability, efficiency or adaptability, which describe properties a similar manner as has been done in this paper for the targeted
that influence a system’s sustainability during certain lifecycle time scope of research. The defined aspects attribute sustain-
phases. ability research to the development process, the maintenance
RQ3: How did the definition develop over time? Sustain- process, system production or system usage. A systematic
ability as a concern in SE is, all in all, a fairly recent topic. mapping study performed later by the same authors [16] maps
The majority of analyzed papers date from after 2010, after publications to knowledge areas, but identifies different areas
which the topic has seen a strong increase in attention. An than in this paper as hotspots, with Models & Methods and
examination of the distribution of time scopes for sustainability Software Design being more prominent. This might be due to
of these papers shows that sustainability during Software the fact that Green IT was not included as a topic specifically.
Runtime has become a major focus of research. Time-wise, Notable venues, authors and guidelines have been identified in
this coincides with the popularization of cloud technology, and a mapping study by Berntsen et al.[3].
ever-growing data-centers world-wide have created a need for A research paper by Penzenstadler [13] takes up the con-
more resource-efficient solutions, a development that has led cept of aspects for defining and delineating sustainability in SE
to awareness for Green IT and energy-efficient data-centers. as presented in the previously described SLR, and elaborated
Other than this increase in research activity, there is no clear on these aspects. A comprehensive description of the state
change in focus over time visible in the data collected from the of the art with an overview about established descriptions,
papers considered for this research. The papers might have to definitions and models, with a focus on the Green IT aspect
be analyzed in more detail than the approach of a systematic of sustainability, is given by Calero and Piattini [6].
mapping study allows to get a clearer picture. Apart from this.
all related work with a similar goal is still fairly recent, so it The frequently referenced GREENSOFT model publishes
remains to be seen how these efforts to bring structure and by Naumann et al. [9] aims to provide a structure and strategies
attention to sustainability research in Software Engineering in order to facilitate Green IT and provide a basis for sus-
impacts the field. Section V will present more details on these tainable software projects. The Generic Sustainable Software
works. Model presented by Amri et al. [1] characterizes software
sustainability according to the 5 dimensions environmental,
B. Threats to Validity technical, social, individual and economic. Sustainability val-
ues are assigned to a SE project according to these dimensions.
We identified the following possible threats to validity:
Unusual among the list of related work is the so-called
Bias of keywords: The keywords of the search as Karlskrona Manifesto for Sustainability Design [2], signed
described in section II-B could have created a bias by a large number of researchers in the field, which aims
to further the dialogue about and foster awareness for sus- [10] F. Osborne, P. Lago, H. Muccini, and E. Motta. Reducing the effort
tainability design in Software Engineering. The accompanying for systematic reviews in software engineering. Empirical Software
paper delineates current definitions of sustainability and related Engineering, 2017. In preparation.
aspects and concepts. [11] F. Osborne and E. Motta. Klink-2: integrating multiple web sources
to generate semantic topic networks. In International Semantic Web
Finally, in order to foster awareness for sustainability Conference, pages 408–424. Springer, 2015.
already in young academics, Penzenstadler and Fleischmann [12] F. Osborne, A. Salatino, A. Birukou, and E. Motta. Automatic
[15] propose a plan for integrating sustainability concepts into classification of springer nature proceedings with smart topic miner.
In International Semantic Web Conference, pages 383–399. Springer,
the curriculum of Master students in Software Engineering. 2016.
