Architecture Design Decisions in Distributed Teams: An Assessment of

Tool Support

Mahum Adil

1 a

, Ilenia Fronza

1 b

, Outi-Sievi-Korte

2 c

and Claus Pahl

1 d

Free University of Bozen-Bolzano, Italy

Tampere University, Finland

ﬁ

Keywords:

Architecture Design Decisions (ADD), ADD Tools, Distributed Collaborative Design (DCD), Global

Software Engineering (GSE).

Abstract:

Background. Global Software Engineering (GSE) teams develop software artifacts across multiple locations.

Designing and maintaining software in such a setting requires continuous collaboration to record design de-

cisions between distributed teams. The literature presents different architecture design decision (ADD) tools

to support design thinking and decision-making. However, recent studies highlight the challenges of man-

aging ADDs in a distributed environment. Aim. This study aims to present a list of assessment metrics to

evaluate whether existing ADD tools support distributed collaborative design (DCD) in the GSE environment.

Method. We used Goal-Question-Metric (GQM) method and expert survey to deﬁne and validate the assess-

ment metrics. Result. Six assessment metrics are presented to evaluate ADD tool support for DCD in the GSE

environment. Conclusion. The tool analysis based on assessment metrics provides insights into the current

practices used for ADD process and its conformance to the distributed environment.

1 INTRODUCTION

Since the global pandemic, many software organiza-

tions have started working in distributed teams at a

larger scale (Matos and Franc¸a, 2022), which requires

software engineers to understand the importance of

collaboration fully. In a Global Software Engineer-

ing (GSE) setting, software developers work in differ-

ent geographical and temporal boundaries (Herbsleb,

2007), sometimes including different cultures. As a

result, GSE was extensively adopted in unprecedented

circumstances by COVID-19 (Marek et al., 2021).

The existing literature highlights the challenge of

designing software architecture in distributed teams

due to a need for practices to efﬁciently manage the

distributed nature of the architecture design process

(Adil et al., 2022; Sievi-Korte et al., 2019). The soft-

ware architecture represents architecture design deci-

sions (ADD) (Jansen and Bosch, 2005), which com-

prise software elements evolved in the software de-

velopment life-cycle (Alexeeva et al., 2016). Related

https://orcid.org/0000-0001-6452-6085

https://orcid.org/0000-0003-0224-2452

https://orcid.org/0000-0002-4956-8989

https://orcid.org/0000-0002-9049-212X

research presented various ADD tools to manage the

design decision process (Bhat et al., 2020). How-

ever, whether these tools support collaboration in dis-

tributed teams while managing software architecture

in a GSE environment is still being determined.

To address this issue, we present a list of as-

sessment metrics to evaluate how ADD tools support

distributed collaborative design (DCD). Five experi-

enced researchers in the ﬁeld of GSE research val-

idated the deﬁnition of the assessment metrics. To

comprehensively review the existing ADD tools, we

used a semi-structured literature review presented by

Bhat et al. (Bhat et al., 2020) to analyze ADD tools

based on the assessment metrics to identify DCD

support in a GSE environment. The evaluation of

the existing ADD tools highlights the current gap in

supporting team collaboration for managing design

thinking and decision-making in a distributed envi-

ronment.

The paper is structured as follows. Section 2

presents background and related work on ADD tools.

Section 3 introduces the research objective, and Sec-

tion 4 presents the research methodology. Section 5

describes the results, which are then discussed in Sec-

tion 6. Section 7 concludes the paper and shares the

future work for the research.

Adil, M., Fronza, I., Sieve-Korte, O. and Pahl, C.

Architecture Design Decisions in Distributed Teams: An Assessment of Tool Support.

DOI: 10.5220/0011848000003467

In Proceedings of the 25th International Conference on Enterprise Information Systems (ICEIS 2023) - Volume 2, pages 65-74

ISBN: 978-989-758-648-4; ISSN: 2184-4992

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

2 RELATED WORK AND

BACKGROUND

2.1 Related Work

During the past two decades, collaboration has be-

come a fundamental component of the software de-

sign process (Larsson, 2003), involving several ac-

tivities grouped into three phases: problem analy-

sis, solution design, and product development (Shaw,

1995). These activities are complex and require a

sound knowledge of design principles and, more im-

portantly, their skillful application (Jolak et al., 2020).

Therefore, software design in a collaborative envi-

ronment largely depends on design thinking and de-

cision practices to support knowledge and architec-

tural planning across all teams. Rodriguez et al. pre-

sented a systematic mapping study discussing tools

used in GSE (Portillo-Rodr

ıguez et al., 2012). How-

ever, the tools categorized as software design focus on

design diagrams instead of the design decision pro-

cess. A systematic mapping study on architecture de-

cisions discussed different aspects of architecture de-

cisions, such as decisions based on use cases, qual-

ity attributes, or group decisions (Tofan et al., 2014).

