AGILE DEVELOPMENT OF INTERACTIVE INSTALLATIONS

Two Case Studies

Pedro Campos

CCM, University of Madeira and INESC ID Lisbon

Campus Universitário da Penteada, 9000-390 Funchal, Portugal

Keywords: Human-Computer Interaction, Interactive Installations, Agile Development, User-Centered Design.

Abstract: We argue that agile methods can be particularly effective when designing and developing interactive instal-

lations, as long as the agile methods are correctly tailored to this application domain. Based on significant

experience, which was built upon ethnographic observation and participation in about a dozen industrial

projects related to interactive installations’ design and development, we present agile strategies which

proved effective when dealing with the industry’s typical tight production schedules, and we also provide

the data from two case studies, discussions and conclusions. Using real world case studies such as these, re-

searchers can obtain more insight into best practices that could be useful for promoting innovation during

the agile process.

1 INTRODUCTION

Agile Software Development (ASD) is a software

engineering approach that has come to an age of

relative maturity, while keeping itself open to inte-

gration or cross-pollination among approaches that

previously represented sharply opposing postures

(Ergdomus, 2007).

On the other hand, the principles underlying the

ASD movement, firstly advocated by a set of several

practitioners, are based on concrete best practices,

which were outlined through a large body of

knowledge. That body of knowledge included

essentially a vast amount of success and failure case

studies and experiences with many software

projects, in many different application domains.

While each practitioner has an individual

perspective about how to successfully approach

software development, there were key factors, which

seemed common to each and every project. These

included tight collaboration between developers and

business managers, face-to-face, informal

collaborations as opposed to bureaucratic production

of project documentation and formal modelling,

frequent iterations and rapid deliveries of functional

prototypes, embracing requirements’ change as

opposed to well-defined, fixed requirements right at

the start of a project, as well as other principles that

have become widely known by the general software

engineering community (Fowler and Highsmith,

2001).

Since the early Agile Manifesto days, there has

been a growing number of research papers around

this theme, some of them focusing on the success

factors brought by the agile practices, such as

customer commitment derived from early and

continuous delivery of valuable software, decision

time, likely made within short timeframes, corporate

culture (agile methodologies are not appropriate to

bureaucratic organizations (Cockburn, 2002)), and

dynamism and uncertainty, i.e. being able to handle

dynamism and the uncertainty built in it.

There are also many tailored approaches and

variants of agile development. The most popular

ones include Extreme Programming (XP) (Hedin et

al., 2003), SCRUM (Schwaber, 2004) and Crystal

Methodologies (Cockburn, 2002).

However, there is currently a lack of agile

strategies to designing a novel kind of software:

software built to support interactive installations

such as interactive floors, walls, multi-touch tables

and - in general - software for novel interaction

paradigms. This new kind of software is quite

different from the traditional GUI-oriented software,

which is essentially based in WIMP (Windows,

Icons, Menus and Pointing Devices) (Want and

Pering, 2005) user interfaces.

Based on a significant experience, which was

174

Campos P. (2010).

AGILE DEVELOPMENT OF INTERACTIVE INSTALLATIONS - Two Case Studies.

In Proceedings of the 12th International Conference on Enterprise Information Systems - Human-Computer Interaction, pages 174-179

DOI: 10.5220/0002956401740179

 SciTePress

built upon ethnographic observations and

participation in about a dozen industrial projects

related to interactive installations' design and

development, we present agile strategies which

proved effective when dealing with the industry's

typical tight production schedules, and we also

provide the data from two case studies, discussions

and conclusions.

Our contribution is two-fold: (i) we present the

results from applied case studies, following the spirit

of the agile movement - which was itself based on a

practical experience body of knowledge - in a novel

software development context that is becoming

increas-ingly important: multimodal user interfaces

(Klein and Myers, 2005) sometimes also referred to

as ubiquitous computing (Latoschik, 2005); and (ii)

we describe a novel set of agile strategies aimed at

supporting this development context in the tourism

market. Both of these contributions are, to our

knowledge, original, although several researchers

have applied similar research methods to similar

domains.

