Impact of End User Human Aspects on Software Engineering

John C. Grundy

Department of Software Systems and Cybersecurity, Faculty of Information Technology,

Monash University, Melbourne, Australia

Keywords:

Software Engineering, Stakeholders, End Users, Human Aspects, Human Factors.

Abstract:

Software is designed and built to help solve human problems. However, much current software fails to take

into account the diverse end users of software systems and their differing characteristics and needs eg. age,

gender, culture, language, educational level, socio-economic status, physical and mental challenges, etc. I give

examples of some of these diverse end user characteristics and the need to better incorporate them into require-

ments engineering, design, implementation, testing, and defect reporting activities in software engineering. I

report on some of our work trying to address some of these issues, including: use of personas to better charac-

terise diverse end user characteristics; extending requirements and design models to capture diverse end user

needs; analysis of app reviews and JIRA logs to identify problems and ways developers try to address them;

analysis of approaches to improve the accessibility of software designs for diverse end users; adaptive user

interfaces and model-driven engineering with human aspects; improved human-centric defect reporting ap-

proaches; and use of living lab co-design approaches to ensure end users are ﬁrst class contributors during all

phases of software development. I ﬁnish by outlining a research roadmap aiming to improve the incorporation

of end user human aspects into software engineering.

1 INTRODUCTION

Humans build software systems to help solve human

problems, be they for industry, leisure, health and well

being, social interactions, and so on (Rashid et al.,

2017; Strengers and Kennedy, 2020; Curumsing et al.,

2019; Grundy and Grundy, 2013). Yet, humans are

different in many ways, including having diverse age,

gender, culture, language, educational level, technical

proﬁciency, preferences in interaction and problem

solving styles, personality, emotional reaction to us-

ing software, mental and physical challenges, and so

on (Burnett et al., 2016; Perez, 2019; Grundy, 2020;

Curumsing et al., 2019; Cruz et al., 2015).

There has been considerable research into co-

operative and human aspects of software engineering

– from the perspectives of software engineers as hu-

mans – for many years (Cruz et al., 2015; Lenberg

et al., 2015; Hidellaarachchi et al., 2021). How-

ever, there has been much less research into how

diverse END USER human aspects impact software

engineering from requirements, design, implementa-

tion, testing, deployment and defect reporting and

ﬁxing perspectives (Lopez-Lorca et al., 2014; Al-

https://orcid.org/0000-0003-4928-7076

shayban et al., 2020; Grundy, 2020; Hidellaarachchi

et al., 2021; Burnett et al., 2016; Yusop et al., 2020).

While human-computer interaction, design science,

psychology, anthropology, and other disciplines have

researched these software end user human impacts

for many years, particularly on design and usabil-

ity evaluation, few software engineers know about

many of the theories involved, techniques and tools

invented, and integrate little or none of these ﬁnd-

ings into contemporary software engineering prac-

tices (Yusop et al., 2016; Grundy, 2020; Madampe

et al., 2020; Curumsing et al., 2019; Alshayban et al.,

2020; Cruz et al., 2015).

Given the increasing use of software for all as-

pects of modern living and working, and the increas-

ing diversity of end users and end user challenges

this software thus has, software engineers must bet-

ter understand and take account of diverse end user

human aspects. In this keynote talk I discuss how

the lack of accounting for some of these end user hu-

man aspects produces not-ﬁt-for-purpose software so-

lutions. I then discuss some of the work my Human-

iSE (Human-centric Software Engineering) team

doing to address different current deﬁciencies dur-

https://www.monash.edu/it/humanise-lab

Grundy, J.

Impact of End User Human Aspects on Software Engineering.

DOI: 10.5220/0010531800090020

In Proceedings of the 16th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2021), pages 9-20

ISBN: 978-989-758-508-1

ing requirements engineering, design, implementa-

tion, evaluation and the overall software process.

The rest of this paper is organised as follows. Sec-

tion 2 outlines some of the key end user human as-

pects and issues of failing to address them during

software development. I then discuss in Sections 3,

4, 5 and 6 some of our work in improving require-

ments engineering, design, model-driven engineering

and evaluation by incorporating end user human as-

pects into key steps of development. In Section 7 I

discuss progress to date and key planned future work

to extend this research and to deliver practical out-

comes for software engineers. In ﬁnish in Section 8

with key conclusions from this keynote talk.

2 END USER HUMAN ASPECTS

The term “human factors” has been used extensively

in HCI and Engineering and Design disciplines to

characterise human issues involved in designing and

using complex technologies (Woodson et al., 1992).

The term “human aspects” has been used for some

time in software engineering but usually to discuss

human issues of software engineers, as individuals

and teams (Hazzan and Tomayko, 2005). In this pa-

per I use the term “end user human aspects” – or

just human aspects – to describe human issues of the

target end users of the software systems we build.

While this concept is related to human aspects of the

software engineers building the systems – and I re-

turn to this interplay at the end of this keynote talk

– they have quite different implications. Similarly

while human factors of complex systems have been

well-studied for many years, how to best support soft-

ware engineers in addressing the diverse human as-

pects of their software end users is a related but quite

different – and under-researched – focus area.

A wide variety of end user human aspects impact

software and its usage. This is becoming more promi-

nent as more and more people need to use software

solutions more and more often (Lopez-Lorca et al.,

2014; Grundy, 2020). Below I summarise some, but

by no means all, of these end user human aspects that

software engineers increasingly need to consider and

address effectively in their work.

