Architecture Decision Records (ADRs) are lightweight tools for documenting critical design decisions, designed to enhance collaboration and maintain architectural alignment in software projects. While ADRs have gained traction in open-source communities, their adoption remains inconsistent, and their impact on both repositories and developers is poorly understood. Existing studies have explored ADR usage at a surface level but have not addressed key questions regarding their influence on project metrics or the people behind their creation and maintenance. In this paper, we present an in-depth analysis of ADR practices in open-source software. Using a dataset of 921 GitHub repositories from earlier ADR studies, we investigate two dimensions: (1) repository-centric patterns, focusing on ADR attributes such as length, frequency of updates, and their predominantly small correlations with project quality metrics such as code smells and issue resolution time; and (2) developer-centric factors, examining who contributes to ADRs, contrasting core and peripheral developers, and how their contributions relate to repository activity, productivity metrics, and ADR maintenance. The results obtained present a nuanced picture of ADR adoption. On one hand, it is clear that the community is exploring its potential as a tool for documenting critical architectural decisions and the processes behind them. On the other hand, ADR adoption practices in open-source projects remain inconsistent; for example, of a total of 5,800 ADRs, approximately 3,674 (about 63%) were opened directly with a status of “accepted” or equivalent, thereby bypassing the deliberative context that ADRs are intended to capture. These findings highlight both the promise and the challenges of integrating ADRs into open-source software development.