2 BACKGROUND

2.1 Agile Methods

Agile methodologies fit nicely into the context of

small software development companies, since they

evolved as a reaction against the so-called 'heavy-

weight' methods, which are regarded as bureaucratic

and slow, like the waterfall model. The main idea

advocated by agile practitioners, is that short itera-

tions turn the methods more responsive to changes

in the environment - this is especially useful in the

application domain this paper is focused on. In par-

ticular, artists and interior designers, who were

stakeholders in the projects we will describe, have a

particular work style, which is based on constant

change as a way to create better results. It's a cultural

and professional issue that clashes against the more

formal and compartmented software engineering

process.

The agile approach quickly became mainstream

in the software industry. The agile community is

defined by a core set of beliefs and practices, in a

“practice what you preach” philosophy. The Mani-

festo for Agile Software Development, the most

well-known conjugation of agile principles and be-

liefs, states:

“(...) we have come to value:

• individuals and interactions over processes and

tools

• working software over comprehensive documen-

tation

• customer collaboration over contract negotiation

• responding to change over following a plan”

Agile methods pay significant attention to users

and their needs. They bring users to the development

process from the early stages, when the requirements

are discovered, and give them an opportunity to

speak up and say what they really need and want.

There are some studies, such as (Mitra and

Gupta, 2005), regarding how innovation can be

brought to the XP and Agile processes. However,

little research is based on extensive experience and

observations from real world case studies. Innova-

tion and agility are essential factors to supporting the

creation of value in a fast-paced global knowledge

economy.

2.2 Novel Interaction Paradigms

Interactive installations based on novel interaction

paradigms; such as gesture recognition or multi-

touch surfaces are typically designed through User-

Centred approaches, the so-called UCD methods

(Vredenburg et al., 2002). There are, however, many

similarities between the XP practices and UCD.

What XP argues as iterations, small increments,

UCD advocates as prototyping. The XP “Planning

game” is very similar to the UCD concept of “Focus

Groups”, which are essentially, focused discussions

where a moderator leads a group of participants

through a set of questions on a particular topic (Dix

et al., 2004). XP’s story cards, task cards and user

stories are very alike to UCD’s scenarios, user roles

and task models. The XP practice of having an on-

site customer representative is analogous to the UCD

notion of user-centred design, or user participation.

The UCD equivalent of XP’s tests (or test stories) is

the evaluation session or usability inspections. And

finally, XP’s metaphors correspond to UCD’s con-

ceptual and mental models (Vredenburg et al.,

2002).

2.3 Innovation Processes

Creativity support tools have the power to accelerate

discovery and innovation (Shneiderman, 2007). The

question is posed in terms of how can designers of

programming interfaces, interactive tools, and rich

social environments enable more people to be more

creative more often (Shneiderman, 2007).

Ben Shneiderman, one of the most prominent

leaders of the human-computer interaction field,

advocates that Leonardo da Vinci could help as an

AGILE DEVELOPMENT OF INTERACTIVE INSTALLATIONS - Two Case Studies

175

inspirational muse for the new computing (Shnei-

derman, 2005). Shneiderman says his example could

push designers to improve quality through scientific

study and more elegant visual design. Leonardo’s

example can guide us to the new computing, which

emphasizes empowerment, creativity, and collabora-

tion.

Shneiderman (2000) also proposes a four-stage

framework for creativity that can assist designers in

producing the right tools their users: (1) Collect:

learn from previous works stored in libraries, the

Web, etc.; (2) Relate, consult with peers and mentors

at early, middle, and late stages; (3) Create: explore,

compose, evaluate possible solutions; and (4) Do-

nate: disseminate the results and contribute to the

libraries. He also emphasizes that “Education could

expand from acquiring facts, studying existing

knowledge, and developing critical thinking, to in-

clude more emphasis on creating novel artefacts,

insights, or performances.” (Shneiderman, 2000).