However, the approaches discussed in the study are

only proposed solutions and are not validated empir-

ically in a real-time environment. The evolution of

architecture decisions and knowledge management is

presented in (Capilla et al., 2016) and (Bhat et al.,

2020), which address the success, shortcomings, and

acceptance of existing ADD tools in software organi-

zations.

2.2 Historical Review of ADD Tools

To analyze the evolution and the shift of focus on

collaboration in the design decision-making process,

we reviewed the ADD tools given in Bhat et al.,

(Bhat et al., 2020) semi-structured systematic map-

ping study, and we distinguished three generations

(2006-2019). The ﬁrst-generation ADD tools focus

on capturing architecture elements and visualization

processes. The second-generation ADD tools are

solution-based research focused on existing limita-

tions in the ﬁeld. The third-generation ADD tools

are advanced automated ADD practices that support

the idea of the collaborative design decision process.

Moreover, later sections in the study will detail how

we structured our research approach to compare these

tools with regard to GSE objectively.

The Architecture Design Decision Support System

(ADDSS) (Capilla et al., 2006) is a research web-

based tool to capture, manage, and store architecture

decisions. ADDSS captures architecture decisions

following an iterative software development process

to store the architecture knowledge evolution for soft-

ware architects to build possible solutions. However,

the tool does not support any communication model

to update the team regarding updates in requirements,

design decisions, or architecture stored in the tool.

ADD Visualization (Lee and Kruchten, 2008b) is a

research-based visualization tool to represent design

decisions to inspect the system’s architecture graphi-

cally. The tool supports four views to provide a de-

pendency list of design, design decisions visualiza-

tion, design decision history, and the impact of de-

sign decisions. However, the view of the tool is solely

based on the architect, which restricts any communi-

cation process within the team.

Decisionstick is based on ADD visualization to

support three approaches to capture ADD, such as

formal identiﬁcation and elicitation, from existing

project documentation and source code annotation

(Lee and Kruchten, 2008a). The authors validated

the feasibility of the tool with practitioners, whereas

no further work was done to enhance the tool. De-

cision Management tool integrates design decisions

with modeling (Uniﬁed Modeling Language - UML)

tool (Konemann, 2009). The idea is to update the

modeling diagrams based on the design decisions

continuously; however, the tool was not validated in

the real-time environment. Ontology Driven Visual-

ization (ODV) supports practitioners in ADD process

(De Boer et al., 2009): the authors used table and ma-

trix representation of ADD with quality attribute to

support product quality evaluation. However, the tool

only shows the negative or positive effect of the qual-

ity criteria on the architectural decision with no com-

munication process with the software quality analyst

to understand the impact of the decision on the soft-

ware system.

Second Generation (2010-2014). The cus-

tomized ADD model-based tool Architectural De-

sign Decisions Management support (ADDMS) (Chen

et al., 2010) manages and organizes architecture

knowledge for the team working on the project. Some

extensions for the tool have been proposed (Chen and

Babar, 2010); however, validation was performed in

a controlled environment. The service-oriented ar-

chitecture design tool Rationale Visualization (Shahin

et al., 2010) allows the user to explore and visual-

ize the ADDs. The tool is integrated using the Com-

pendium tool to store scope, design drivers, and alter-

natives for each design decision. The tool was pro-

posed to support collaboration; however, no commu-

nication process was used.

Traceability between software architecture and

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

other software artifacts is another solution given in

research to main software evolution. The Language

for integrated software architecture (LISA) toolkit is

a research-based solution to build a semi-automatic

traceability network between ADDs and software im-

plementation (Buchgeher and Weinreich, 2011). The

tool captures traceability in three steps. First, the de-

veloper selects the design decision; then, the devel-

oper builds the ADD record to specify implementa-

tion elements. Finally, a third review serves to en-

sure compatibility with the software system. The

web-based tool Repertory grid technique (RGT) (To-

fan and Galster, 2014) focuses on the collaborative

design decision process. The tool is available on-

line and focuses on three aspects of architecture de-

cisions; to capture, analyze, and make group deci-

sions on ADDs. Design Practice STream (DPS) DPS

(Nakakoji et al., 2011) supports real-time annotation

of video meetings to help new team member(s) to join

existing projects and build an understanding of ex-

isting ADDS. Architectural Development Using Ar-

chitectural Knowledge (ADUAK) is a web-based re-

search tool proposed by Dhaya and Zayaraz (Dhaya

and Zayaraz, 2012) to explore design patterns by us-

ing architecture knowledge of the system to enhance

ADD process.

To promote the importance of non-functional re-

quirements during the architecture design process,

Architech (Ameller et al., 2012) suggests alterna-

tive ADDs to satisfy non-functional requirements

stated for the software system. The research-based

tool Architectural Design Decision Support Frame-

work (ADvISE) (Lytra et al., 2013) is based on the

QOC (Question, Option, Criteria) approach to model

ADDs, track possible issues, and build the solution.

