Free and open source software package ecosystems have existed for a long time, but such collaborative development practice has surged in recent years. One of the oldest and most popular package ecosystems is Comprehensive R Archive Network (CRAN), the repository of packages of the statistical language R, a popular statistical computing environment. CRAN stores a large number of packages that are updated regularly and depend on many other packages in a complex graph of relations. As the repository grows, its sustainability could be threatened by that complexity or nonuniform evolution of some packages. This paper provides an empirical analysis of the evolution of the CRAN repository in the last 20 years, considering the laws of software evolution and the effect of CRAN's policies on such development. Results show how the progress of CRAN is consistent with the laws of continuous growth and change and how there seems to be a relevant increase in complexity in recent years. Significant challenges are raising related to the scale and scope of software package managers and the services they provide; understanding how they change over time and what might endanger their sustainability are key factors for their future improvement, maintenance, policies, and, eventually, sustainability of the ecosystem.

中文翻译：

---

综合 R 档案网络（CRAN）包生态系统的演进与展望

免费和开源的软件包生态系统由来已久，但近年来这种协同开发的实践激增。最古老和最受欢迎的软件包生态系统之一是综合 R 档案网络 (CRAN)，它是统计语言 R 的软件包存储库，是一种流行的统计计算环境。CRAN 存储了大量定期更新的包，这些包依赖于复杂关系图中的许多其他包。随着存储库的增长，其可持续性可能会受到某些包的复杂性或不均匀演变的威胁。本文提供了过去 20 年 CRAN 存储库演变的实证分析，考虑了软件演变的规律以及 CRAN 政策对这种发展的影响。结果表明 CRAN 的进展如何与持续增长和变化的规律相一致，以及近年来复杂性似乎有相关的增加。与软件包管理器及其提供的服务的规模和范围相关的重大挑战正在增加；了解它们如何随时间变化以及什么可能危及它们的可持续性是它们未来改进、维护、政策以及最终生态系统可持续性的关键因素。