Some successful examples of creativity and in-

novation processes come from universities, espe-

cially examples where cross-disciplinary design re-

search is involved. Ellen Yi-Luen Do and Mark

Gross (2007) engaged their students in this line of

action and describe parameters and principles that

they found helpful in organizing and conducting this

kind of work. A variety of projects that have been

developed in their group illustrated their parameters

and principles. The focus is on making and they

have come to see creativity as grounded in the abil-

ity to make things.

3 METHOD

This research was mainly carried out in the company

WowSystems, while the case studies we focus on

happened at the World Expo Zaragoza 2008 (Spain),

and at a Cultural Exhibition in Madeira (Portugal).

WowSystems is specialized in new digital media,

novel interaction paradigms and interactive installa-

tions. Because of the very nature of its core business,

innovation and agility are main concerns of the

company.

We followed an interpretive research approach

(Walsham, 1995). Interpretive case studies can make

a valuable contribution to Information System (IS)

theory and practice and the volume and range of

such studies are limited. Some researchers (Hedin et

al., 2003) agree that there is a need for more inter-

pretive stances in the future in the IS field and soft-

ware engineering as well.

This means that the analysis of data was based

“on understanding a complex whole from precon-

ceptions of about the independent meanings of its

parts and their interrelationships” (Klein and Myers,

1999). We also followed in some way the spirit of

ethnographic research, taking field observations and

tracking artefacts such as post-it notes, desktop

items, whiteboard collaborative writing sessions and

similar ones. Ethnographic analysis is derived from

anthropology. Field observations are taken at a site

of a possible user. These observations also gather the

sequence of work and interruptions that determine

the user’s typical day.

Our method included the gathering of data re-

lated to (i) participant observations, in an ethno-

graphic study manner, over the course of the pro-

jects; (ii) semi-structured interviews to stakeholders

and (iii) informal meetings and discussions. The

following are the time periods and projects the au-

thor participated, in order to conduct the research

following the previously mentioned method:

• Cultural Interactive Exhibition: January to April

2008;

• Portuguese Pavilion in Expo Zaragoza, Spain:

April to September 2008;

The meetings involved a large assortment of pro-

fessions: programmers, visual de-signers, project

managers, interior architects, artists, government

persons, marketing personnel and researchers. We

took notes during the observation sessions, and au-

dio-recorded some of the interviews. Photos were

taken at relevant and/or interesting moments

throughout the development. The authors were also

able to participate in informal meetings.

Following the ethnographic research spirit, we

also conducted several in-depth structured inter-

views with developers and engineers, project man-

agers, and customer representatives, which included

the professions mentioned before.

4 CASE STUDIES

WowSystems had chosen to follow agile develop-

ment methods in all projects developed by the com-

pany, since it fitted very well into the tight schedules

the clients demanded. There were, however, two

major concerns expressed by the developers: how to

achieve innovative solutions and how to deal with

the communication issues. These communication

issues were derived not only from the background

differences between artists and engineers but also

from the very nature of interactive installations as

“physical pieces” of software, sometimes even cal-

ICEIS 2010 - 12th International Conference on Enterprise Information Systems

176

led as tangible user interfaces (Ishii, 2008).

We will begin by analysing the Cultural Interac-

tive Exhibition, and afterwards we will describe the

Tourism Information Office case study. At the end

of each subsection, we provide a detailed discussion

organized around the main challenges and processes;

in particular we outline how agile processes should

be tailored in order to be more effective when de-

signing interactive installations like these.

4.1 Cultural Interactive Exhibition

In April 2008, WowSystems designed a set of sen-

sor-based installations in a cultural exhibition organ-

ized by the Direction of Cultural Affairs, which

aimed at showing the visitor the cultural richness

that formed the streets of Funchal (Portugal). The

concepts of the exhibition revolved around promot-

ing awareness about, and foster a better under-

standing of, the cultural tourism that can be per-

formed by simply walking through strategic streets

and watching certain buildings, sites, and heritage.