Software Architecture Warehouse (SAW) is a web-

based research tool (Nowak and Pautasso, 2013), suit-

able for distributed collaborative design decision pro-

cess. However, the tool does not support integration

with existing architecture knowledge tools and does

not comply with system and software architecture

standards (e.g., ISO 40210). Decision Architect de-

veloped by Manteuffel et al., (Manteuffel et al., 2014)

is a plug-in for Sparx System Enterprise Architect to

manage design decisions and design rationale for all

the developers involved in the project.

Third Generation (2015-2019). To synchronize

architecture decision-related knowledge with design

documentation, the Eclipse plug-in DecDoc (Hesse

et al., 2016) helps a team to capture ADDS and col-

laborate on design documentation by reﬂecting on the

decision-making process and enhance the knowledge

for building solutions. Similarly, Ontology-based rec-

ommender (Bhat et al., 2017), based on DBpedia on-

tology, automatically identiﬁes architecture elements

from software design documentation and proposes

a list of possible solutions to enhance the decision-

making process. Since the ontology-based recom-

mender is part of DBpedia ontology, the accuracy of

the recommendation can be affected by it. Quiver is a

web-based design decision recommender (Gopalakr-

ishnan and Biswal, 2017) that captures and stores ref-

erence architecture knowledge and architecture arti-

facts to provide recommendations for design deci-

sions. However, the tool has not been validated to un-

derstand the accuracy of its recommendations. Evolu-

tion Visualization for Architecture (EVA) (Nam et al.,

2018) visualizes architecture evolution from multiple

facets. The tool enhances team collaboration as the

developers can visualize the software architecture and

explore the impact of design decisions on the software

product. However, empirical analysis has yet to be

done to check the performance of the tool in a real-

environment.

Recently, researchers shared studies to show the

importance of choosing experienced developers with

architectural expertise to address any possible short-

comings in software systems (Bhat et al., 2018). To

support the idea, the expert recommendation system

ADeX (Amelie - Decision Explorer) (Bhat et al., 2019)

provides a list of experts based on their expertise in

the ﬁeld. The tool uses a bottom-up approach to auto-

matically detect and generate speciﬁc views on archi-

tecture knowledge to support the collaborative design

decision-making process.

3 RESEARCH OBJECTIVE

This study presents a list of assessment metrics to

evaluate whether the existing ADD tools support

DCD in the GSE environment. This evaluation will

help identify the possible gap in existing tools for

managing collaborative design thinking and decision-

making process in distributed teams. The aim of the

study is presented through the three levels of the Goal-

Question-Metrics (GQM) measurement model (Basili

et al., 2014) where:

1. Goal: to evaluate whether existing ADD tools

support DCD in the GSE environment.

2. Question: do ADD tools support the collaborative

design decision for managing software architec-

ture in the GSE environment?

3. Metrics: six assessment metrics are deﬁned to

evaluate existing ADD tools quantitatively.

To verify the goal achievement of the study, we

used the GQM template to provide a structured def-

Architecture Design Decisions in Distributed Teams: An Assessment of Tool Support

inition of the measurement (Table 1) by specifying

the purpose (what object and reason of measurement),

perspective (what aspect to evaluate and who will do

it) and context (the environment) characteristics.

Table 1: GQM template for study.

Analyze ADD tools support for

DCD in GSE

For the purpose of Understanding

With respect to Tool

From the viewpoint Researcher

In the context of ADD process

4 RESEARCH METHOD

To identify the assessment metrics, we used the Goal-

Question-Metrics (GQM) measurement model (Basili

et al., 2014) (shown in Figure 1) and an expert sur-

vey to deﬁne metrics empirically evaluated by the

GSE ﬁeld experts. To deﬁne the assessment met-

rics, we used the extended taxonomy of GSE (Britto

et al., 2016) and features of GSE tools presented in a

systematic mapping study (Portillo-Rodr

ıguez et al.,

2012). To validate metrics, we identiﬁed GSE authors

based on the excellent articles published in the ﬁeld

on Google Scholar. Through the search, we contacted

14 experts working in the ADD/GSE ﬁeld via email.

The experts received a detailed document describing

the metrics deﬁnition and, for each metric, were asked

to approve, disapprove, or suggest to avoid by using

an online survey. If they disapproved, they were asked

to suggest changes; if they suggested avoiding a met-

ric, they could motivate their answer. We received

ﬁve responses in total. All the answers given by the

experts are kept anonymous, and the credentials are

given in Table 2 after their shared consent.

In order to ﬁnalize the assessment metrics, all

the experts’ responses were discussed among the re-

searchers of this study. All the researchers simulta-

neously agreed on the name and deﬁnition the experts

approved; however, in case the names or deﬁnitions of

the assessment metrics were disapproved or suggested

to avoid, the experts shared new names or deﬁnitions

to keep it coherent with the GSE taxonomy. Metrics

names and deﬁnitions were changed according to the

given suggestions and the researchers’ common con-

sensus.

4.1 Deﬁnition of Assessment Metrics

A detailed description of the identiﬁed and validated

assessment metrics follows below. The aim is to sep-

Table 2: Expert credentials.

