Documentação de tarefas em Software Crowdsourcing : um estudo empírico sobre a plataforma TopCoder

• Ano de defesa: 2018
• Autor(a) principal: Vaz, Luis Fernandes
• Orientador(a): Marczak, Sabrina dos Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Pontifícia Universidade Católica do Rio Grande do Sul
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação
• Departamento: Escola Politécnica
• País: Brasil
• Palavras-chave em Português: Documentação de Tarefas; Engenharia de Requisitos; Estudo Empírico
• Palavras-chave em Inglês: Software Crowdsourcing; TopCoder; Task Documentation; Requirements Engineering; Empirical Study
• Área do conhecimento CNPq: CIENCIA DA COMPUTACAO::TEORIA DA COMPUTACAO
• Link de acesso: http://tede2.pucrs.br/tede2/handle/tede/8313

This research aimed to investigate task documentation in Software Crowdsourcing, more specifically, in the TopCoder platform. It also aimed to identify the elements that should be considered in the documentation of a task in this kind of software development. This research is of importance when considering that a Task is the component that links the other components of the software crowdsourcing model, which are: the Buyer, the Platform, and the Crowd. It is the task that expresses the Buyer’s need to the crowd members. We followed a qualitative research approach and conducted a Case Study with newcomers in Software Crowdsourcing and a Field Study with industry professionals. Data was analyzed using the Content Analysis technique. We found that, for the Case Study novices, the documentation of the task had a secondary role in the task selection. However, the need of a clear documentation become more relevant during the development of the task given that this is the moment that the instructions within the documentation need to be decoded by the developer and turned into a solution to be later submitted to the platform. For the Field Study participants, the most relevant elements related to the documentation of a task were how clear the description of a task is and their prior knowledge about the task content in order to influence its selection. Inspired on our studies’ results, we propose a model for task documentation in TopCoder. We believe this model will likely aid the description of tasks in software crowdsourcing and will, as a consequence, help crowd members in their task development journey.