To better complement the exhibition’s traditional

large-format printed panels, the organizers wanted to

have the interactive factor as a means to add value to

the visitor’s experience.

The final set of installations included: (i) a vir-

tual encyclopaedia that could be browsed by simple

page-flipping gestures performed in mid-air; (ii) an

interactive floor that illustrated the evolution of the

transportation means along the years; (iii) an interac-

tive timeline using a touch-screen and (iv) a panel

with projected images that would change through

waving. These installations are shown in Fig. 1.

Figure 1: Innovation examples for cultural heritage and

museums: the installations and interaction styles employed

throughout the 2008 “Cultural Tourism” exhibition

(clockwise from top left): page-flipping, walking over,

touching and waving.

The use of sensor-based interactive installations,

in particular installations involving infrared motion

sensors as well as cameras coupled with real time

video processing algorithms, have been receiving

considerable interest both from industry and acade-

mia (Hornecker and Stifter, 2006) During the design

and evaluation of interactive exhibitions, much can

be learned about interaction design for public set-

tings like these.

An important principle, upon which the agile

methodologies are based, is the close relations be-

tween developers and users or customers. Being

agile means giving priority to customer's satisfaction

through early and continuous delivery of software

where changes are appreciated. The main customers

for this Culture Interactive Exhibition were archi-

tects, artists and designers who were interested in

conceiving the best possible ways to provide an in-

teresting exhibition. The design and development

processes were therefore a collaborative effort be-

tween WowSystems’ team and these user groups.

“Change was constant - and communication was

a true challenge, since it was difficult for them to

communicate us the whole point of. And when we

moved from the laptop to an actual kiosk or projec-

tion, we noticed how different their opinion was re-

garding every aspect of the design and development”

(Developer/Designer, WowSystems).

Another interesting observation was:

“They were completely focused on the MS

PowerPoint model - they thought kiosks and interac-

tive installations had to be designed as if they were

PowerPoint presentations” (Developer, WowSys-

tems).

The experience of designing and evaluating an

interactive exhibition featured four different interac-

tion styles to control digital contents: touching,

walking over, waving and page-flipping. The design

approach was based on tailoring the interaction

styles to the exhibition’s contents and making a

creative use of sensor-based technology, with the

explicit goal of reducing the distance between visi-

tors and cultural heritage.

While some of the more than fifty interactive in-

stallations already deployed were solely created as

experiential activities, providing an increase in the

level of learning by adding facts to an already well-

formed conceptual model, others were designed to

enact a reflective activity, thus supporting a restruc-

turing learning where new conceptual frameworks

need to be built. Another issue that drives the devel-

opment team is the observation of the visitors’ and

users’ behaviours, particularly finding out how col-

AGILE DEVELOPMENT OF INTERACTIVE INSTALLATIONS - Two Case Studies

177

laborative activities can be supported as feedback

mechanisms to enhance engagement and learning

motivation.

Generally, the observations were focused around

four issues: (i) usability, how easy and intuitive it is

to interact with the products? (ii) interaction model,

i.e. how did the interaction model was learned and

reapplied, (iii) social interaction, which types of in-

ter-action triggered more collaborative activity and

how did this activity affect the performance of the

interaction and finally (iv) learning effectiveness,

did visitors actually learn anything? Whether it’s for

a museum, tourism or a brand, all these issues apply.

If we’re designing for a brand, the visitor should

learn all about it and memorize it. If we’re de-

signing for tourism then the destination itself is the

brand. And if we’re designing for a museum or a

science park, learning is one of the most important

goals to be attained by the product’s usage.

Particularly important for designing for innova-

tion is the interaction model and how it is learned

and reapplied. If the interactive product is too inno-

vative, then it could be difficult to learn at first hand.

On the other hand, conventional, well-established

models of interaction (e.g. touch-screen kiosk) aren’t

innovative and therefore we can conclude that there

is, naturally, a dichotomy between the learning curve

and innovation degree of an interactive product.