Age. End users have a wide range of ages from the

very young to the very old. Different aged people

have quite different challenges in using software, and

may have quite different expectations and reactions

to the same software (Grundy et al., 2018; Nouwen

et al., 2015; Williams et al., 2013). Failure to take

account of differently aged end users may result in

software that uses wrong terminology, poor interfaces

and workﬂow, is overly complex or confusing, and is

not sufﬁciently enjoyable or engaging.

Physical and Mental Challenges. Many software so-

lutions have been developed to assist with people liv-

ing with physical and/or cognitive challenges. Some

must be designed and implemented to speciﬁcally

take account of them to assist with accessibility of

the software or to ensure the software supports and

not harms vulnerable end users (Carcedo et al., 2016;

Sierra and Togores, 2012). These challenges include

poor mental health, various degrees of cognitive im-

pairment and a wide variety of physical challenges,

such as limited or impaired sight, hearing, mobility

and speech (Alshayban et al., 2020; Stock et al., 2008;

Zhao et al., 2020; Rashid et al., 2017). Failure to

account for diverse physical and mental challenges

results in accessibility problems, but also inefﬁcient

and ineffective software solutions, confusing and even

dangerous systems, end user frustration, and creates

digital living barriers and obstacles that add to and/or

exacerbate physical ones.

Emotions. Using the same software system often

generates positive and negative emotions in differ-

ent people. For example, positive reactions might

include a home monitoring system providing a feel-

ing of safety, to negative reactions to the same soft-

ware, such as feeling lack of control or being moni-

tored intrusively (Curumsing et al., 2019). These re-

actions can have a major impact of acceptance and use

of solutions, but are very poorly supported by most

existing software requirements and design methods

(Miller et al., 2015; Taveter et al., 2019).

Personality and Cognitive Style. There has been rel-

atively little research about the impact of personality

and cognitive style differences of end users on soft-

ware usage to date (McElroy et al., 2007; Barnett

et al., 2015; Burnett et al., 2016). Studies with soft-

ware engineers and others have shown signiﬁcant im-

pacts of personality and different cognitive styles –

these may thus have a major impact on end user per-

ceptiosn and usage of software solutions (Cruz et al.,

2015; Kanij et al., 2015).

Engagement and Entertainment. Many end users are

highly driven by enjoyment, entertainment and ‘fun’

aspects of using software – such as with computer

games and gamiﬁcation-based approaches. Similarly,

different people may engage with the same software

at different levels and in different ways (Fensel et al.,

2017; Kumar, 2013). Failure to take these differences

into account will likely result in less appealing soft-

ware for different end user groups.

Human Values. These include values such as inclu-

siveness, equality, privacy, openness, etc. (Winter

et al., 2018). These human values have been found

ENASE 2021 - 16th International Conference on Evaluation of Novel Approaches to Software Engineering

to be mis-aligned with software they use, with many

software systems and developers have values that con-

ﬂict with one or more end user values (Obie et al.,

2021). As a result, these mis-aligned values can cause

severe expectation mis-matches of end users and their

software solutions, reducing the software take-up and

usage.

Gender. Gender bias has been shown to be highly

problematic in many modern technologies including

many software systems (Perez, 2019; Strengers and

Kennedy, 2020; Burnett et al., 2016). Different prob-

lem solving styles of different genders have been

shown to have a major impact on software acceptance

and usability (Burnett et al., 2016).

Ethnicity and Culture. Culture can be used to de-

scribe different beliefs and behaviours of different

groups of people (Alsanoosy et al., 2019). Software

that is biased in terms ethnicity of people is highly

problematic, especially for many emerging AI-based

smart living and surveillance systems used by police

and other agencies (Garvie and Frankle, 2016). Like

many other human aspects, software developers often

have different ethnicity and culture than many of their

target end user groups. Current software development

methods do not assist developers in better understand-

ing and accounting for culture and ethnic differences

of their end users.

Figure 1: Part of a persona used for smart parking app de-

velopment.

Language. Language differences occur in several

forms in end users of software. People speak dif-

ferent languages, use different forms of jargon, have

differing educational attainment levels, have differ-

ent language competencies, and use different dialects

and slang (Roturier, 2015). Failure to take these

language differences into account means software is

much harder to understand and interact with for many

end users. Again, software engineers lack sufﬁcient

techniques and tools to help them take into account

these end user language differences.

Socio-economic Status. There is a huge digital and

physical divide between those with jobs, money and

luxuries, and those in precarious living, little/no work,

and who struggle to access some of the necessities of

life (Ahmed, 2007). There is still a massive imbalance

in the world’s wealth and access to physical world as

well as digital services, including access to commu-

nity support services via digital means (Grundy and

Grundy, 2013). Most software engineers are rela-

tively wealthy and highly educated – supporting more

vulnerable members of society in software solutions

is not always straightforward, obvious or even con-

sidered (Newman et al., 2015).

3 HUMAN-CENTRIC

REQUIREMENTS

ENGINEERING

I outline several projects we are undertaking to en-

hance requirements engineering practices and tools to

better support diverse end user human aspects.

3.1 Use of Personas to Model Users

Software engineers need to better understand the wide

differences in their end users, in terms of their diverse

human aspects outlined in Section 2 as well as oth-

ers (Lopez-Lorca et al., 2014; Grundy, 2020). One

approach used extensively in human-computer inter-

action and design domains, but less frequently in soft-

ware engineering, is the persona. Figure 1 shows an

example persona from a smart parking app develop-

ment project one of our student teams and PhD stu-

dent developed to aid them in better understanding the

range of end users of this app. This example shows

an elder woman user and outlines her demographics,

goals and frustrations relating to transport and park-

ing. We developed several diverse personas and used

them to help capture richer requirements, develop a

more complete design to support all these possible

end users, and to evaluate and reﬁne the prototype

new smart parking app. We aim to develop tools to

help build richer end user personas with diverse hu-

man aspects, and to support their use more extensively

throughout phases of software development.

