SCENARIO-BASED DESIGN - AN ESSENTIAL INSTRUMENT FOR

AN INNOVATIVE TARGET APPLICATION

Case Report

L. S. Garc

ıa, A. I. Direne, M. A. Castilho, L. E. Bona, F. Silva and M. S. Sunye

Departamento de Inform

atica, Universidade Federal do Paran

a, Jardim das Am

ericas, Curitiba, Brazil

Keywords:

Scenario, System-design, User-centered.

Abstract:

Scenario-based design is a largely accepted method within the literature on Human-Computer Interaction

and, for certain cases, also within the literature on Software Engineering. However, the lack of integration

between these two areas, in addition to the lack of attention paid to the actual (and still quite infrequent)

use of scenario-based design, stresses the need for increased emphasis on the relevance of scenario-based

design applied to projects of truly innovative technological artefacts. In the present paper, we will present a

case report whereby the above mentioned method, when applied to problem-analysis, led to the project of a

differential user-interface environment when compared to the human process prior to the introduction of the

computational application.

1 INTRODUCTION

Scenario-based design is a largely accepted method

within the literature on Human-Computer Interaction

(HCI) and, for certain cases, also within the litera-

ture on Software Engineering (SE) behind object ori-

ented use-cases (Button and Sharrock, 1994; Cock-

burn, 1997; Sommerville, 2000; Bittner and Spence,

2002; Bezerra, 2007). However, the lack of formal,

solid integration between these two areas (i.e. HCI

and SE), in addition to the lack of attention paid to

the actual (and still quite infrequent) use of scenario-

based design, stresses the need for increased emphasis

on the relevance of scenario-based design applied to

projects of truly innovative technological artefacts.

A scenario is a story with characters, events, prod-

ucts and environments. They help the designer to ex-

plore ideas and the ramiﬁcations of design decisions

in particular, concrete situations. Storyboards, one al-

ternative representation for scenarios, are sequences

of snapshots which focus on the main actions in a

possible situation. By using these techniches, design-

ers are able to move from existing (documented) to

potential interactions and hence to antecipate prob-

lems. (Preece et al., 1994) Scenarios force design-

ers to consider the range of users who will use the

system and the range of activities for which they will

use it. The user space is deﬁned by the variety of

users, work and environment in which the interaction

will take place. Scenarios are used to male concrete

particular combinations of these dimensions. (Tog-

nazzini, 1992 apud Preece et al., 1994) A signiﬁcant

number of authors claim thar scenarios help extract-

ing requirements (Achour, 1999; B”dker, 1999; Dour-

ish, 2001; Suchman, 1987).

In the present paper, we will present a case report

whereby the focus on the development of scenario-

based design applied to the activity to be supported by

the application led to the project of a differential user-

interface environment when compared to the human

process prior to the introduction of the computational

application.

The present case report concerns the activities

of a non-governmental organisation of a develop-

ing country whose main objective is to lower the

countrys infant mortality rates, particularly in the

poorer areas. This Institution has a hierarchical struc-

ture, namely a National Coordination (which is made

up of experts from several different knowledge ar-

eas, amongst which are Paediatrics, Gynaecology,

Obstetrics, Epidemiology and Nutrition) and volun-

tary workers (community leaders). Other organisa-

tions also support the Institution such as universi-

ties, governmental organisations, as well as other non-

governmental organisations in the ﬁeld of health and

other areas of interest and so do individual volun-

teers mainly researchers and technicians who dedi-

cate part of their time to the initiatives of the Institu-

tion. All potential contributors (both individuals and

enterprises) may take part in the institutional activi-

113

García L., Direne A., Castilho M., Bona L., Silva F. and S. Sunye M. (2009).

SCENARIO-BASED DESIGN - AN ESSENTIAL INSTRUMENT FOR AN INNOVATIVE TARGET APPLICATION - Case Report.

In Proceedings of the 11th International Conference on Enterprise Information Systems - Human-Computer Interaction, pages 113-117

DOI: 10.5220/0002005101130117

 SciTePress

ties, with equal rights.

In addition to the organisational structure and the

community leaders relentless work, what actually en-

sures the efﬁciency and effectiveness of the Institu-

tions work is the Leaders Guide. The Leaders Guide

is a technical-instructive manual divided into chap-

ters dedicated to crucial issues concerning the ﬁght

against infant mortality and, ultimately, the achieve-

ment of full civil rights for all community members.

Guided by this material, the leaders go through their

communities and instruct families mainly about ba-

sic care during pregnancy and the neonatal period,

as well as about the main measures against infant

malnutrition. However, health and nutrition are not

the only topics included in the Leaders Guide, see-

ing as other factors also contribute to the subhuman

life some of these families lead, hence worsening any

possible cases of infant malnutrition.

