SELECTION CRITERIA FOR SOFTWARE DEVELOPMENT

TOOLS FOR SMES

SMEs and Cooperatives in Venezuela

Lornel Rivas, María Pérez, Luis E. Mendoza and Anna Grimán

Processes and Systems Department, Simón Bolívar University

PO Box 89000, Caracas 1080-A, Venezuela

Keywords: Software development, methodology, SME, tools, selection.

Abstract: Software engineering tools have regained interests in recent years due to different changes affecting soft-

ware developing organizations. These organizations carry out activities that might be undertaken in a plan

driven and agile manner with the support of such tools. A proper balance between both approaches and the

effective tool adoption will help organizations to meet their objectives and evolve. Small and medium en-

terprises and Cooperatives (S&C) share common characteristics throughout Latin America. Small and me-

dium enterprises (SMEs) lack of formality in their roles and relationships among interacting individuals,

whereas Cooperatives are usually small companies with weaknesses as to management techniques and tech-

nological equipment. In fact, both have difficulties when finding the right personnel and tools that best suit

their needs. Considering Venezuela as our study subject, we have herein proposed some criteria to assist

S&C in the tools selection that support their development processes while fostering the balance required be-

tween agility and discipline. Such criteria were formulated based on the characterization of five factors

aimed at determining this balance. These contributions will help subsequently identifying methodological

and technical aspects to provide guidance to S&C in the improvement of their development processes.

1 INTRODUCTION

Small and medium enterprises and Cooperatives

(S&C) are increasingly consolidating their presence

in the countries’ economies. In Venezuela, there are

83.68% of local software developing organizations

(Rivero et al., 2007). However, S&C have character-

istics that might affect their competitiveness. Such

characteristics influence the organizations’ software

developing activities and determine their particular

needs, as to the support required from software en-

gineering tools. Boehm & Turner (2004) proposed

five factors to determine the balance between agility

and discipline for software development projects.

Our main goal is, from the analysis of such factors

(size, criticality, dynamism, personnel, and culture)

within the S&C particular context, to propose a set

of criteria for tool selection at Venezuelan S&C.

2 RELATED WORK

An adequate agility-discipline balance may signifi-

cantly support compliance with the objectives of a

software development organization (Boehm &

Turner, 2004). Parada et al. (2008) proposed a deci-

sion network aimed at determining the organiza-

tions’ necessity of applying either an agile or a plan-

driven methodology. If it is located in the center of

the network, an agile methodology is recommended

for the organization, but if pointing towards the

outer edge, a plan-driven methodology seems to be a

suitable decision (see Figure 1).From this network,

we infer that S&C require agility, but when they

undertake critical, complex or medium-sized pro-

jects, they might require a higher discipline level in

order to improve their probability of success. Like-

wise, if they want become more competitive, they

should apply more disciplined methodologies. The

perfect balance lies in obtaining a degree of disci-

pline that does not make them lose their essence of

S&C, with little-specialized highly-proactive re-

sources. Based on this categorization, we identified

functional and non-functional characteristics for

those software engineering tools, which focus on the

development processes undertaken at S&C to gather

a set of criteria that facilitate their selection.

252

Rivas L., Pérez M., E. Mendoza L. and Grimán A. (2008).

SELECTION CRITERIA FOR SOFTWARE DEVELOPMENT TOOLS FOR SMES - SMEs and Cooperatives in Venezuela.

In Proceedings of the Tenth International Conference on Enterprise Information Systems - ISAS, pages 252-255

DOI: 10.5220/0001696402520255

 SciTePress

Figure 1: Decision network of Venezuelan software developing SMEs and Cooperatives (Parada et al., 2008).

3 S&C CHARACTERISTICS

In SMEs, complete functions are concentrated on

few individuals. Flexibility in the roles performed by

the personnel is a common pattern to most SMEs.

This pattern is evidenced in the lack of formality of

relations among individuals and their roles. This is

also observed in the organizations’ flat structure as

well as in the flexible workflow driven by the needs

of a specific moment (Erard, 2005). SMEs’ main

difficulty is finding the right personnel and neces-

sary tools to perform their work.

Cooperatives show fluid communication and

work. A group of people integrate an ad-hoc team

any time a need arises. The composition of that

team will not probably be influenced by the person-

nel’s degree or work position within the organiza-

tion, but it will be driven by the current work needs

and skills of the available personnel (Molina & Gar-

cía, 2005). Cooperatives usually are small organiza-

tions. They are mostly managed by volunteers, who

have little managerial skills and remarkable techno-

logical deficiencies in their productive and service

processes (Muller, 2001).