Impact of End User Human Aspects on Software Engineering

3.2 Identiﬁcation and Dialogue with

Stakeholders

Related to persona building to represent end users,

we are working on developing improved approaches

to identify software stakeholders in general, not just

end users of the software. For example, some e-

health sofwtare systems are not used directly by hos-

pital managers or carers of patients, but these are crit-

ical stakeholders whose needs from the software also

needs to be carefully identiﬁed, modelled and taken

into account. These stakeholders themselves have

diverse human aspects that may different from end

users. Improving dialogue with stakeholders and end

users to more effectively identify and capture human

aspects is needed (McManus, 2004). We plan to cap-

ture these with personas and extended modelling lan-

guages (see Section 4 below for an example).

3.3 Extraction of Human-centric

Requirements from Documents

There has been much work done on extracting func-

tional and non-functional requirements from natural

language documents over many years (Osama et al.,

2020). We want to explore the identiﬁcation, ex-

traction, modelling and reasoning about requirements

relating to human aspects of end users and related

needs. For example, extracting the requirements re-

lating to smart parking for the persona illustrated in

Figure 1 semi-automatically would assist software en-

gineers in identifying and using these end user hu-

man aspect-related requirements. We also plan de-

sign critics that assist software engineers in explor-

ing end user human aspect-related requirements gaps,

incomplete requirements or seemingly-erroneous re-

quirements for the domain/target end users (Ali et al.,

2013).

4 HUMAN-CENTRIC DESIGN

Following on from the previous section, I discuss

some our current projects relating to enhancing the

design phase of software engineers to better address

end user human aspects.

4.1 Modelling Human-centric Aspects

of End Users

Figure 2 shows a wireframe model that one of our

student teams has extended to identify different target

end user groups and to describe different interfaces

Figure 2: Modelling human-centric characteristics of end

users (from (Jim et al., 2021)).

needed for these diverse groups (Jim et al., 2021).

In this example we model different age group needs

from an app interface that needs to take into account

age-related differences of these end users. These in-

clude different language, font, colour, layout, size

and other UI design choices. Such design differences

are not just restricted to appearance/interaction design

choices but also overall problem solving workﬂow for

(parts of) the app. We want to generalise this ap-

proach to other modelling languages for requirements

and design – such as i

∗

, UML, BPMN, user stories

and so on – to enable better capture and usage of a

wide range of diverse end user human aspects during

software development.

4.2 Design with Personas

In Figure 1 we showed one of our smart parking app

personas. Figure 3 shows some of the screens from

the redesigned smart parking app that one our student

teams prototyped using a range of diverse personas.

Some users have physical and age-related challenges

e.g. eye-sight, colour-blindness, mobility skills – that

all need to be carefully designed in and evaluated.

Others have diverse parking needs relating to human

aspects e.g. their children, job, living location, per-

sonal preferences, personal emotions and preferences,

and language and cultural differences. All of these

need to be elicited, modelled and then used during

design, implementation and evaluation of the app pro-

totype. In this example, some text is enlarged for

users with eyesight challenges (1); different prebook-

ing, selection and driving direction approaches are

provided based on different end user needs and pref-

ENASE 2021 - 16th International Conference on Evaluation of Novel Approaches to Software Engineering

Figure 3: Example of human-centric parking app design.

Figure 4: Example of user-deﬁned data visualisation (from

(Avazpour et al., 2015)).

erences (2); image-based recognition of license-plate

avoids typographical errors (3); and different reserv-

ing/booking/leaving workﬂows support different user

personal circumstances (4).

4.3 Addressing Human Values and

Accessibility of Web Sites and Apps

We have several projects looking at analysing web

sites and mobile apps for issues with human values

violations (Obie et al., 2021), privacy and accessibil-

ity issues (Haggag et al., 2021), and user with vari-

ous human aspect-related challenges e.g. sight, hear-

ing, ageing and so on (Grundy, 2020; Grundy et al.,

2018). We are also conducting indepth surveys and

interviews of developers and end users exploring rea-

sons for these issues, why they are challenging to ad-

dress, and how they are ﬁxed when found (Sham-

sujjoha et al., 2021). We aim to determine from

this analysis key design failures in the apps and web

sites which can be detected during design-time via

improved techniques and tools for developers. We

also aim to equip developers with improved guidance

on how to correct these issues with improved design

decision guidelines and implementation automation

tools.

5 HUMAN-CENTRIC

MODEL-DRIVEN

ENGINEERING

5.1 End-user Visualisation Development

For many years we have been developing tools to sup-

port end-user development of complex software sys-

tems (Khalajzadeh et al., 2020; Hirsch et al., 2010;

Grundy et al., 1998). This helps to avoid the clas-

sic problem with conventional software engineering

– reliance on a highly trained expert workforce (soft-

ware engineers) to build and make necessary changes

to software systems. Instead, end users can specify

and generate their own solutions, often for very com-

plex domains, incorporating their own human aspect-

related needs into their own solutions.

Figure 4 shows an example of one such tool, Con-

VErT, used by end users to specify and generate com-

plex information visualisations and data mapping sys-

tems (Avazpour et al., 2019; Avazpour et al., 2015).