The National Coordination, together with com-

munity representatives, other organisations and indi-

vidual volunteers, put the Leaders Guide together a

couple of years ago. Due to its dynamic nature (a

certain eradicated disease may afﬂict the population

once again, or a new epidemic may call national at-

tention), the National Coordination updates the mate-

rial on a yearly basis. The updates are based upon

suggestions collected throughout the previous year

through personal contacts, via telephone, fax and e-

mail. They also take into account the suggestions

made at their Annual Conference, where the new is-

sue of the Leaders Guide is only one of many topics

on the agenda. The amount of effort put into these

updates is, therefore, signiﬁcant, since all suggestions

obtained through several different means of commu-

nication must be compiled, after which they must be

discussed and passed, one by one, during the Annual

Conference. Despite the colossal amount of work it

takes, these annual updates ensure the efﬁciency and

effectiveness of the Institutions initiatives, as its Co-

ordinators assert and reassert every year.

According to the potential users, the main prob-

lem that led to the application yet to be developed

consisted of the difﬁculties which, in turn, comes

be easily solved by WEB applications, namely the

geographic distribution within such a large country

(which makes it difﬁcult for contributors to get to-

gether) and the precariousness of the means of com-

munication available to collect all suggestions (which

prevents people to contribute as often as they could).

In this context, the initial objective was to develop

a WEB environment for successfully collecting these

suggestions.

2 METHODOLOGY

We based our methodology on the development of the

scenario-based design according to the hypothetical

problems and the main topic of the present project, i.e.

collecting suggestions (data entry) across the country.

This scenario was developed to ﬁt both the existing

human procedures pointing, additionally, to proce-

dural hindrances not originally reported and the pro-

posed WEB environment gradually revealing the in-

novative potential of the technological artefact, restat-

ing the cycle of technology evolution (Carroll, 2003).

The development of sub-scenarios associated to

data entry, which in turn was made possible through

the analysis of the documents associeted to the human

procedures of suggestion collection already described

above, dynamically delineated our methodology. In-

deed, thanks to this methodology, we were able to sig-

niﬁcantly improve both the process of data entry (en-

try of suggestions, discussion and decision) and the

product itself (the WEB application).

The methodology consisted of the following steps:

1. Identiﬁcation of the Several Elements which

make up a Chapter of the Leaders Guide. This

step consisted of the exhaustive analysis of the

suggestions included in the documents available.

During this step, we learned that a typical chap-

ter always contains the following elements: a ti-

tle, subtitles, a biblical quotation, initial call to

the leaders, the body of the text (in paragraphs),

photos with keys, closing call to the leaders (see

Figure 1). We also noticed that only the elements

belonging to the middle of the chapters (i.e. para-

graphs and photos) could be repeated and ran-

domly interposed. Finally, all pages contained

contextualisation elements, such as chapter title

and page number;

2. Identiﬁcation of Suggested Actions for each

Kind of Chapter Element. At this point, we re-

alised that for the text itself there were suggestions

concerning the style, the spelling and the register

of the text (such as the substitution of sentences

as, for instance, Oral hygiene must be carried out

with the toothbrush at an angle of 45 to the teeth,

for simpler, easily understandable sentences con-

sidering the target public). As for the photos,

some suggestions referred both to technical as-

pects (such as contrast problems) and to their con-

tent itself (as, for example, the replacement of the

photo of a female leader for the photo of a male

leader, so as to discourage the discrimination of

men who engage themselves in the project). We

associated each of the chapter elements to a set

of actions, all derived from the documents on the

ICEIS 2009 - International Conference on Enterprise Information Systems

114

human procedures of data collection;

3. “Semantic regularity” of the actions. The main

objective of this step was to complete, from the

point of view of semantic expressiveness, the set

of actions ascribed to each chapter element. The

main reason why this step is so essential is the

fact that the documents analysed were the result

of one instance of the process and, therefore, were

incomplete in terms of time. This is why some of

the additional actions were complementary, con-

troversial, etc.;

4. Analysis of the Actions Ascribed to each Chap-

ter Element so as to Determine its Relevance to

other Chapter Elements. An analogous reason

justiﬁes this step. One example is spelling correc-

tion, which is relevant not only for the body of the

text, but also for titles, keys, references, etc.;

5. Identiﬁcation and Classiﬁcation of other Po-

tential Suggestion Targets. Through this step we

mainly established that, in addition to the content

of the chapters, a potential suggestion target could

be the thematic focus of the chapter. Furthermore,

we assessed the complexity and comprehensive-

ness of the topics on the one hand we have topics

to which more than one chapter is dedicated and,

on the other hand, we have topics whose nature

is interdisciplinary. Indeed, this additional insight

revealed the inherent, underlying difﬁculty in the

process of suggestion entry;

6. Analysis of Ambiguous Suggestions made by

Community Leaders during the Annual Con-

ference. This was one of the most relevant steps

because the leaders, despite having an enormous

potential to help due to their proximity to commu-

nity members, are mostly poorly educated people.