Country Afﬁliation Academic

rank

Involved

in GSE

research

Germany Academic Professor Since 2003

Spain Academic Professor Since 2004

Ireland Academic,

Applied

Research

Professor Since 2006

Spain Academic,

Software

industry

Professor Since 2006

Sweden Academic,

Software

industry

Senior

Re-

searcher

Since 2012

arate the main concerns of ADD and DCD in a GSE

setting using the GQM approach.

4.1.1 M1: Synchronous Collaborative Work

The ﬁrst dimension of consideration is the tool’s sup-

port for team communication and collaboration. To

do so, we check if the tool supports team members

collaborating or communicating simultaneously on a

task. Herein, we used GSE taxonomy (Britto et al.,

2016) to assess whether synchronous collaborative

work is supported by the tool for distributed teams. To

categorize, we used binary values, i.e., Yes when the

synchronous collaboration is supported and No when

the synchronous collaboration is not supported.

4.1.2 M2: Software Development Methodology

The software development methodology is used to un-

derstand the practices incorporated in different sites

to conduct GSE projects. The development method-

ology used in different sites can help to classify the

effort associated with the development and coordina-

tion with each site. We used GSE taxonomy (Britto

et al., 2016) to categorize the two dimensions of soft-

ware development methodology used in the tool as

follows:

• Plan-driven, when the tool requires detailed re-

quirements deﬁnition against each team role,

which minimizes the collaboration within the

team.

• Agile, when the tool does not require detailed re-

quirements deﬁnition and the team collaboration

is increased to build the solution.

4.1.3 M3: Communication Mode

In GSE projects, the communication between dis-

tributed teams is mediated through different elec-

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

Figure 1: Proposed assessment metrics using the GQM measurement (Basili et al., 2014) and expert survey.

tronic communication media. To understand the

mode of communication used in tools, we used a

GSE taxonomy (Britto et al., 2016), and communica-

tion features listed for GSE tools (Portillo-Rodr

ıguez

et al., 2012) to categorize three levels of synchronicity

as follows:

• Low, when the tool supports asynchronous com-

munication via, e.g., forum discussion (F), com-

ments on software items (Cmt), or email discus-

sion (E);

• High, when the tool supports synchronous com-

munication via, e.g., audio/ video-conference call

• Balanced when the tool supports both syn-

chronous and asynchronous when needed.

4.1.4 M4: Collaboration Awareness

In GSE projects, it is crucial to collaborate with the

other team members to participate and avoid any pos-

sible conﬂict (Trainer and Redmiles, 2018) during

ADD process. Here we used the awareness feature

listed for GSE tools (Portillo-Rodr

ıguez et al., 2012)

to deﬁne collaboration activities as follows:

• Presence awareness, when the tool allows to know

where team members are geographically located

(global awareness) or who is connected in a ses-

sion (session awareness);

• Change awareness, when the tool provides infor-

mation on who is doing a task independently (vi-

sual awareness) or supports asynchronous com-

munication to inform any possible changes in the

project (email awareness).

4.1.5 M5: Control Activities

In distributed teams, it is crucial to monitor the

project’s progress, track any possible issues, and con-

tinuously update the changes in the related docu-

ments. We used the control and coordination feature

listed in GSE tools (Portillo-Rodr

ıguez et al., 2012)

to deﬁne control activities as:

• Issue tracking (IT), when the tool supports the

identiﬁcation of any possible issue in the project

and team coordination accordingly.

• Version control (VC), when the tool allows keep-

ing track of changes in the related documents

and coordinating the changes between distributed

teams.

• Build management (BM) when the tool supports

the automated source code of the software project.

4.1.6 M6: Knowledge Management Type

In order to enhance coordination and collaboration

within distributed teams, knowledge management is

used to support knowledge acquisition, sharing, and

distribution for the software development process.

We used the knowledge management features used in

GSE tools (Portillo-Rodr

ıguez et al., 2012), covering:

• Wiki, when the tool includes a platform to share

knowledge or seek consult from the rest of the

team;

• Document management system (DMS), when the

tool allows storing all the related documents of the

software project in electronic format with proper

indexing in case of information retrieval.

• Blog, when the tool supports brainstorming ses-

sions to exchange knowledge in a distributed en-

vironment.

Architecture Design Decisions in Distributed Teams: An Assessment of Tool Support

5 RESULTS

Table 3 summarizes the analysis of 20 ADD tools

(Bhat et al., 2020) based on the assessment metrics

to evaluate DCD support in GSE environment. We

address the six criteria one by one in the following.

Synchronous Collaborative Work. In order to

manage ADDs in a distributed environment, contin-

uous team communication and collaboration are re-

quired for brainstorming, listing all the possible so-

lutions, and establishing consensus on a ﬁnal set of

ADDs for the software system (Yang et al., 2021).

In the analysis, we noticed that the support of syn-

chronous collaborative work was captured in 5 of 20

ADD tools discussed in the third generation of tools

mentioned in Section 2. However, most of these

tools lack empirical validation studies to assess per-

formance in a real-time environment.