S&C have characteristics that allow for ap-

proaching the degree of agility and discipline they

should have to succeed; all this in order to adopt or

adapt their methodologies for achieving this pur-

pose. The initial effort for implementing processes

aimed at improving product and service quality at

Latin American (LA) software systems development

(SSD) organizations is high and complex. This is

mainly due to the fact that most organizations are

small- or medium-sized, lack any monthly budget

and plans, and have no control on the productivity

and processes involved in SSD (Herrera, 2003).

Only a small fraction of LA organizations has been

awarded official CMM or ISO 9000 certifications

from international entities (Herrera, 2003).

These organizations account for a significant

percentage of SSD companies in LA. In Venezuela,

the Software National Industry (SNI) is mostly com-

prised by SMEs with less than 51 employees (ap-

proximately 83.68%), the main layer being organiza-

tions with 21 and 51 employees (42.68%) (Rivero et

al., 2007). Most companies of SNI do not use mature

models and methods for their productive processes.

Now, only a little percentage of the surveyed organi-

zations could be qualified at CMMI level 2, which is

a relevant and high-priority factor for reaching inter-

national competitiveness (Rivero et al., 2007).

For LA S&C, and particularly in Venezuela, the

question is: In accordance to the factors proposed by

Boehm, which parameters determine the level of

agility and discipline of a methodology to make

these organizations competitive in the global mar-

ket? Consequently, which implications should soft-

ware developing S&C consider when deciding on

the tools required to support their development proc-

esses? Next, we present the Boehm’s factors analy-

sis according to S&C context in Venezuela.

Size. This factor’s determines the application of

a methodology with a high agility level that allows

SELECTION CRITERIA FOR SOFTWARE DEVELOPMENT TOOLS FOR SMES - SMEs and Cooperatives in

Venezuela

253

for distributing team’s work.

Culture. This factor determines the application

of a methodology that leads to flexibility in decision-

making processes and problem resolution within a

formal framework, so as to achieve agility without

risking team work planning.

Personnel. This factor determines the applica-

tion of an agile methodology that allows experts to

solve immediate problems with a sufficient level of

discipline aimed at work management for the rest of

the group.

Criticality. This factor determines the applica-

tion of a plan-driven methodology.

Dynamism. This factor determines the applica-

tion of a plan-driven methodology.

4 TOOLS SELECTION CRITERIA

Although software engineering supporting tools

were developed to support the stakeholders’ tasks

and processes throughout the information system life

cycle (Lundell & Lings, 2004), they share multiple

scopes and are subjected to constant changes derived

from the advances in the software engineering

knowledge field. For S&C, the decision of acquiring

a tool may have significant implications in terms of

costs and implementation. Hence, the need for mak-

ing well-grounded decisions regarding technology-

adoption processes.

For this reason, we intend proposing a set of cri-

teria for tools selection in S&C. These criteria estab-

lish a framework for decision making processes and

serve as referential parameters for the evaluation and

selection of tools, so to offer advantages for achiev-

ing the necessary agility-discipline balance through-

out the software development processes at S&C.

Such criteria have been established based on the

S&C characteristics analyzed herein and the factors

affecting the required balance (see Figure 2).

Flexible Project Management. S&C have time,

people and financial restrictions, i.e., S&C need an

efficient management of their resources. This poses

some expectations regarding support to flexible pro-

ject management, which are related to the tool func-

tional characteristics, are: (a) Organizational man-

agement: objectives, time, tasks -per individuals and

groups- goals, events, unplanned actions, and correc-

tive actions; (b) Metrics management and generation

of reports on metrics related to the project follow-

up: percentage of completion, planned/executed

progress, percentage of participation.

Flexible Process Management. High competi-

tive environment, and productivity pressures, de-

mands to S&C to be prepared for handle changes

that arise during the project development. Then,

considerations about personnel skills and distribu-

tion of software development activities among ex-

perts and senior developers are very important. Bal-

ance of agility and discipline is required to solve

problems and manage work, respectively. Criteria

proposed include (a) methodology customization

support; key device generation in accordance with

time, objectives and available personnel; selection or

omission of activities according to the context; (b)

strengths in analysis activities; UML support, use

case-oriented analysis; iterative development sup-

port; (c) prototypes’ design; (d) role management

support: role selection, fluent communication among

roles

Coordination, Communication, and Coopera-

tion. The process requires mechanisms to facilitate

for efficient interaction among the team members,

since development teams are usually diverse, shared

and distributed. Therefore, it is necessary to evaluate

tools with features such as: (a) portability, thus fa-

cilitating users’ handling in different time-zones and

locations; (b) allow for efficient information ex-

change: notification mechanisms; support at reposi-

tories and virtual or distributed environments.

