Cleidson R. B. de Souza

The scale and evolution of coordination needs in large-scale distributed projects: implications for the future generation of collaborative tools

The past decade has witnessed the development of a new class of coordination tools that focus on automatically providing individuals a rich context for facilitating the coordination of their work. Despite their valuable contributions, current coordination tools have mostly been designed without taking into account scalability aspects beyond the small-group level. The increasing pervasiveness of large-scale projects suggests that those mechanisms need to scale dramatically to adequately support such work settings. In this paper, we used data from five distinct large-scale projects from three different companies to study the scale, range, and volatility of the coordination requirements that emerged over time within those projects. Our results showed that coordination requirements tend to be quite volatile, vary significantly in their magnitude across project members and a significant proportion of the coordination requirements cut across organizational and geographical boundaries. Furthermore, new coordination requirements represent, on average, a third of the coordination requirements faced by a project member on a monthly basis. The implications of these results for the design of collaborative tools are discussed.

Towards a classification of logical dependencies origins: a case study

Logical dependencies are implicit relationships established between software artifacts that have evolved together. Software engineering researchers have investigated this kind of dependency to assess fault-proneness, detect design issues, infer code decay, and predict likely changes in code. Despite the acknowledged relation between logical dependencies and software quality, the nature of the logical dependencies is unknown in the literature. Most authors hypothesize about their origins, but no empirical study has been conducted to investigate the real nature of these dependencies. In this paper, we investigated the origins of logical dependencies by means of a case study involving a Java FLOSS project. We mined the project repository, filtered out irrelevant data based on statistical analyses, and performed a manual inspection of the logical dependencies to identify their origins using information from the revision comments, code diffs, and informal interviews held with the developers of the analyzed project. Preliminary results showed that logical dependencies involved files that changed together for a series of different reasons, which ranged from changing software license to refactoring classes that belonged to the same semantic class.

Information needs of system administrators in information technology service factories

In this paper we describe the results of an empirical study about the information needs of system administrators. This study is based on an electronic survey with more than 200 systems administrators, or sysadmins, working on incident management in a large scale IT service delivery organization. The survey covered their information needs in both complex and routine situations. The results of the survey described in this paper go beyond previous work on system administrators by presenting a much more complex picture, suggesting that sysadmins make low usage of knowledge management tools; largely adopt personal communication and collaboration tools: and finally, need to gather information about customers from a complex set of stakeholders. The system administrators also indicated in our survey that the most useful sources of information in handling complex incidents are: (i) the customer; (ii) the customer account team; and (iii) other employees who were experts both in the customer and in particular aspects of the delivery of services. This study indicates that knowledge management in IT service factories is very challenging and possibly should evolve from the often adopted passive model to a dynamic knowledge management style emphasizing both knowledge reusability through information technologies and knowledge sharing through informal discussions among employees.

RaisAware: Uma Ferramenta de Auxílio à Engenharia de Software Colaborativa Baseada em Análises de Dependências

This paper presents RaisAware, a collaborative software development tool. RaisAware is one of the few tools aimed to support the long known relationship between software architecture and coordination of work. Our design is based on dependency analysis of software development artifacts and software developerspsila activities. We describe the theoretical and analytical motivations behind this work, detail the design and implementation of RaisAware, and provide recommendations for future work in this direction.

Cooperative and human aspects of software engineering (CHASE 2009)

The CHASE 2009 workshop is concerned with exploring the cooperative and human aspects of software engineering, and providing a forum for discussing high-quality research. Accepted papers reflect the diversity of the field of software engineering - ranging from requirements to testing, and from ethnographic research to experiments. Moreover, the background of attendees reflects the diversity of researchers in this domain, ranging from sociology to psychology, from informatics to software engineering. CHASE 2009 met its goals in presenting high-quality research and building community through a mixture of presentations, discussions, posters, and social activities.

