Socio-technical and Ethical Dimensions of Architecture Practices in FLOSS

I. INTRODUCTION

• FLOSS is critical infrastructure, but emergent architecture leaves rationale and constraints scattered, while authority concentrated among maintainers can bottleneck review, onboarding, and change.Prior work also identifies architectural erosion as a threat to maintainability and long-term evolution.Source: §I · p. 1, §II · p. 1
• The project will study FLOSS architecture practices and develop a conceptual framework and lightweight interventions for practice and education.Its planned work includes multi-method case studies and formative evaluation of interventions.Source: §I · p. 1, §II · p. 2

II. BACKGROUND AND RESEARCH GAP

• Research on FLOSS architecture is fragmented across strands that rarely integrate architectural practice with governance and ethical concerns.The strands include artifacts and erosion, communication, governance, ethical participation, and education.Source: §II · p. 1
• Architectural rationale is distributed across issues, pull requests, and mailing lists, while practitioner discussions may underrepresent early decision-making and rationale.This motivates examining where decisions are negotiated and whether rationale becomes durable project knowledge.Source: §II · p. 1
• FLOSS governance research describes formalization, decentralization, control, and role arrangements, but rarely connects them explicitly to architectural negotiation and stewardship.Role arrangements shape who can approve and coordinate changes.Source: §II · p. 1
• Ethical concerns such as stewardship, responsibility, and inclusivity are invoked in FLOSS, but are seldom examined in day-to-day architecture work or operationalized as observable indicators.Examples include participation pathways, rationale traceability, and newcomer access to architectural knowledge.Source: §II · pp. 1–2
• FLOSS education offers authentic systems and collaboration contexts, while architecture education emphasizes trade-offs; it remains unclear whether teaching addresses architecture as socio-technical work.Relevant concerns include responsibility distribution, participation, and rationale capture.Source: §II · p. 2
• The central gap is how artifacts, communication routines, and governance intersect in FLOSS architecture and relate to stewardship, responsibility, and inclusivity.Evidence is also limited on low-overhead practices that make architectural knowledge more durable and accessible, and on translating them into education.Source: §II · p. 2

III. RESEARCH QUESTIONS AND OBJECTIVES

• The research design is iterative, with early findings informing refinement and refocusing of subsequent questions.Source: §III · p. 2, A. Sub-questions · p. 2

A. Sub-questions

• Selected projects are studied to understand how they document, communicate, and coordinate architecture, and how those practices evolve.Source: A. Sub-questions · p. 2
• The project examines how governance, communication structures, and ethical considerations shape participation and preservation of rationale.Source: A. Sub-questions · p. 2
• It investigates how FLOSS-based courses teach architectural decision-making and related socio-technical responsibilities.Source: A. Sub-questions · p. 2
• It asks which lightweight practices and educational formats can improve explicitness and accessibility without increasing maintainer workload.Source: A. Sub-questions · p. 2
• Education is a core project component: teaching formats stem from empirical findings and are evaluated with the same rigor as practice-oriented interventions.Source: A. Sub-questions · p. 2

B. Objectives

• O1 systematically characterizes architecture artifacts, communication channels, and decision-making mechanisms across a set of FLOSS projects.Source: B. Objectives · p. 2
• O2 analyzes how socio-technical and ethical factors shape architectural practices and knowledge distribution in FLOSS projects.Source: B. Objectives · p. 2
• O3 examines how FLOSS-based teaching presents software architecture and whether it addresses socio-technical and ethical aspects.Source: B. Objectives · p. 2
• O4 develops a conceptual framework linking FLOSS architecture practices, socio-technical conditions, and ethical considerations.Source: B. Objectives · p. 2
• O5 designs and preliminarily evaluates lightweight architectural practices and educational patterns that support more explicit and reflective architecture work in FLOSS.Source: B. Objectives · p. 2

IV. METHODOLOGY

• The three phases are information gathering, solution generation, and evaluation, combining repository mining, communication analysis, and semi-structured interviews to capture artifacts and tacit decision-making knowledge.Educational materials will also be analyzed to assess how FLOSS-based teaching represents architecture practices and whether socio-technical and ethical dimensions are sufficiently addressed.Source: §IV · p. 2
• Each case will describe the project’s architectural knowledge base, including its artifacts and their traceability.Source: §IV · p. 2
• Each case will map decision points and participation pathways, identifying who proposes, discusses, and approves architecture-relevant changes.Source: §IV · p. 3
• Each case will identify socio-technical and ethical factors shaping architectural work.Source: §IV · p. 3

A. Phase 1: Information Gathering and Interpretation

• A systematic literature review will cover FLOSS architecture practice, architectural erosion and smells, architecture knowledge communication, and FLOSS-based education.Source: A. Phase 1: Information Gathering and… · p. 3
• The study will compare 3–4 domain-pairs with comparable architectural cores but differing governance, selected based on repository and communication-channel accessibility.Candidate pairs are LibreOffice/OpenOffice.org, GnuPG/Sequoia-PGP, Pacman/Zypper, Matrix/SimpleX, and Rocq/Z3; selection criteria will be refined after a literature scan and pilot inspection.Source: A. Phase 1: Information Gathering and… · p. 3
• Repository artifacts and dependency graphs will support architectural recovery, while quantitative proxies will assess governance, participation, decision-power distribution, ethical stewardship, and inclusivity.Proxies include architectural authority and permissions, contributor and newcomer participation, ownership and ADR approvals, ethical mentions and participation pathways, role diversity, and discussion sentiment or politeness; feasible automated measures will be validated through interviews and triangulated with qualitative data.Source: A. Phase 1: Information Gathering and… · p. 3
• Issues, pull requests, and mailing-list archives will be mined for architecture threads, alongside 10–20 interviews and a corpus of FLOSS-oriented course materials.Keyword, label, and document-link retrieval will be checked with a stratified manual sample; decision and constraint criteria will be refined iteratively per project.Source: A. Phase 1: Information Gathering and… · p. 3
• Phase 1 will describe current practices, provide initial insights into socio-technical and ethical factors, and establish a baseline for architecture teaching in FLOSS courses.These findings will guide refinement of research questions, sampling strategies, and data collection instruments for subsequent phases.Source: A. Phase 1: Information Gathering and… · p. 3