This example shows two end user-speciﬁed data vi-

sualisations – a building layout (left) from a CAD

tool dataset and categorisations tree from an ERP

systen (middle). A data mapping between parts of

each data model can be speciﬁed by drag and drop

between elements (arrows connecting), used to gen-

erate code to implement CAD tool to ERP system

data export. Such end user development tools allow

end users to tailor their software solutions themselves

without waiting for software engineers to do it.

Impact of End User Human Aspects on Software Engineering

Figure 5: Need for human aspects in MDE (from (Khambati et al., 2008)).

5.2 Adding Human Aspects to Code

Generators

Figure 5 shows an example of the Visual Care Plan

Modelling Language (VCPML) tool in use to specify

a care plan for those with obesity (Khambati et al.,

2008). This provides a set of visual languages al-

lowing end users – clinicians – to specify complex

health care plans. They can then tailor generic care

plans to speciﬁc patient needs and have the tool gen-

erate a fully working mobile app from the speciﬁed

care plan and associated desired screen deﬁnitions.

While this idea is good in theory, the VCPML tool

does not capture any speciﬁc human aspects of dif-

ferent users e.g. different language, gender, age, cul-

ture, accessibility issues etc. The tool thus generates

a one-size-ﬁts-all mobile app. We are working on in-

corporating such human aspects of different end users

into the modelling languages and code generators of

such MDE-based tools. We will then be able to tai-

lor the generated code to range of different individual

end user human aspect-related differences, better ad-

dressing these in a “personalised” app generated for

each different user.

5.3 Adaptive User Interfaces

Related to the above generated app example, we have

also been experimenting with using an alternative ap-

proach of highly run-time adaptable UI components

(Grundy and Hosking, 2002; Grundy and Zou, 2004).

These allow different end users to specify a range of

preferences e.g. fonts, size, colour scheme, colour

mapping, language preferences etc and have their web

site or app adapt at run-time to these preferences. Fig-

ure 6 shows two example conﬁguration approaches.

A large conﬁguration menu (1) allows users to spec-

ify a wide range of sight- and cognition-related con-

straints that need to be satisﬁed by the web site e.g.

colour blindness, eyesight limitations, dyslexia, etc.

Figure 7 shows an example of an adaptive web site

prototype using these conﬁgurations to adapt a food

information website to a particular end user’s human

aspects e.g. enlarged font size, dyslexia-friendly font

face, color blindness friendly colour scheme, colour

blindness friendly image colour ﬁltering and site auto

re-layout. The second small app conﬁguration panel

in Figure 6 is used by the smart parking app proto-

type from Figure 3 to adapt various UI components in

the app to different user human aspect needs e.g. lan-

guage, font size, style, colour scheme etc. We are also

exploring AI-based semi-automated adaptation of UI

components to different user human aspect needs, to

compliment these manual conﬁguration approaches.

ENASE 2021 - 16th International Conference on Evaluation of Novel Approaches to Software Engineering

Figure 6: Approaches to UI/app conﬁguration.

Figure 7: Example of adapted UI.

6 HUMAN-CENTRIC SOFTWARE

EVALUATION

6.1 A New Human-centric Defect

Reporting Taxonomy

We discovered when exploring the domain of usabil-

ity defect reporting that existing usability defect clas-

siﬁcation schemes are severely limited (Yusop et al.,

2020). To this end we developed a revised usabil-

ity defect classiﬁcation taxonomy to aid the devel-

opment of improved usability defect reports (Yusop

et al., 2018). In a similar way, we want to develop a

taxonomy of end user human aspects in software en-

gineering (Grundy, 2020), and use these to improve

the classiﬁcation of “end user human aspect defects”

in software applications. Such a taxonomy will aid

end users in reporting human aspect-related software

defects in their applications, but also aid software en-

gineers in understanding these defects, the end users

reporting them, what impact they have, and how they

might be ﬁxed. This work also leverages our use of

personas to describe different end users of software

application and analysis of app reviews to determine

human aspect and human value defects with apps,

both described above.

6.2 Human-centric Defect Reporting

An interesting challenge for end users in reporting

human aspect-related defects in their software is the

poor usability and human aspect defects in current de-

fect reporting tools! Similar to the poor usability and

suitability of current usability defect reporrting tools

(Yusop et al., 2018; Yusop et al., 2020), we need to

provide end users with human aspect defect report-

ing tools. Figure 8 shows an example of a human as-

pect defect reporting prototype tool developed by one

of our student teams. This uses an end user human

aspect preference setting panel, similar to the smart

parking and adaptive user interface component ones

in Figure 6, to conﬁgure the reporting tool to differ-

ent end user sight, hearing, cognitive and language

challenges. The left hand side image shows the ﬁrst

stage of reporting a defect. The right hand side im-

age shows further defect information being speciﬁed.

It also shows an alternative way of capturing this in-

formation, using more “human-centric” voice record-

ing and voice transcription. This recorded voice ap-

proach can better suit some end users reporting de-

fects than typing a lot of text. We use a set of personas

to represent different groups of end users with differ-

ent human aspect challenges to group these defect re-

ports. When shown to the software engineers, these

personas help them to see the defects from the per-

spectives of end users with different human aspects to

each other and to the software engineers themselves.

The idea is to help the software engineers reproduce

– and ﬁx and test – the defect through the eyes of the

diverse defect reporters.