Software Development Methodology. Agile

practices are used in the software development pro-

cess to support collaborative design decisions for

managing software architecture in a GSE environ-

ment (Camara et al., 2020). However, our analysis no-

ticed that no ADD tool was designed to support agile

practices for software architecture and development

processes.

Communication Mode. The global distance be-

tween distributed teams requires more effort to ease

communication during software design thinking and

decision-making. While the communication chal-

lenges in distributed teams are well discussed in liter-

ature (Hummel et al., 2013; Sievi-Korte et al., 2019),

12 of 20 tools lack support for communication mode

(synchronous and asynchronous) to seek team con-

sensus on the software architecture of the project.

Collaboration Awareness. The ADD process

is a group-based activity requiring continuous coor-

dination among teams to make distributed collabo-

ration effective (Bosch and Bosch-Sijtsema, 2010).

The analysis shows that 11 of 20 ADD tools support

change awareness, which shows how tasks are allo-

cated to each member. Furthermore, 6 of 20 ADD

tools do not support collaboration activities, thereby

restraining distributed team members from instanta-

neously increasing shared understanding of the soft-

ware system.

Control Activities. To monitor the project’s

progress and track any possible issues in distributed

teams, 10 of 20 ADD tools support issue tracking

(IT). This feature is used to centralize the project

repository and prioritizes issues according to sever-

ity. However, distributed teams need to have up-to-

date project documentation to resolve the given is-

sues. This allows the distributed teams to review the

related documents and systematically propose solu-

tions. In this analysis, only 6 of 20 tools support issue

tracking (IT) and version control (VC) to coordinate

related changes with distributed teams.

Knowledge Management Type. To support

knowledge acquisition, sharing, and distribution in

distributed teams, 15 of 20 ADD tools used a docu-

ment management system (DMS) to store the related

documents. However, these tools offer limited inte-

gration capabilities and collaboration features for dis-

tributed environments.

6 DISCUSSION

This section discusses the results of this work to ad-

dress the possible barriers of DCD in a distributed en-

vironment. It is important to note that some of the

ﬁndings discussed here reﬂect known concerns, such

as communication and collaboration barriers in GSE

(Noll et al., 2011) and beneﬁts of using agile practices

in GSE (Jalali and Wohlin, 2012), but are valuable

here to provide a comprehensive perspective. Hence,

this study acknowledges the existing gaps in the ﬁeld.

The following observations concern the focus of

the reviewed tools and speciﬁc features that are con-

sequently well or insufﬁciently supported.

Active software process model. The results pre-

sented in Table 3 show that the plan-driven software

process model is actively supported in all generations

of ADD tools. We found that 16 of 20 tools follow the

plan-driven method since the process activities were

predeﬁned and the project tasks were executed in a se-

quential task-speciﬁc manner. As a result, this feature

limits the collaboration within the team to discuss any

design-related issues. In order to increase collabora-

tion in distributed teams, the agile software process

should be used within ADD tools to emphasize the

importance of team communication and coordination

practices for successful projects (S

eguin et al., 2012).

Research based tools are in majority. Regarding

the current support of ADD tools for the collaborative

design decision process, we found that 13 of 20 tools

were prototypes developed by researchers, while two

tools shared the prototype on online repositories (e.g.,

gitHub) and can be used without any license payment.

Five of the 20 tools are commercial tools developed

by practitioners as part of a software product, and the

license can be obtained through payment. The results

show that most tools are research-based projects, so

validating their effectiveness in real-time is challeng-

ing.

Communication process is not well supported.

The support for the communication process in tools

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

Table 3: ADD tools analysis using the proposed assessment metrics.

Name Sync. col-

lab. work

Software dev.

methodology

Communic.

mode

Collaboration

awareness

Control activ-

ities

Knowledge

manage. type

ADDSS No Plan-driven NA NA VC DMS

ADD Visual-

ization

No NA NA Change: Vi-

sual

IT, VC DMS

Decision

Stickies

No Plan-driven Low: F Change: Vi-

sual

IT, VC DMS

Decision

management

tool

No Plan-driven NA Change: Vi-

sual

IT Wiki

ODV No Plan-driven NA Change: Vi-

sual

BM NA

ADDMS No NA Low: F Change: Vi-

sual, email

IT DMS

Rationale vi-

sualization

No Plan-driven NA Change: Vi-

sual

IT DMS

LISA No NA Low: F Change: Vi-

sual

IT, BM NA

RGT Yes Plan-driven Low: Cmt Presence:

Session

IT DMS

DPS No NA NA NA IT NA

ADUAK No Plan-driven NA NA IT DMS

ArchiTech No Plan-driven NA Change: Vi-

sual

IT DMS

ADvISE No Plan-driven NA NA IT, BM DMS

SAW Yes Plan-driven Low: F, Cmt Presence:

Session

IT DMS

Decision Ar-

chitect

No Plan-driven NA Change: Vi-

sual

IT, VC DMS

DecDoc Yes Plan-driven Low: Cmt Change: Vi-

sual

IT, VC DMS

Ontology-

based recom-

mender

No Plan-driven Low: Cmt Change: Vi-

sual

IT DMS

Quiver No Plan-driven NA NA IT NA

EVA Yes Plan-driven Low: F, Cmt Change: Vi-

sual

IT, VC DMS

ADeX Yes Plan-driven NA NA IT, VC DMS

was another focus point for the study since it encour-

ages continuous information ﬂow within distributed

teams. (Alexeeva et al., 2016). According to our anal-

ysis, 7 out of 20 tools supported low asynchronous

communication to track any issue or changes done

in architecture presentation. Furthermore, no tool

supported high synchronous or balanced communica-

tion process. This leads to a research gap regarding

how to support the communication process to control

the progress of tasks, track issues and collaborate for

decision-making while working in a distributed envi-

ronment.

Few tools exist for collaborative design decision.

To improve architecture presentation and DCD pro-

cess, 5 out of 20 tools support team collaboration to

be continuously aware of the activities performed by

team members. This activity is tightly coupled with

the communication process used in the tool. In these

ﬁve tools, the communication support consisted of fo-

rum discussions and comments on software items to

inform the possible changes in the project. Further-

more, only 2 of these ﬁve tools support the session

awareness to know who is currently connected within

the team. This is another research gap that needs to be

addressed by researchers and practitioners, as work-

ing in distributed teams is a new normal software de-

velopment process.

Architecture Design Decisions in Distributed Teams: An Assessment of Tool Support

6.1 Threats to Validity

We analyzed possible threats to validity following the

guidelines provided by Wohlin et al. (Wohlin et al.,

2012) to mitigate any research bias. Concerning the

presented results, this study poses research limitations

as we only investigated the tools mentioned in Bhat et

al., (Bhat et al., 2020). Some existing tools that have

not been included here may have been published after

the related research. However, to limit this threat, we

searched systematic mapping studies and literature re-

views through research sources published after 2020

focused on the same area of research. Apart from the

tools mentioned in this study, we may have excluded

some commercial tools since many commercial tools

are not presented through research publications.

Concerning construct validity, a possible threat re-

lates to how the expert survey results are obtained. To

avoid any threats to construct validity, we contacted

the experts in the ﬁeld to share their opinion on the

assessment metrics through an anonymous survey to

ensure data conﬁdentiality and avoid evaluation ap-

prehension. During the survey, we noticed that ex-

perts suggested different terms to what was given to

them in the initial document. As a result, it helped to

avoid any inﬂuence of the authors of this paper on the

proposed assessment metrics.

External validity concerns the ability to generalize

the results for a speciﬁc setting. The proposed assess-

ment metrics can be used by other researchers in the

GSE ﬁeld, focusing on tool support for a broader per-

spective of ADD process in a distributed environment.

7 CONCLUSIONS AND FUTURE

WORK

While tool support for architecture design for dis-

tributed development in a global software engineer-

ing context exists, recent years have emphasized the

need to speciﬁcally consider distributed collaborative

design (DCD)(Yang et al., 2021).

In this paper, we comprehensively reviewed ADD

tools in terms of their support for DCD in distributed

environments by building on a semi-structured liter-

ature review presented by Bhat et al. (Bhat et al.,

2020) to identify the tools. To do so, we identiﬁed,

deﬁned, and validated six assessment metrics with the

support of experts of GSE. Based on the assessment

metrics analysis, existing ADD tools are not designed

to incorporate agile practices for the software design

decision process. Furthermore, among the consid-

ered tools, only ﬁve support collaborative practices to

manage software architecture in a distributed environ-

ment. This shows the need for synchronous collabo-

rative work to model ADDs and support DCD in ex-

isting tools, showing the need for additional research

to prevent knowledge vaporization.

In the future, we aim to go beyond this tool review

by conducting a systematic mapping study to analyze

ADD approaches and tools to discover which design

thinking and decision-making practices are currently

used in software organizations. Based on the results,

our long-term goal is to design an ADD tool that over-

comes the identiﬁed limitations to enhance the collab-

orative design in distributed teams.

REFERENCES

Adil, M., Fronza, I., and Pahl, C. (2022). Software design

and modeling practices in an online software engi-

neering course: The learners’ perspective. In CSEDU

(2), pages 667–674.

Alexeeva, Z., Perez-Palacin, D., and Mirandola, R.

(2016). Design decision documentation: A literature

overview. In European Conference on Software Archi-

tecture, pages 84–101. Springer.

Ameller, D., Collell, O., and Franch, X. (2012). Architech:

Tool support for nfr-guided architectural decision-

making. In 2012 20th IEEE International Require-

ments Engineering Conference (RE), pages 315–316.

IEEE.

Basili, V., Trendowicz, A., Kowalczyk, M., Heidrich, J.,

Seaman, C., Mnch, J., and Rombach, D. (2014).