B. Phase 2: Solution Generation Based on Findings

• The phase will build a framework linking observed architectural practices with socio-technical factors and ethical considerations, alongside lightweight practices for FLOSS projects.Examples include adapted decision-documentation templates, newcomer-inclusive architecture discussions, and ways to make architectural constraints visible.Source: B. Phase 2: Solution Generation Based… · p. 3
• Prototype educational formats and resources will foreground architecture as a socio-technical and ethical practice.Examples include course assignments, reflective exercises, and teaching cases.Source: B. Phase 2: Solution Generation Based… · p. 3
• Co-design sessions or feedback interviews with FLOSS contributors and educators will refine solutions grounded in Phase 1 findings and stakeholder input.The approach is iterative and aims to produce context-sensitive, realistic solutions.Source: B. Phase 2: Solution Generation Based… · pp. 3–4

C. Phase 3: Evaluation of Proposed Solutions

• Pilots in collaborating FLOSS projects and teaching settings will assess feasibility, usefulness, understanding, and participation, with before-and-after cues guiding refinement rather than causal claims.Cues include template adoption and completion, links between ADRs or design discussions and code changes, and participation in architecture-relevant discussions; results will refine the framework and guidelines and support actionable recommendations for communities and educators.Source: C. Phase 3: Evaluation of Proposed… · p. 4

D. Validity, Trustworthiness, and Ethics

• Trustworthiness measures combine triangulation across data sources with a documented, versioned audit trail and sharing scripts and deidentified or aggregated data where feasible,.The study will compare converging and diverging evidence, document sampling and inclusion decisions, and version the evolving codebook and analysis scripts.Source: D. Validity, Trustworthiness, and Ethics · p. 4
• Peer debriefing, participant feedback on interpreted interview themes, and active searches for counterexamples will challenge interpretations and refine explanations.The researcher will discuss coding, themes, and alternatives with peers; share short summaries with interviewees; and examine cases where explicit architectural knowledge coexists with concentrated participation, or vice versa.Source: D. Validity, Trustworthiness, and Ethics · pp. 4–5
• The researcher will reflect on positionality and bias while addressing risks from limited maintainer access, incomplete rationale traceability, case coverage, and evolving generative AI [Table III].Planned mitigation actions are summarized in Table III.Source: D. Validity, Trustworthiness, and Ethics · p. 5, Table III · p. 5
• The study will record whether architectural decisions or ADRs were produced or edited with LLM assistance to assess possible effects on decision authority and knowledge distribution.This lightweight capture is intended to assess systematic impact without making AI a primary focus or expanding the project’s scope.Source: D. Validity, Trustworthiness, and Ethics · p. 5
• Because developers may interpret text-based communication differently, sensitive trace-based interpretations such as conflict will be triangulated through interviews and reported cautiously.Source: D. Validity, Trustworthiness, and Ethics · p. 5
• Interview participation requires informed consent and permits withdrawal; recordings and transcripts will be secured, findings aggregated, and trace data anonymized where appropriate.Direct quotations will be paraphrased or deidentified where needed, and sensitive interpretations will not be attributed to individuals.Source: D. Validity, Trustworthiness, and Ethics · p. 5

V. EXPECTED RESULTS

• The project expects a grounded cross-case account of how architectural practices are enacted and communicated in FLOSS, including artifacts, communication patterns, and recurring challenges.This account is intended to support more explicit and reflective architectural work.Source: §V · p. 5
• Roles, governance, and community norms are expected to explain how knowledge distribution, participation, and architectural decision-making are shaped in FLOSS.Source: §V · p. 5
• The project expects a conceptual framework linking architecture practices with socio-technical conditions and ethical considerations in FLOSS.Source: §V · p. 5
• Lightweight practices and educational patterns are expected to support explicit, inclusive, and reflective architecture work in FLOSS projects and FLOSS-based teaching.By connecting empirical software engineering with educational design, the project aims to inform researchers and practitioners and complement practitioner-oriented perspectives with a holistic lens on architectural design, quality, and evolution.Source: §V · p. 5

VI. RESEARCH PLAN AND MILESTONES

• The schedule may be refined based on access to projects and early empirical findings.Source: §VI · p. 5
• Table IV outlines the planned timeline and milestones for the research plan.Source: Table IV · p. 5

• Which proxies and analysis strategies can defensibly characterize architectural responsibility and decision-power distribution from FLOSS traces?Source: §VII · p. 5
• Is the planned case pairing and diversity suitable for cross-case synthesis, or should cases be narrowed to one per domain?Source: §VII · p. 5
• Which lightweight interventions are realistic in FLOSS practice and education without increasing maintainer workload, and how should feasibility be assessed?Source: §VII · p. 5