4.2 Tourism Booth, Expo Zaragoza’08

Another similar innovation case study was the instal-

lation at the Portuguese Pavilion in Expo Zaragoza,

Spain, in 2008. WowSystems was commissioned by

the Tourism Board to design, develop and install an

interactive floor of 5x2m at Expo Zaragoza, the

world’s largest exhibition, dedicated to “Water and

sustainable development”. The installation was a

recreation of the famous Madeira “Levadas”, the

name given to watercourses built by man in order to

carry the water from the mountains down to the vil-

lages.

Figure 2 illustrates the final product: we can see

a picture of a visitor walking over an interactive rec-

reation – an interactive floor – that reacted to the

user’s steps or gestures. Innovation was once again

present, and the team wanted to achieve something

more significant, but very rapidly and in an agile

way. Since the idea was to recreate the environment,

and since that idea influenced the design process, the

team added the true sounds of the forest’s bird spe-

cies, and even added a “scent projector” that spread

the scents and aromas of the forest as well. This

way, the visitor could really immerse herself into the

scenery, in a multi-sensorial experience.

Figure 2: The virtual walkway.

The innovation process followed in this project

was similar to the previous one. Like the previous

project, one of the innovative ideas that were put

into practice was the alignment of the interaction

style to the message being conveyed by the product:

In this case, visitors of the pavilion get to know the

“levadas” the same way they would as if they were

visiting the real ones - by walking over the interac-

tive “levada”. However, some significant differences

occurred because the goal was to bring a little bit of

life to the pavilion. The innovation process that was

undertook in this case, was to brainstorm about how

to recreate a touristic site using technology. After

two sessions, one member of the design team pro-

posed to add sounds and scents to the installation,

and this turned out to be an aspect that the end users

and the client both appreciated, much more than

expected.

5 CONCLUSIONS

In this paper, we focused on describing the innova-

tion aspects that occur using an agile development

approach, when creating software products that em-

ploy novel user interaction paradigms. Using real

world case studies such as these, one can obtain

more insight into best practices that could be useful

for promoting innovation during the agile process.

Incubators, applied research centres, company’s

R&D departments, and innovation centres in gen-

eral: All these can benefit from learning other com-

panies’ experiences and projects.

Interactive, Digital Media projects are often con-

ducted by multi-disciplinary teams that usually in-

clude programmers, software engineers, project

managers, interior designers, architects, graphical

artists, and a high-level of client involvement.

ICEIS 2010 - 12th International Conference on Enterprise Information Systems

178

A great difficulty arises when the final product is

actually deployed: interactive installations are diffi-

cult to prototype and many aspects are impossible to

model and test by means of early prototypes. This

doesn’t happen with, e.g. mobile applications, where

the designer has full access to the end product look

& feel anytime and anywhere. Since innovative

products are much harder to propose to clients than

conventional products, this difference between early

prototypes and final product is a significant chal-

lenge for small companies, since it involves a large

degree of imagination to describe.

Finally, many governmental programs, which are

aimed at sparking innovative companies and sup-

porting creative entrepreneurs, entail a large degree

of bureaucratic forms, laws, regulations and similar

“red tape” that are easy creativity-killers. This is a

major difficulty should be addressed. Facilitating or

diminishing the amount of red tape in incentives

programs doesn’t necessarily diminishes the credi-

bility and transparency of the program.

One of the limitations of a study like this is re-

lated to measuring results: it is not easy, in a busi-

ness context, to effectively measure innovation

processes or even to explain exactly what went well

and what went wrong during the agile process. Much

of the practitioners’ knowledge is tacit, in the sense

that they don’t know themselves how they accom-

plish their everyday tasks. Only through extensive

observation in situ can researchers obtain a clearer

picture on the innovation processes that companies

and research institutes follow and consequently how

can software developers take appropriate measures

in order to improve their competitiveness and effi-

ciency. This is, however, a first step towards that

goal.