Aligning Organizations Through Measurement: The

GQM+Strategies Approach. Springer Publishing

Company, Incorporated.

Bhat, M., Shumaiev, K., Biesdorf, A., Hohenstein, U., Has-

sel, M., and Matthes, F. (2017). An ontology-based

approach for software architecture recommendations.

AMCIS 2017.

Bhat, M., Shumaiev, K., Hohenstein, U., Biesdorf, A., and

Matthes, F. (2020). The evolution of architectural de-

cision making as a key focus area of software archi-

tecture research: A semi-systematic literature study.

In 2020 IEEE International Conference on Software

Architecture (ICSA), pages 69–80.

Bhat, M., Shumaiev, K., Koch, K., Hohenstein, U., Bies-

dorf, A., and Matthes, F. (2018). An expert recom-

mendation system for design decision making: Who

should be involved in making a design decision? In

2018 IEEE International Conference on Software Ar-

chitecture (ICSA), pages 85–8509. IEEE.

Bhat, M., Tinnes, C., Shumaiev, K., Biesdorf, A., Hohen-

stein, U., and Matthes, F. (2019). Adex: A tool

for automatic curation of design decision knowledge

for architectural decision recommendations. In 2019

IEEE International Conference on Software Architec-

ture Companion (ICSA-C), pages 158–161.

Bosch, J. and Bosch-Sijtsema, P. (2010). Coordination be-

tween global agile teams: From process to architec-

ICEIS 2023 - 25th International Conference on Enterprise Information Systems

ture. In Agility Across Time and Space, pages 217–

233. Springer.

Britto, R., Wohlin, C., and Mendes, E. (2016). An ex-

tended global software engineering taxonomy. Jour-

nal of Software Engineering Research and Develop-

ment, 4(1):1–24.

Buchgeher, G. and Weinreich, R. (2011). Automatic tracing

of decisions to architecture and implementation. In

2011 Ninth Working IEEE/IFIP Conference on Soft-

ware Architecture, pages 46–55. IEEE.

Camara, R., Alves, A., Monte, I., and Marinho, M. (2020).

Agile global software development: A systematic lit-

erature review. In Proceedings of the 34th Brazilian

Symposium on Software Engineering, pages 31–40.

Capilla, R., Jansen, A., Tang, A., Avgeriou, P., and Babar,

M. A. (2016). 10 years of software architecture

knowledge management: Practice and future. Jour-

nal of Systems and Software, 116:191–205.

Capilla, R., Nava, F., P

erez, S., and Due

nas, J. C. (2006). A

web-based tool for managing architectural design de-

cisions. ACM SIGSOFT software engineering notes,

31(5):4–es.

Chen, L. and Babar, M. A. (2010). Supporting customizable

architectural design decision management. In 2010

17th IEEE International Conference and Workshops

on Engineering of Computer Based Systems, pages

232–240.

Chen, L., Babar, M. A., and Liang, H. (2010). Model-

centered customizable architectural design decisions

management. In 2010 21st Australian Software Engi-

neering Conference, pages 23–32. IEEE.

De Boer, R. C., Lago, P., Telea, A., and Van Vliet, H.

(2009). Ontology-driven visualization of architectural

design decisions. In 2009 Joint Working IEEE/IFIP

Conference on Software Architecture & European

Conference on Software Architecture, pages 51–60.

IEEE.

Dhaya, C. and Zayaraz, G. (2012). Development of mul-

tiple architectural designs using aduak. In 2012 In-

ternational Conference on Communication and Signal

Processing, pages 93–97. IEEE.

Gopalakrishnan, A. and Biswal, A. C. (2017). Quiver—an

intelligent decision support system for software archi-

tecture and design. In 2017 International Conference

On Smart Technologies For Smart Nation (SmartTech-

Con), pages 1286–1291. IEEE.

Herbsleb, J. D. (2007). Global software engineering: The

future of socio-technical coordination. In Future

of Software Engineering (FOSE’07), pages 188–198.

IEEE.

Hesse, T.-M., Kuehlwein, A., and Roehm, T. (2016). Dec-

doc: A tool for documenting design decisions collab-

oratively and incrementally. In 2016 1st International

Workshop on Decision Making in Software ARCHitec-

ture (MARCH), pages 30–37. IEEE.

Hummel, M., Rosenkranz, C., and Holten, R. (2013).

The role of communication in agile systems develop-

ment. Business & Information Systems Engineering,

5(5):343–355.

Jalali, S. and Wohlin, C. (2012). Global software engineer-

ing and agile practices: a systematic review. Journal

of software: Evolution and Process, 24(6):643–659.

Jansen, A. and Bosch, J. (2005). Software architecture as

a set of architectural design decisions. In 5th Work-

ing IEEE/IFIP Conference on Software Architecture

(WICSA’05), pages 109–120. IEEE.