Scalable Repository Management. The tool’s

ability to facilitate coordination, communication and

cooperation, demands integrated data management

and safeguarding. It is relevant to count on tools

capable of (a) interoperations, by being compatible

with other data storage tools, protocols, standards,

shared databases; (b) providing access security and

appropriate role and team restrictions; (c) offering

accurate results by being consistent in the integrated

exchange of data recorded in the repository.

Support to the Methods and Approaches. In

order to support planning required by S&C, we ex-

pect the satisfaction of the following criteria: (a)

support to the methods, models, and notation of

analysis and designs used by the organization, (b)

adherence to standards (c) availability of documen-

tation on the methods used, (d) support to the devel-

opment approach used by S&C.

Quality Assurance. Normally S&C accept chal-

lenges of increasingly critical projects; therefore, we

have conceived this criterion as support to activities

mainly related to software internal features, such as

availability of resources for (a) Audit and inspection

checklists; (b) Code standard analyzability; (c) Test-

ability: black-and- white box tests, (d) model verifi-

cation (logic checks, error messages, ease of debug-

ging, trace files step, dynamic display of elements

and display of the workflow path).

ICEIS 2008 - International Conference on Enterprise Information Systems

254

Figure 2: Proposed criteria for tool selection in S&C.

User-friendly and Self-learnable. S&C have

difficulties for acquiring specialized people. People

are few. Then, S&C could need tools that provide

ease of learning, so that people can getting benefits

in short time. Hence, the relevance of considering

criteria related to tool usability: (a) user-friendly

management of appealing interfaces, reducing

change resistance; (b) model representativity and

functionality; (c) tutorials and on-line documentation

to encourage effectiveness and efficiency of the

learning process (d) availability of libraries support-

ing modeling efficiency.

5 CONCLUSIONS AND FUTURE

WORK

In this work, we have established a set of criteria for

the selection of software engineering tools that best

suit S&C’ reality. Criteria pursuing

a higher level of

agility are essential for proper management of soft-

ware development activities within time, resources,

and personnel restrictions, which are common to

these organizations. Criteria pursuing

a higher level

of discipline are key to ensure a higher control over

the team, tasks and objectives, thus reducing the

possibility of chaos and supporting efforts’ integra-

tion.

Next step in this research is identifying which

elements should be

included in the agile methodolo-

gies to achieve these goals, i.e. a proper disciplined

balance and refined tool selection criteria susceptible

of being evaluated and instantiated.

ACKNOWLEDGEMENTS

This research was supported by National Fund of

Science, Technology and Innovation, Venezuela,

under contract S1-2005000165.

REFERENCES

Boehm, B. & Turner, R., 2004. Balancing Agility and

Discipline: A Guide for the Perplexed. Addison-

Wesley Professional.

Erard, P. ¿Como vencer los obstáculos al desarrollo de la

pyme? El aporte de Fundes. Debates Iesa, 5, 9-19, Ca-

racas, Venezuela.

Herrera, E. 2003. A methodology for self - diagnosis for

software quality assurance in small medium-sized in-

dustries in Latin America, In Americas Conference on

Information Systems.

Lundell, B. and Lings, B. (2004), Changing perceptions of

case technology, J SYST SOFT, 7(2), 271-280.

Molina, C. & García, A., 2005. Cooperativa ¿qué es y

cómo funciona?, PANAPO: Caracas, Venezuela.

Muller, A., 2001. Acerca de la Reforma de la Ley de Co-

operativas de Venezuela. Revista Venezolana de Eco-

nomía Social, 1, 1-19. Mérida, Venezuela.

Parada, D, Di Paula, G., Mendoza, L., Pérez, M., 2008.

Disciplina y agilidad en el proceso de desarrollo de

software para SMEs y cooperativas en Latinoamérica.

In VII Jornadas iberoamericanas de ing. de software e

ing. de conocimiento. Guayaquil, Ecuador.

Rivero, D., Montilva, J., Granados, J., Barrios, J. Besem-

bel, I., Sandia, B., 2007, La Industria de Software en

Venezuela: Una Caracterización de su Recurso Huma-

no. In Encuentro Venezolano sobre Tec. de la Inf. e

Ing. del Software, Margarita, Venezuela.

SELECTION CRITERIA FOR SOFTWARE DEVELOPMENT TOOLS FOR SMES - SMEs and Cooperatives in

Venezuela

255