6.3 Developer Surveys and GitHub,

JIRA Developer Discussion Analysis

We are interested to better understand (1) what end

user human aspects software engineers have experi-

ence (or little experience) addressing in their soft-

ware; (2) which aspects they ﬁnd harder to address

and why; (3) how they currently talk about human

aspect-related requirements and defects; and (4) what

Impact of End User Human Aspects on Software Engineering

Figure 8: Example of human-centric defect reporting.

tools and techniques they currently use to help them

address different end user human aspects in their soft-

ware. To this end, we have been carrying our a large

survey and targeted interviews of software engineers.

We have also been looking for JIRA and GitHub issue

logs and StackOverﬂow posts that reﬂect discussion

of different human aspects by software engineers. We

hope to link these to end user defect reports and app

reviews and better understand how these are under-

stood (or not) and addressed (or not) by software en-

gineers. This, we hope, will lead us to better guide-

lines, techniques and tools for software engineers to

improve end user human aspect defect ﬁxing, but also

improved design and implementation decision mak-

ing to avoid or reduce the severity of these human-

centric defects.

6.4 Living Lab Co-design Approach

A key issue we have identiﬁed with realising our vi-

sion of achieving improved addressing of end user hu-

man aspects in software engineering is the current ap-

proach used to develop software. Despite agile soft-

ware development’s promise of people-oriented de-

velopment (Hoda et al., 2018), there is still a ”them

vs us” approach taken by software engineers, us =

software engineers and them = everyone else. We are

exploring the use of a living-lab based approach to

software development, popularised by development

approaches used for eHealth systems among others

(Andersen et al., 2017).

Figure 9 outlines our approach (Grundy et al.,

2020). (1) We build focus groups of stakehold-

ers and developers to co-create software solutions as

Figure 9: Living lab co-design approach (from (Grundy

et al., 2020)).

equals. (2) We enrich software requirements and de-

sign processes, models, techniques and tools to cap-

ture and reason about diverse human aspect-related

needs of stakeholders and end users. (3) We use

model-driven engineering techniques to generate and

conﬁgure software allowing for faster development

and response to emerging end user needs. (4) We sup-

port more human-centric defect reporting and ﬁxing,

via proactive feedback loops.

7 DISCUSSION

I summarise some of our key ﬁndings and progress

to date, key gaps still to address, and brieﬂy discuss

software engineers and their human aspects.

We have been focusing on better understanding

current developer approaches to addressing their end

users’ human aspects, key open challenge areas in

this domain, and then researching new techniques and

tools to address these. We conducted a detailed sur-

vey, answered by 59 developers and managers, and

interviewed 12 developers, to better understand these

issues from a software engineering perspective. Fig-

ure 10 summarises their rating of key critical end user

human aspects in their work. Figure 11 summarises

the key reasons given why they ﬁnd these end user

human aspects challenging to address. Key reasons

included the different languages and (lack of) comfort

with technology of different user groups; range of end

user human differences that exist; complexity of user

ENASE 2021 - 16th International Conference on Evaluation of Novel Approaches to Software Engineering

Figure 10: End user human aspects survey respondents judge to be critical (or not) in their work.

Figure 11: Reasons these human aspects are challenging and how taken into consideration in survey respondents work.

interfaces; different problem solving styles of many

end user groups; and differences in terminology used,

digital literacy and need to carefully consider text and

icon usage.

We asked developers what would help them to

improve development of their software to better ad-

dress some of these diverse end user human aspects.

Examples we were given included: the need to de-

velop better development processes to improve target

end user involvement in software development; pro-

vide developers with better guidelines and practices

to follow to address diverse end user human aspects;

better requirements capture and human aspect mod-

elling support; AI-based tools to automatically advise

on missing end user human aspect issues; more live

testing with representative end users; a need for bet-

ter education of software engineers about diverse end

user human aspects and their impact on software us-

age; simpler interfaces in software for many end user

groups; better defect reporting to enable end users

to more easily identify, describe and report problems

they have with their software; better participant re-

cruitment approaches ensuring more diverse end users

are included; better design frameworks and tooling to

address a range of end user human aspects; and more

research into human aspects in software engineering.

To address some of these issues, we have been de-

veloping the techniques and tools outlined in this talk,

among many others. Key examples include:

• improved processes to including diverse end user

perspectives – living lab based approaches

• better identiﬁcation of stakeholders and improv-

ing capture and modelling of diverse stakeholder

and end user human aspect-related requirements

• improved design and implementation support,

ranging from better tools to model end user hu-

man aspects, include these in software code gen-

erators and run-time adaption frameworks, better

design guidance for developers to address differ-

ent end user human aspects in software, and AI-

based tools to proactively deploy this guidance

• improved human-centric defect reporting tools,

with improved support for developers to locate

human aspect-related defects, identify possible

ﬁxes, make ﬁxes, and inform end users of ﬁxes

made

Some of the key remaining gaps in this area we have

identiﬁed include:

• lack of a taxonomy of end user human aspects

Impact of End User Human Aspects on Software Engineering

including keywords, phrases and examples – this

makes talking about end user human aspects more

difﬁcult and impacts all of the techniques and

tools we want to provide software developers

• lack of studies focusing on how software engi-

neers and software engineering teams address end

user human aspects in software – many design and

HCI works exist but few if any have made their

way into impacting software engineering practice,

for various reasons

• lack of tools to identify challenging end user hu-

man aspects to address during requirements en-

gineering, including extraction, modelling, 3Cs

checking, and validation

• a range of design and evaluation guidelines and

tools but lack of connectivity, consistency, and ap-

plicability of these tools in many domains e.g. for

mobile app development

• overly-complex, inaccessible and incomplete de-

sign and implementation guidelines to address

many challenging end user human aspects

• difﬁculty in end users reporting human aspect de-

fects in software, difﬁculty in software engineers

understanding these defects – or even appreciating

them – and little or no guidelines, techniques and

tools for ﬁxing these human aspect-related defects

when they occur

• development processes and techniques that still

don’t sufﬁciently include diverse stakeholder and

end user perspectives – hence the co-creational

living lab approach we are trialling to address this

• deﬁciencies in the education of software engi-

neers regarding human aspects of their end users

and the need to carefully elicit, understand and

address these – one of our ﬁnal year Bachelor of

Engineering in Software Engineering honors team

members mentioned when working on the adap-

tive UI components “... someone mentioned ac-

cessibility issues once, I think, a few years ago in

a UX lecture...”

Finally, software engineers and software engineering

teams themselves have many human aspects that im-

pact how they work, think about their work, and think

about their software end users, including their end

user human aspects. Software engineers are humans

themselves and thus are impacted by all of the hu-

man aspects end users have, to greater or lesser de-

gree. Exactly how these software developer human

aspects and end user human aspects interplay has not

yet been researched in any way to our knowledge

(Hidellaarachchi et al., 2021). It may be the case that

various differences in developer and end user human

aspects have signiﬁcant impact on how the later issues

are addressed (or not) in produced software.

8 SUMMARY

I have described the need to better take into account

human aspects of end users during software develop-

ment. This includes diverse age, ethnicity, gender,

personality, emotional reactions, engagement, socio-

economic status, language and language skills, cul-

ture, physical and mental challenges, cognitive prob-

lem solving style, and likely many more. These di-

verse human aspects of users can have a very sig-

niﬁcant impact on the usability and ﬁt-for-purpose of

the software. Currently we lack even a taxonomy to

characterise and discuss such diverse human aspects

of software end users. I have outlined some of our

projects to address end user human aspects during re-

quirements engineering, design, model-driven engi-

neering, implementation, defect reporting and defect

ﬁxing. I have described our co-creational living lab

approach aimed at having end users be fully involved

during the whole development process for future soft-

ware systems. This will, we hope, ensure their diverse

human aspects are fully supported in future software

systems.

ACKNOWLEDGEMENTS

Efforts of the HumaniSE team, including Lisa Mc-

Givern, Bran Salic, Anuradha Madugalla, Jen McIn-

tosh, Ingo Mueller, Hourieh Khalajzadeh, Tanjila

Kanij, Waqar Hussain, Humphrey Obie, Mojtaba

Shahin and Rashina Hoda on these and related

projects are greatly acknowledged. Several of these

project outcomes have been produced by ﬁnal year

project teams, Masters and and PhD students whose

enthusiasm for human-centric sofwtare engineering

is wonderful to see. Aspects of this work have been

supported by ARC Discovery Projects DP200100020,

DP170101932 and DP140102185, Industry Transfor-

mation Research Hub IH170100013, and ARC Lau-

reate Fellowship FL190100035.

REFERENCES

Ahmed, A. (2007). Open access towards bridging the

digital divide–policies and strategies for developing

countries. Information Technology for Development,

13(4):337–361.

Ali, N. M., Hosking, J., and Grundy, J. (2013). A tax-

ENASE 2021 - 16th International Conference on Evaluation of Novel Approaches to Software Engineering

onomy and mapping of computer-based critiquing

tools. IEEE Transactions on Software Engineering,

39(11):1494–1520.

Alsanoosy, T., Spichkova, M., and Harland, J. (2019). Cul-

tural inﬂuence on requirements engineering activities:

a systematic literature review and analysis. Require-

ments Engineering, pages 1–24.

Alshayban, A., Ahmed, I., and Malek, S. (2020). Ac-

cessibility issues in android apps: state of affairs,

sentiments, and ways forward. In 2020 IEEE/ACM

42nd International Conference on Software Engineer-

ing (ICSE), pages 1323–1334. IEEE.

Andersen, T. O., Bansler, J. P., Kensing, F., and Moll, J.

(2017). From prototype to product: Making partici-

patory design of mhealth commercially viable. Stud

Health Technol Inform, 233:95–112.

Avazpour, I., Grundy, J., and Grunske, L. (2015). Spec-

ifying model transformations by direct manipulation

using concrete visual notations and interactive recom-

mendations. Journal of Visual Languages & Comput-

ing, 28:195–211.

Avazpour, I., Grundy, J., and Zhu, L. (2019). Engineering

complex data integration, harmonization and visual-

ization systems. Journal of Industrial Information In-

tegration, 16:100103.

Barnett, T., Pearson, A. W., Pearson, R., and Kellermanns,

F. W. (2015). Five-factor model personality traits

as predictors of perceived and actual usage of tech-

nology. European Journal of Information Systems ,

24(4):374–390.

Burnett, M., Stumpf, S., Macbeth, J., Makri, S., Beck-

with, L., Kwan, I., Peters, A., and Jernigan, W.

(2016). Gendermag: A method for evaluating soft-

ware’s gender inclusiveness. Interacting with Com-

puters, 28(6):760–787.

Carcedo, M. G., Chua, S. H., Perrault, S., Wozniak, P.,

Joshi, R., Obaid, M., Fjeld, M., and Zhao, S. (2016).

Hapticolor: Interpolating color information as haptic

feedback to assist the colorblind. In Proceedings of

the 2016 CHI Conference on Human Factors in Com-

puting Systems, pages 3572–3583.