Jolak, R., Savary-Leblanc, M., Dalibor, M., Wortmann,

A., Hebig, R., Vincur, J., Polasek, I., Le Pallec, X.,

erard, S., and Chaudron, M. R. (2020). Software en-

gineering whispers: The effect of textual vs. graphical

software design descriptions on software design com-

munication. Empirical Software Engineering, 25(6).

Konemann, P. (2009). Integrating decision management

with uml modeling concepts and tools. In 2009 Joint

Working IEEE/IFIP Conference on Software Architec-

ture & European Conference on Software Architec-

ture, pages 297–300. IEEE.

Larsson, A. (2003). Making sense of collaboration: The

challenge of thinking together in global design teams.

In Proc. of the 2003 Intl. ACM SIGGROUP Confer-

ence on Supporting Group Work.

Lee, L. and Kruchten, P. (2008a). Customizing the cap-

ture of software architectural design decisions. In

2008 Canadian Conference on Electrical and Com-

puter Engineering, pages 000693–000698. IEEE.

Lee, L. and Kruchten, P. (2008b). A tool to visualize archi-

tectural design decisions. In International Conference

on the Quality of Software Architectures, pages 43–54.

Springer.

Lytra, I., Tran, H., and Zdun, U. (2013). Supporting consis-

tency between architectural design decisions and com-

ponent models through reusable architectural knowl-

edge transformations. In European Conference on

Software Architecture, pages 224–239. Springer.

Manteuffel, C., Tofan, D., Koziolek, H., Goldschmidt, T.,

and Avgeriou, P. (2014). Industrial implementation

of a documentation framework for architectural deci-

sions. In 2014 IEEE/IFIP Conference on Software Ar-

chitecture, pages 225–234. IEEE.

Marek, K., Wi

nska, E., and Dabrowski, W. (2021). The

state of agile software development teams during the

covid-19 pandemic. In International Conference on

Lean and Agile Software Development, pages 24–39.

Springer.

Matos, J. and Franc¸a, C. (2022). Pandemic agility: Towards

a theory on adapting to working from home. In Pro-

ceedings of the 15th International Conference on Co-

operative and Human Aspects of Software Engineer-

ing, pages 66–75.

Nakakoji, K., Yamamoto, Y., Matsubara, N., and Shirai, Y.

(2011). Toward unweaving streams of thought for re-

ﬂection in professional software design. IEEE soft-

ware, 29(1):34–38.

Nam, D., Lee, Y. K., and Medvidovic, N. (2018). Eva: A

tool for visualizing software architectural evolution.

In Proceedings of the 40th international conference

on software engineering: companion proceeedings,

pages 53–56.

Architecture Design Decisions in Distributed Teams: An Assessment of Tool Support

Noll, J., Beecham, S., and Richardson, I. (2011). Global

software development and collaboration: barriers and

solutions. ACM inroads, 1(3):66–78.

Nowak, M. and Pautasso, C. (2013). Team situational

awareness and architectural decision making with the

software architecture warehouse. In European Con-

ference on Software Architecture, pages 146–161.

Springer.

Portillo-Rodr

ıguez, J., Vizca

ıno, A., Piattini, M., and

Beecham, S. (2012). Tools used in global software

engineering: A systematic mapping review. Informa-

tion and Software Technology, 54(7):663–685.

eguin, N., Tremblay, G., and Bagane, H. (2012). Ag-

ile principles as software engineering principles: An

analysis. In International Conference on Agile Soft-

ware Development, pages 1–15. Springer.

Shahin, M., Liang, P., and Khayyambashi, M. R. (2010).

Improving understandability of architecture design

through visualization of architectural design decision.

In Proceedings of the 2010 ICSE Workshop on Shar-

ing and Reusing Architectural Knowledge, pages 88–

95.

Shaw, M. (1995). Making choices: A comparison of styles

for software architecture. IEEE Software, 12(6).

Sievi-Korte, O., Richardson, I., and Beecham, S. (2019).

Software architecture design in global software devel-

opment: An empirical study. Journal of Systems and

Software, 158:110400.

Tofan, D. and Galster, M. (2014). Capturing and making

architectural decisions: an open source online tool.

In Proceedings of the 2014 European Conference on

Software Architecture Workshops, pages 1–4.

Tofan, D., Galster, M., Avgeriou, P., and Schuitema, W.

(2014). Past and future of software architectural

decisions–a systematic mapping study. Information

and Software Technology, 56(8):850–872.

Trainer, E. H. and Redmiles, D. F. (2018). Bridging the gap

between awareness and trust in globally distributed

software teams. Journal of Systems and Software,

144:328–341.

Wohlin, C., Runeson, P., H

ost, M., Ohlsson, M. C., Reg-

nell, B., and Wessl

en, A. (2012). Experimentation in

software engineering. Springer Science & Business

Media.

Yang, Z., Xiang, W., You, W., and Sun, L. (2021). The inﬂu-

ence of distributed collaboration in design processes:

an analysis of design activity on information, problem,

and solution. International Journal of Technology and

Design Education, 31(3):587–609.

ICEIS 2023 - 25th International Conference on Enterprise Information Systems