We found two kinds of ambiguity. The ﬁrst was

the omission of the action to be carried out in re-

lation to the object mentioned, such as “branchial

clef”, “pregnancy card”, “Gertrude, the doll” (doll

used for teaching the sexual organs), all of which

are noun phrases used in chapters content of the

Leaders Guide. The second kind was named total

ambiguity, i.e. suggestions whose objects could

not be identiﬁed, such as “inﬂuence of the envi-

ronment on childrens health” (nothing about that

topic is mentioned in the Guide), “diet for babies

over six months-old” (the Guide does not men-

tion that either), and “How to use enriched ﬂour”.

The main objective of the present step was to de-

ﬁne which kind of information the system should

display, as well as which kind of information it

should required so as to minimise the occurrence

of ambiguities in the virtual environment;

7. Design of the User-interface Environment.

Based on all previous steps, we carried out this

step aiming at developing a truly innovative tech-

nological tool. This way, by the end of the project,

we had created a collaborative environment that

offered not only a universal and ongoing sugges-

tion posting mechanism thus fulﬁlling the initial

objective , but also, and more importantly, new

possibilities and functions which were not avail-

able in the previous process, prior to the introduc-

tion of technology.

Figure 1: Leaders Guide: Chapter Elements (element type).

SCENARIO-BASED DESIGN - AN ESSENTIAL INSTRUMENT FOR AN INNOVATIVE TARGET APPLICATION -

Case Report

115

3 PROPOSED ENVIRONMENT: A

SKETCH

Taking both the abovementioned steps and the discus-

sions with potential users, which together helped de-

termine some of the innovative requirements associ-

ated to the main scenario (i.e. suggestion entry for

the Leaders Guide updates), we sketched the envi-

ronment. Our solution basically consists of the divi-

sion of suggestion entry into two phases, as follows:

ﬁrstly, users have to set the target object of the sugges-

tion and, then, enter the suggestion itself. In the fol-

lowing sections we will describe all information and

procedures that concern the process as a whole.

3.1 Setting the Target Object of the

Suggestions

During this phase of target setting, we took into con-

sideration the different object types derived from the

actual suggestions in the documents available. We

then came up with three possibilities, as follows: the

“Table of Contents” (displayed as hypertext, enabling

access to its speciﬁc components on request), the

“Topic” itself and “Others” (comprising those sugges-

tions which do not ﬁt into the other two categories).

This last type would include socio-political sugges-

tions concerning the Leaders Guide as a whole, as

well as suggestions about the work methodology it-

self. An instance of a suggestion that would be in-

serted into this last category is the following: “My

impression is that the Leaders Guide instructs leaders

cognitively, but fails to teach about the exact method-

ology through which all this knowledge can be ap-

plied to the target families”. The several types of in-

formation that make up this screen are shown in Fig-

ure 2.

The interaction dynamics of the process of choos-

ing one of the Guides components basically consists

of selecting the section of interest by clicking the cor-

respondent link in the Table of Contents. After that,

the screen of the beginning of the session will be dis-

played, and users will be able to choose from the ele-

ments available. Once the element has been selected,

the system will then display a list of suggestions un-

der discussion, so as to avoid repetitions. Only after

this discussion page has been thoroughly examined

can users access the actual suggestion entry page.

If instead users choose to select a topic as the ob-

ject of their suggestion, they will then be asked to se-

lect from the thematic tree in which all topics men-

tioned in the Leaders Guide are organised. It is essen-

tial to bear in mind that a single topic may be men-

tioned in more than one chapter; similarly, a single

chapter may mention more than one topic. This is

why we decided to enable the Table of Contents to-

gether with the thematic menu, as an alternative. An-

other aspect to be taken into consideration is the in-

terdisciplinary nature of some of the topics. In this

case, the system will automatically choose the area to

which the topic is most closely related. This corre-

lation between topics and specialisation areas will be

established by the head of each area within the Na-

tional Coordination, aiming at correctly associating

suggestions and areas to facilitate dialogue modera-

tion once the suggestion has been posted.

