The development of open-source software (OSS) projects usually have been driven through collaborations among contributors and strongly relies on volunteering. Thus, allocating software practitioners (e.g., contributors) to a particular task is non-trivial and draws attention away from the development. Therefore, a number of bug bounty platforms have emerged to address this problem through bounty rewards. Especially, Gitcoin, a new bounty platform, introduces a bounty reward mechanism that allows individual issue owners (backers) to define a reward value using cryptocurrencies rather than using crowdfunding mechanisms. Although a number of studies have investigated the phenomenon on bounty platforms, those rely on different bounty reward systems. Our study thus investigates the association between the Gitcoin bounties and their outcomes (i.e., success and non-success). We empirically study over 4,000 issues with Gitcoin bounties using statistical analysis and machine learning techniques. We also conducted a comparative study with the Bountysource platform to gain insights into the usage of both platforms. Our study highlights the importance of factors such as the length of the project, issue description, type of bounty issue, and the bounty value, which are found to be highly correlated with the outcome of bounty issues. These findings can provide useful guidance to practitioners.

翻译：开源软件（OSS）项目的开发通常通过贡献者之间的协作驱动，并严重依赖自愿参与。因此，将软件从业者（如贡献者）分配到特定任务并非易事，且会分散开发精力。为此，许多漏洞赏金平台通过赏金奖励机制来解决这一问题。特别是，新型赏金平台Gitcoin引入了一种赏金奖励机制，允许单个问题发起人（支持者）使用加密货币而非众筹机制来定义奖励价值。尽管已有诸多研究探讨了赏金平台的现象，但这些研究基于不同的赏金奖励系统。因此，本研究旨在探究Gitcoin赏金与其结果（即成功与非成功）之间的关联。我们通过统计分析和机器学习技术，对超过4,000个带有Gitcoin赏金的问题进行了实证研究。同时，我们与Bountysource平台进行了对比研究，以深入了解这两个平台的使用情况。研究发现，项目长度、问题描述、赏金问题类型以及赏金价值等因素与赏金问题的结果高度相关。这些发现可为从业者提供有益指导。