[13] B. Penzenstadler. Towards a Definition of Sustainability in and for
VI. C ONCLUSION Software Engineering. In Proceedings of the 28th Annual ACM
Symposium on Applied Computing, SAC ’13, pages 1183–1185, New
The systematic mapping study presented in this paper had York, NY, USA, 2013. ACM.
the goal to give an overview as to how researchers in the [14] B. Penzenstadler, V. Bauer, C. Calero, and X. Franch. Sustainability in
domain of Software Engineering defined and approached the software engineering: A systematic literature review. In 16th Interna-
omnipresent issue of sustainability. Taken on one hand from tional Conference on Evaluation Assessment in Software Engineering
a previously executed literature study and on the other hand (EASE 2012), pages 32–41, May 2012.
extracted automatically, we analyzed a total of 1035 papers [15] B. Penzenstadler and A. Fleischmann. Teach sustainability in software
on sustainability and Green IT. Given a suitable data set of engineering? In 2011 24th IEEE-CS Conference on Software Engineer-
ing Education and Training (CSEE T), pages 454–458, May 2011.
publications as a basis, this automatic extraction algorithm
[16] B. Penzenstadler, A. Raturi, D. Richardson, C. Calero, H. Femmer,
significantly decreases the manual effort necessary for the and X. Franch. Systematic Mapping Study on Software Engineering
execution of a study, however, it does not entirely eliminate for Sustainability (SE4s). In Proceedings of the 18th International
the manual review process. Conference on Evaluation and Assessment in Software Engineering,
EASE ’14, pages 14:1–14:14, New York, NY, USA, 2014. ACM.
The analysis of the finally selected papers emphasizes [17] K. Petersen, S. Vakkalanka, and L. Kuzniarz. Guidelines for conduct-
how relatively new sustainability is as an issue in software ing systematic mapping studies in software engineering: An update.
engineering. Research is strongly focused on the environmental Information and Software Technology, 64:1–18, Aug. 2015.
aspect and the direct effects of the software engineering
process and the resulting product. Systemic effects of software
and the SE process are still getting relatively few attention.
Efforts in related work to formalize sustainability as part of
the SE process have been made only during recent years and
have yet to make their way into official standards and models.

R EFERENCES
[1] R. Amri and N. B. B. Saoud. Towards a Generic Sustainable Software
Model. In 2014 Fourth International Conference on Advances in
Computing and Communications, pages 231–234, Aug. 2014.
[2] C. Becker, R. Chitchyan, L. Duboc, S. Easterbrook, B. Penzenstadler,
N. Seyff, and C. C. Venters. Sustainability Design and Software: The
Karlskrona Manifesto. In 2015 IEEE/ACM 37th IEEE International
Conference on Software Engineering, volume 2, pages 467–476, May
2015.
[3] K. R. Berntsen, M. R. Olsen, N. Limbu, A. T. Tran, and R. Colomo-
Palacios. Sustainability in Software Engineering - A Systematic
Mapping. In SpringerLink, pages 23–32. Springer, Cham, Oct. 2016.
[4] P. Bourque and R. E. Fairley, editors. Guide to the Software Engineering
Body of Knowledge (SWEBOK(R)): Version 3.0. IEEE Computer
Society Press, Los Alamitos, CA, USA, 3rd edition, 2014.
[5] G. H. Brundtland. Our common future - Call for action. Environmental
Conservation, 14(04):291–294, 1987.
[6] C. Calero and M. Piattini. Introduction to green in software engineering.
In C. Calero and M. Piattini, editors, Green in Software Engineering,
pages 3–27. Springer, 2015.
[7] E. Jagroep, J. Broekman, J. M. E. M. van der Werf, S. Brinkkemper,
P. Lago, L. Blom, and R. van Vliet. Awakening awareness on energy
consumption in software engineering. In Proceedings of the 39th Inter-
national Conference on Software Engineering: Software Engineering in
Society Track, ICSE SEIS, pages 76–85. IEEE Press, 2017.
[8] P. Lago, R. Kazman, N. Meyer, M. Morisio, H. A. Müller, and
F. Paulisch. Exploring initial challenges for green software engineering:
Summary of the first GREENS workshop, at ICSE 2012. SIGSOFT
Softw. Eng. Notes, 38(1):31–33, Jan. 2013.
[9] S. Naumann, M. Dick, E. Kern, and T. Johann. The GREENSOFT
Model: A reference model for green and sustainable software and
its engineering. Sustainable Computing: Informatics and Systems,
1(4):294–304, Dec. 2011.

View publication stats