Finally, based on the analysis of the suggestions

collected through the human procedure, we feel that

the option “Others” is paramount. Indeed, a wide

range of suggestions referred to the material as a

whole, to its political focus, to the work methodology

within the Institution, etc.

Figure 2: Menu for setting the target object of the sugges-

tion to be posted: information structure.

3.2 The Process of Suggestion Posting

Itself

The process of suggestion posting about one of the

Guides components is enabled through a screen with

data entry, as shown in Figure 3.

However, when the suggestion concerns a “Topics”

or belongs to the “Others” category (the latter in-

ICEIS 2009 - International Conference on Enterprise Information Systems

116

Figure 3: Suggestion entry screen: Information structure

(commentaries for the readers).

cluding more reﬂections than actual suggestions) the

system will display an open entry screen. It is im-

portant to mention that most contributions belonging

to these categories are entered by both governmental

and non-governmental socio-technical organisations,

which means that for this case the readability is often

high. Thanks to this factor, we believe that a mere re-

production of the human procedure (i.e. open entry)

is relevant and justiﬁable.

4 CONCLUSIONS AND FUTURE

WORK

In the present article, we described a case in which

the use of scenarios supported by the application led

to the project of a user-interface environment with a

differential interaction when compared to the human

procedure, prior to the introduction of technology.

Designed, at ﬁrst, to solve problems of geographic

extension and difﬁculties in enabling ongoing contri-

butions, the environment we proposed ended up offer-

ing possibilities before inexistent, thereby character-

ising an innovative technological artefact.

In summary, through a methodology developed to

make full use of the semantic potential of the docu-

ments on suggestion entry (prior to the introduction of

technology), we created an environment capable of,

amongst others things, minimising ambiguity (or at

least signiﬁcantly minimising the likelihood of its oc-

currence) in the most commonly posted suggestions

(i.e. the ones belonging to one of the Guides compo-

nents); lending visibility to each suggestion posted;

encouraging “democratic” debates (with ongoing ac-

cess to all different users proﬁles, allowing users to

comment on all suggestions); ensuring that a techni-

cal reply will be provided, including a justiﬁcation to

the author of the suggestion (visible to the entire com-

munity), thus showing respect to each individual con-

tributor; registering suggestions, decisions and justiﬁ-

cations (also enabling future access); generating, by

the end of the systems availability period, a report

with all suggestions posted, along with its respective

decisions and justiﬁcations, so as to assist in the pro-

cess of compilation of the new issue of the Leaders

Guide.

The example described here makes it clear that

scenario-based design is a feasible method for the

analysis and development of real systems, constitut-

ing a differential factor in projects of innovative in-

formation systems.

As future work we can quote the systema imple-

mentation followed by an experiment with real users.

REFERENCES

Achour, B. (1999). Extracting Requirements by Analysing

Text Scenarios. PhD thesis, Universit Paris - 6.

B”dker, S. (1999). Scenarios in user-centered design - set-

ting the stage for reﬂection and action. In HICSS ’99:

Proceedings of the Thirty-Second Annual Hawaii In-

ternational Conference on System Sciences-Volume 3,

page 3053, Washington, DC, USA. IEEE Computer

Society.

Bezerra, E. (2007). Princpios de Anlise e Projetos de Sis-

temas com UML. Ensevier Editora Ltda.

Bittner, K. and Spence, I. (2002). Use Case Modeling.

Addison-Wesley Longman Publishing Co., Inc.

Button, B. and Sharrock, W. (1994). Occasioned practices

in the work of software engineers. pages 217–240.

Carroll, J. M. (2003). HCI Models, Theories, and Frame-

works: Toward a Multidisciplinary Science. Morgan

Kaufmann.

Cockburn, A. (1997). Structuring use cases with goals.

Journal of Object-Oriented Programming, Sept-Oct

and Nov-Dec.

Dourish, P. (2001). Where the action is: the foundations

of embodied interaction. MIT Press, Cambridge, MA,

USA.

Preece, J., Rogers, Y., Sharp, H., Benyon, D., Holland, S.,

and Carey, T. (1994). Human-Computer Interaction:

Concepts And Design (ICS). Addison Wesley.

Sommerville, I. (2000). Software Engineering (6th Edition).

Addison Wesley.

Suchman, L. A. (1987). Plans and situated actions: the

problem of human-machine communication. Cam-

bridge University Press, New York, NY, USA.

SCENARIO-BASED DESIGN - AN ESSENTIAL INSTRUMENT FOR AN INNOVATIVE TARGET APPLICATION -

Case Report

117