Cruz, S., da Silva, F. Q., and Capretz, L. F. (2015). Forty

years of research on personality in software engineer-

ing: A mapping study. Computers in Human Behav-

ior, 46:94–113.

Curumsing, M. K., Fernando, N., Abdelrazek, M., Vasa,

R., Mouzakis, K., and Grundy, J. (2019). Emotion-

oriented requirements engineering: A case study in

developing a smart home system for the elderly. Jour-

nal of Systems and Software, 147:215–229.

Fensel, A., Tomic, D. K., and Koller, A. (2017). Contribut-

ing to appliances? energy efﬁciency with internet of

things, smart data and user engagement. Future Gen-

eration Computer Systems, 76:329–338.

Garvie, C. and Frankle, J. (2016). Facial-recognition soft-

ware might have a racial bias problem. The Atlantic,

Grundy, J. (2020). Human-centric software engineering

for next generation cloud-and edge-based smart living

applications. In 2020 20th IEEE/ACM International

Symposium on Cluster, Cloud and Internet Computing

(CCGRID), pages 1–10. IEEE.

Grundy, J. and Grundy, J. (2013). A survey of australian hu-

man services agency software usage. Journal of tech-

nology in human services, 31(1):84–94.

Grundy, J. and Hosking, J. (2002). Developing adaptable

user interfaces for component-based systems. Inter-

acting with computers, 14(3):175–194.

Grundy, J., Hosking, J., and Mugridge, W. (1998). Support-

ing large-scale end user speciﬁcation of workﬂows,

work coordination and tool integration. Journal of

Organizational and End User Computing (JOEUC),

10(2):38–48.

Grundy, J., Khalajzadeh, H., and Mcintosh, J. (2020). To-

wards human-centric model-driven software engineer-

ing. In ENASE, pages 229–238.

Grundy, J., Mouzakis, K., Vasa, R., Cain, A., Curumsing,

M., Abdelrazek, M., and Fernando, N. (2018). Sup-

porting diverse challenges of ageing with digital en-

hanced living solutions. In Global Telehealth Confer-

ence 2017, pages 75–90. IOS Press.

Grundy, J. and Zou, W. (2004). Auit: Adaptable user in-

terface technology, with extended java server pages.

Multiple User Interfaces. John Wiley & Sons, New

York, pages 149–167.

Haggag, O., Haggag, S., Grundy, J., and Abdelrazek, M.

(2021). Covid-19 vs social media apps: Does privacy

really matter? 2021 International Conference on Soft-

ware Engineering.

Hazzan, O. and Tomayko, J. E. (2005). Reﬂection and ab-

straction in learning software engineering’s human as-

pects. Computer, 38(6):39–45.

Hidellaarachchi, D., Grundy, J., Hoda, R., and Madampe,

K. (2021). The effects of human aspects on the re-

quirements engineering process: A systematic litera-

ture review. IEEE Transactions on Software Engineer-

ing, (01):1–1.

Hirsch, C., Hosking, J., and Grundy, J. (2010). Vikibuilder:

end-user speciﬁcation and generation of visual wikis.

In Proceedings of the IEEE/ACM international con-

ference on Automated software engineering, pages

13–22.

Hoda, R., Salleh, N., and Grundy, J. (2018). The rise and

evolution of agile software development. IEEE soft-

ware, 35(5):58–63.

Jim, A., Shim, H., Wang, J., Wijaya, L., Xu, R., Khala-

jzadeh, H., Grundy, J. C., and Kanij, T. (2021). Im-

proving the modelling of human-centric aspects of

software systems. In 16th International Conference

on Evaluation of Novel Approaches to Software Engi-

neering (ENASE2021), online, 26-27 April, 2021.

Kanij, T., Merkel, R., and Grundy, J. (2015). An empirical

investigation of personality traits of software testers.

In 2015 IEEE/ACM 8th International Workshop on

Cooperative and Human Aspects of Software Engi-

neering, pages 1–7. IEEE.

Khalajzadeh, H., Simmons, A. J., Abdelrazek, M., Grundy,

J., Hosking, J., and He, Q. (2020). An end-to-

end model-based approach to support big data ana-

lytics development. Journal of Computer Languages,

58:100964.

Khambati, A., Grundy, J., Warren, J., and Hosking, J.

(2008). Model-driven development of mobile personal

Impact of End User Human Aspects on Software Engineering

health care applications. In 2008 23rd IEEE/ACM In-

ternational Conference on Automated Software Engi-

neering, pages 467–470. IEEE.

Kumar, J. (2013). Gamiﬁcation at work: Designing en-

gaging business software. In International conference

of design, user experience, and usability, pages 528–

537. Springer.

Lenberg, P., Feldt, R., and Wallgren, L. G. (2015). Behav-

ioral software engineering: A deﬁnition and system-

atic literature review. Journal of Systems and software,

107:15–37.

Lopez-Lorca, A. A., Miller, T., Pedell, S., Mendoza, A.,

Keirnan, A., and Sterling, L. (2014). One size doesn’t

ﬁt all: diversifying” the user” using personas and emo-

tional scenarios. In Proceedings of the 6th Inter-

national Workshop on Social Software Engineering,

pages 25–32.

Madampe, K., Hoda, R., and Grundy, J. (2020). Towards

better understanding of agile teams through behavior

change models. In 2020 35th IEEE/ACM Interna-

tional Conference on Automated Software Engineer-

ing Workshops (ASEW), pages 53–56. IEEE.

