arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.03201cs.SE
收藏

TrustOps:持续构建可信软件

TrustOps: Continuously Building Trustworthy Software

Eduardo Brito, Fernando Castillo, Pille Pullonen-Raudvere, Sebastian Werner

AI总结:

针对软件服务信任验证难、现有模型依赖假设存风险的问题,提出TrustOps方法,通过在软件全生命周期收集可验证证据并结合现有工具技术,为构建新信任模型提供适应性原则与未来研发布局。

摘要:

软件服务在日常生活中扮演着至关重要的角色,自动化操作决定着资源和信息的访问权限。信任服务提供商能够公平、准确地执行这些操作至关重要,但用户难以验证这一点。即使有公开可用的代码库,快速的开发节奏和现代部署的复杂性也阻碍了对服务操作的理解和评估,专家也不例外。因此,当前的信任模型严重依赖于服务提供商遵循最佳实践并遵守法律法规的假设,这越来越不切实际且具有风险,导致未被发现的缺陷和数据泄露。在本文中,我们认为在软件开发和运营过程中收集可验证的证据是创建新信任模型的必要条件。为此,我们提出TrustOps,这是一种在软件生命周期所有阶段持续收集可验证证据的方法,它依赖并结合已有的工具和增强信任的技术来实现这一目标。为此,我们介绍了TrustOps的适应性核心原则,并提供了未来研究和开发的路线图。

英文摘要:

Software services play a crucial role in daily life, with automated actions determining access to resources and information. Trusting service providers to perform these actions fairly and accurately is essential, yet challenging for users to verify. Even with publicly available codebases, the rapid pace of development and the complexity of modern deployments hinder the understanding and evaluation of service actions, including for experts. Hence, current trust models rely heavily on the assumption that service providers follow best practices and adhere to laws and regulations, which is increasingly impractical and risky, leading to undetected flaws and data leaks. In this paper, we argue that gathering verifiable evidence during software development and operations is needed for creating a new trust model. Therefore, we present TrustOps, an approach for continuously collecting verifiable evidence in all phases of the software life cycle, relying on and combining already existing tools and trust-enhancing technologies to do so. For this, we introduce the adaptable core principles of TrustOps and provide a roadmap for future research and development.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