Clouds are shareable scientific instruments that create the potential for reproducibility by ensuring that all investigators have access to a common execution platform on which computational experiments can be repeated and compared. By virtue of the interface they present, they also lead to the creation of digital artifacts compatible with the cloud, such as images or orchestration templates, that go a long way—and sometimes all the way—to representing an experiment in a digital, repeatable form. In this article, I describe how we developed these natural advantages of clouds in the Chameleon testbed and argue that we should leverage them to create a digital research marketplace that would make repeating experiments as natural and viable part of research as sharing ideas via reading papers is today.

中文翻译：

---

一线希望

云是可共享的科学工具，通过确保所有研究人员都可以访问可以重复和比较计算实验的通用执行平台，从而创造可重复性的潜力。凭借它们呈现的界面，它们还导致创建与云兼容的数字工件，例如图像或编排模板，这对于以数字化、可重复的方式表示实验大有帮助——有时甚至是一路形式。在这篇文章中，我描述了我们如何在 Chameleon 测试平台中开发云的这些自然优势，并认为我们应该利用它们来创建一个数字研究市场，使重复实验成为研究的自然和可行的一部分，就像通过阅读论文分享想法一样今天。