McElroy, J. C., Hendrickson, A. R., Townsend, A. M., and

DeMarie, S. M. (2007). Dispositional factors in inter-

net use: personality versus cognitive style. MIS quar-

terly, pages 809–820.

McManus, J. (2004). A stakeholder perspective within

software engineering projects. In 2004 IEEE Inter-

national Engineering Management Conference (IEEE

Cat. No. 04CH37574), volume 2, pages 880–884.

IEEE.

Miller, T., Pedell, S., Lopez-Lorca, A. A., Mendoza, A.,

Sterling, L., and Keirnan, A. (2015). Emotion-led

modelling for people-oriented requirements engineer-

ing: The case study of emergency systems. Journal of

Systems and Software, 105:54–71.

Newman, P., Ferrario, M. A., Simm, W., Forshaw, S., Fri-

day, A., and Whittle, J. (2015). The role of design

thinking and physical prototyping in social software

engineering. In 2015 IEEE/ACM 37th IEEE Inter-

national Conference on Software Engineering, vol-

ume 2, pages 487–496. IEEE.

Nouwen, M., Van Mechelen, M., and Zaman, B. (2015). A

value sensitive design approach to parental software

for young children. In Proceedings of the 14th Inter-

national Conference on Interaction Design and Chil-

dren, pages 363–366.

Obie, H. O., Hussain, W., Xia, X., Grundy, J., Li, L.,

Turhan, B., Whittle, J., and Shahin, M. (2021). A ﬁrst

look at human values-violation in app reviews. 2021

International Conference on Software Engineering.

Osama, M., Zaki-Ismail, A., Abdelrazek, M., Grundy, J.,

and Ibrahim, A. (2020). Score-based automatic detec-

tion and resolution of syntactic ambiguity in natural

language requirements. In 2020 IEEE International

Conference on Software Maintenance and Evolution

(ICSME), pages 651–661. IEEE.

Perez, C. C. (2019). Invisible women: Exposing data bias

in a world designed for men. Random House.

Rashid, Z., Meli

a-Segu

ı, J., Pous, R., and Peig, E. (2017).

Using augmented reality and internet of things to im-

prove accessibility of people with motor disabilities in

the context of smart cities. Future Generation Com-

puter Systems, 76:248–261.

Roturier, J. (2015). Localizing Apps: A practical guide for

translators and translation students. Routledge.

Shamsujjoha, M., Grundy, J. C., Li, L., Khalajzadeh, H.,

and Lu, Q. (2021). Human-centric issues in ehealth

app development and usage: A preliminary assess-

ment. In 28th IEEE International Conference on Soft-

ware Analysis, Evolution and Reengineering (SANER

‘21), ERA Track, Online, 9-12 March, 2021. IEEE.

Sierra, J. S. and Togores, J. (2012). Designing mobile apps

for visually impaired and blind users. In The Fifth

international conference on advances in computer-

human interactions, pages 47–52. Citeseer.

Stock, S. E., Davies, D. K., Wehmeyer, M. L., and Palmer,

S. B. (2008). Evaluation of cognitively accessi-

ble software to increase independent access to cell-

phone technology for people with intellectual dis-

ability. Journal of Intellectual Disability Research,

52(12):1155–1164.

Strengers, Y. and Kennedy, J. (2020). The Smart Wife: Why

Siri, Alexa, and Other Smart Home Devices Need a

Feminist Reboot. MIT Press.

Taveter, K., Sterling, L., Pedell, S., Burrows, R., and

Taveter, E. M. (2019). A method for eliciting and

representing emotional requirements: Two case stud-

ies in e-healthcare. In 2019 IEEE 27th Interna-

tional Requirements Engineering Conference Work-

shops (REW), pages 100–105. IEEE.

Williams, D., Alam, M. A. U., Ahamed, S. I., and Chu, W.

(2013). Considerations in designing human-computer

interfaces for elderly people. In 2013 13th Interna-

tional Conference on Quality Software, pages 372–

377. IEEE.

Winter, E., Forshaw, S., and Ferrario, M. A. (2018). Mea-

suring human values in software engineering. In Pro-

ceedings of the 12th ACM/IEEE International Sym-

posium on Empirical Software Engineering and Mea-

surement, pages 1–4.

Woodson, W. E., Tillman, B., and Tillman, P. (1992). Hu-

man factors design handbook: information and guide-

lines for the design of systems, facilities, equipment,

and products for human use.

Yusop, N. S. M., Grundy, J., Schneider, J.-G., and Vasa, R.

(2018). Preliminary evaluation of a guided usability

defect report form. In 2018 25th Australasian Soft-

ware Engineering Conference (ASWEC), pages 81–

90. IEEE.

Yusop, N. S. M., Grundy, J., Schneider, J.-G., and Vasa, R.

(2020). A revised open source usability defect classi-

ﬁcation taxonomy. Information and software technol-

ogy, 128:106396.

Yusop, N. S. M., Grundy, J., and Vasa, R. (2016). Re-

porting usability defects: a systematic literature re-

view. IEEE Transactions on Software Engineering,

43(9):848–867.

Zhao, D., Xing, Z., Chen, C., Xu, X., Zhu, L., Li, G., and

Wang, J. (2020). Seenomaly: Vision-based linting of

gui animation effects against design-don’t guidelines.

In 42nd International Conference on Software Engi-

neering (ICSE’20). ACM, New York, NY.

ENASE 2021 - 16th International Conference on Evaluation of Novel Approaches to Software Engineering