A caracterização das microtasks aplicadas ao desenvolvimento de software crowdsourcing

Abstract

Background: software development has been transformed with the progress of technologies and the particularities that each project can present. Currently, there is a tendency of the productive sector for the adoption of crowdsourcing (CS) in the creation of products and software projects being denominated as CS software development. This trend applies several concepts of distributed and open source development. Despite this, its activities known as “microtasks”, still represent the cause of much contradiction and confusion within the literature. Objective: the aim of this work was to investigate the characterization of microtasks based on four pillars: the use, characteristics, contrasts and complexity of microtasks. Method: the methods and procedures adopted in conducting this study represented a hybrid validation approach. Initially, a case study was conducted to investigate the use of microtasks in CS software development. Based on the experience provided by the analyzed case, the literature was consulted in order to add inputs on the characteristics of a microtask and to compare their contrasts with the software activities. Finally, controlled experimentation was used to model and validate a microtask complexity assessment approach. Results: the results demonstrated the breadth of use of microtasks in different scenarios and steps of the life cycle of a CS project; a comprehensive taxonomy on the characteristics and states that a microtask may possess; the freedom of development that microtasks provide in relation to software activities performed in distributed development; and an approach capable of measuring the execution effort of a microtask. Conclusion: with the conduction of this research, it was possible to identify that microtasks represent a trend of CS software development, providing empirical contributions on its application and execution as well as theoretical contributions on its structure and behavior.