Another limitation of our experience is that it

considers only the perspective of an interactive digi-

tal media company. It would be very interesting to

compare this experience to other businesses and to

other research fields, since some conclusions can be

transversal to the research field.

REFERENCES

Cockburn, A. (2002). Agile software development, Add

son-Wesley Longman Publishing Co., Inc., Boston,

MA, 2002.

Dix, A., Finly, J., Abowd, G. and Beale, R. (2004). Hu-

man-Computer Interaction. Prentice-Hall, 3rd edition.

Do, E. Y. and Gross, M. D. (2007). Environments for

creativity: a lab for making things. In Proceedings of

the 6th ACM SIGCHI Conference on Creativity &

Cognition (Washington, DC, USA, June 13 - 15,

2007). C&C ‘07. ACM, New York, NY, 27-36.

Ergdomus, H. (2007). Agile's Coming of Age... or Not.

IEEE Software, Nov.-Dec. 2007.

Fowler, M. and Highsmith, J. (2001). The Agile Mani-

festo. Software Development Magazine. August 2001.

Hedin, G., Bendix, L., and Magnusson, B. (2003). Intro-

ducing software engineering by means of Extreme

Programming. In Proceedings of the 25th interna-

tional Conference on Software Engineering. IEEE

Computer Society, Washington, DC, 586-593.

Hornecker, E. and Stifter, M. (2006). Learning from inter-

active museum installations about interaction design

for public settings. In Proceedings of the 18th Austra-

lia Conference on Computer-Human interaction. J.

Kjeldskov and J. Paay, Eds. OZCHI’06, vol. 206.

ACM.

Ishii, H. (2008). The tangible user interface and its evolu-

tion. Communications of the ACM, 51, 6 (Jun. 2008),

32-36.

Schwaber, K. (2004). Agile Project Management with

Scrum. Microsoft Press. ISBN 978-0-735-61993-7.

Shneiderman, B. (2005). Leonardo’s laptop: human needs

and the new computing technologies. In Proceedings

of the 14th ACM international Conference on informa-

tion and Knowledge Management (Bremen, Germany,

October 31 - November 05, 2005). CIKM’ 05.

Shneiderman, B. (2007). Creativity support tools: accele-

rating discovery and innovation. Communications of

the ACM, 50, 12 (Dec. 2007), 20-32.

Shneiderman, B. (2000). Creating creativity: user interfac-

es for supporting innovation. ACM Transactions on

Computer-Human Interaction, 7(1) (Mar. 2000), 114-

138.

Want, R. and Pering, T. (2005). System challenges for

ubiquitous & pervasive computing. In Proceedings of

the 27th international Conference on Software Engi-

neering (St. Louis, MO, USA, May 15 - 21, 2005).

ICSE’05. ACM, New York, NY, 9-14.

Latoschik, M. E. (2005). A user interface framework for

multimodal VR interactions. In Proceedings of the 7th

international Conference on Multimodal interfaces

(Torento, Italy, October 04 - 06, 2005). ACM, New

York, NY, 76-83.

Klein, H., & Myers, M. (1999). A set of principles for

conducting and evaluating interpretive field studies in

information systems. MIS Quarterly, 23(1), 67–94.

Mitra, A. and Gupta, A. (2008). Knowledge Reuse and

Agile Processes: Catalysts for Innovation. Information

Science Reference - Imprint of: IGI Publishing.

Vredenburg, K., Mao, J., Smith, P. W. and Carey, T.

(2002). A survey of user-centered design practice. In

Proceedings of the SIGCHI Conference on Human

Factors in Computing Systems: Changing Our World,

Changing Ourselves. CHI ‘02. ACM, New York, NY,

471-478.

Walsham, G. (1995). Interpretive case studies in IS re-

search: Nature and method. European Journal of In-

formation Systems, 4, 74–81.

AGILE DEVELOPMENT OF INTERACTIVE INSTALLATIONS - Two Case Studies

179