Designing an enterprise social tool for cross-boundary communication, coordination, and information sharing

This paper discusses the design of a social tool for cross-boundary communication, coordination, and information sharing in a large organization. Based on insights and requirements gathered in qualitative and quantitative studies conducted within the organization, the Live Corkboard, a virtual message board system enhanced with community features and text/history search is proposed as a tool to enhance communication, group awareness, and information sharing and reuse. We describe the requirements for our tool as well as how they influenced our design. The research was conducted in a large IT services delivery company which has recently changed its organizational structure from a customer-centered to a competency-centered model. Focus group evaluation results suggest that the tool will be useful to the employees in the organization.

The Evolution and Impact of the Research in Distributed Software Development in Brazil

A comunidade mundial de Engenharia de Software tem testemunhado uma mudança significativa na forma com que os projetos de software têm sido desenvolvidos nas últimas duas décadas: as equipes vêm sendo organizadas com seus integrantes geograficamente distantes uns dos outros. O termo Desenvolvimento Distribuído de Software (DDS) tem sido cada vez mais utilizado na academia e na indústria para especificar este contexto de desenvolvimento, bem como suas peculiaridades e problemas. Ao mesmo tempo, esta mudança vem causando impacto não apenas no mercado em si, mas na maneira como os produtos de software estão sendo modelados, construídos, testados e entregues para os clientes. Neste sentido, o DDS tem atraído um grande número de pesquisas na área de Engenharia de Software. No Brasil isto não é diferente. Existem registros de pesquisas sendo desenvolvidas no país desde 1999. O objetivo deste artigo é apresentar e discutir como a pesquisa em DDS tem evoluído no país a partir de uma avaliação histórica de artigos sobre este assunto, qual o impacto desta pesquisa tanto em nível nacional como internacional e o papel da comunidade brasileira de Engenharia de Software no desenvolvimento desta área. Discute-se também a visão de futuro para pesquisas na área.

Task allocation for crowdsourcing using AI planning

Crowdsourcing is a relatively new phenomenon in computer science and software engineering. In crowdsourcing a task is delivered to a crowd of participants who will work on this task. Task allocation is then an important aspect in the context of crowdsourcing. If done properly, it delivers successful results based on the answers provided by the crowd. However, task allocation in crowdsourcing is not a trivial problem. Factors like a task’s requirements, the knowledge required for its resolution, and the size and heterogeneity of the participants in the crowd all impact task allocation, and therefore, the expected quality of the task results. In this case, the execution of actions from a plan, which assist the dynamic tasks’ allocation in crowdsourcing systems, become relevant as an alternative solution. This paper formalizes task allocation in crowdsourcing scenarios as an artificial intelligence planning problem. Our results suggest that task allocation has several challenges when it is observed in distributed, undefined and dynamic environments, like in crowdsourcing scenarios. Our goal is to evaluate if automated planning is appropriate for providing a plan to match skills of crowd workers for the right tasks in software engineering projects. Preliminary results are presented in this paper,

Collaborative refactoring: results of an empirical study using grounded theory

Due to the current market dynamics, changes in requirements are often faced by the software industry, impacting directly on the software system to be produced. To deal with this situation, software development organizations need to use techniques that enable fast responses. Agile methods have been considered adequate to handle these situations. Our research focuses on understanding how Brazilian organizations are adopting agile methods. In this paper we present the results of an empirical study of refactoring activities, which in the organizations we studied are conducted in a collaborative way. We call this collaborative refactoring. Our results suggest that collaborative refactoring promotes knowledge sharing among the development team especially about the software architecture of the system. On the other hand, we also observed that there is a lack of tools to support collaborative refactoring.

Workshop on cooperative and human aspects of software engineering (CHASE 2011)

Software is created by people for people working in varied environments, under various conditions. Thus understanding cooperative and human aspects of software development is crucial to comprehend how methods and tools are used, and thereby improve the creation and maintenance of software. Over the years, both researchers and practitioners have recognized the need to study and understand these aspects. Despite recognizing this, researchers in cooperative and human aspects have no clear place to meet and are dispersed in different research conferences and areas. The goal of this workshop is to provide a forum for discussing high quality research on human and cooperative aspects of software engineering. We aim at providing both a meeting place for the growing community and the possibility for researchers interested in joining the field to present their work in progress and get an overview over the field.