← Reference notes

Summary

Lopes, Accioly, Borba, and Menezes compare two ways teams integrate work in Git. Branch-based workflows keep several remote branches with set roles, such as feature branches, an integration branch, and a main branch, and move changes toward the shared branch as the code matures. They name GitFlow as the best-known example. Trunk-based workflows keep a single remote branch, and developers integrate there directly. The evidence is semi-structured interviews and a survey of software developers in Brazil. The abstract reports that trunk-based development fitted fast projects with smaller, more experienced teams, and that branch-based development fitted larger, less experienced teams, despite management challenges. The abs page lists no conference. Comments say 11 pages with 3 figures. This note follows that abstract page only. It is not the manual for the commit or push commands.

When to apply

• A team is choosing how commits land: one shared branch, or a ladder of branches.
• Someone needs the meaning of git commit, the flags, or how to push. Use the Git documentation for that. This paper is about the workflow around those commands.

Costs and limits

• The findings are the authors' reading of interviews and a survey of Brazilian developers. They are not a universal rule for every team.
• The abstract does not report sample size. This note does not invent one and does not re-check the PDF.

Evidence (detail)

• The abs page (v1, 11 Jul 2025) defines branch-based workflows as several remote branches with roles, with GitFlow as the popular example, and trunk-based workflows as one remote branch where developers integrate directly (Abstract).
• Methods named on the abstract: semi-structured interviews and a survey (Abstract).
• Reported fit: trunk-based development for fast-paced projects with experienced and smaller teams; branch-based development for less experienced and larger teams, despite management challenges (Abstract).
• Comments line: "11 pages with 3 figures". No journal reference on the page.

Facts

• id: F1 claim: The paper contrasts trunk-based Git workflows, which integrate on one remote branch, with branch-based workflows, which move work through several remote branches such as feature, integration, and main. confidence: medium date: 2026-10-18 sources: [S1 Abstract] license: CC BY 4.0
• id: F2 claim: From interviews and a survey of developers in Brazil, the abstract reports that trunk-based development fitted fast projects with smaller experienced teams, and branch-based development fitted larger less experienced teams, despite management challenges. confidence: medium date: 2026-10-18 sources: [S1 Abstract] license: CC BY 4.0
• id: F3 claim: The arXiv abs page shows CC BY 4.0 on v1 (11 Jul 2025) and lists no conference or journal. confidence: high date: 2026-10-18 sources: [S1 abs page] license: CC BY 4.0

Related

• The library miss is "commit git make push". This note covers which branching workflow a team uses when those commits land. It does not replace the Git command reference.

Review notes

• Staging draft. Paraphrase of the arXiv abstract page checked 2026-10-18. The PDF was not re-read.
• License on the abs page: icon by-4.0.png, link text "view license", http://creativecommons.org/licenses/by/4.0/.
• Fresh-eyes recheck 2026-10-18: the abstract says "management challenges", not a coordination or management cost. Licence, authors, and the v1 date unchanged.

@misc{iil_note_yegmqj02,
  author = {{iinventedlight Library}},
  title = {{Choosing the Right Git Workflow: A Comparative Analysis of Trunk-based vs. Branch-based Approaches}},
  howpublished = {iinventedlight Library},
  year = {2026},
  date = {2026-10-18},
  url = {https://iinventedlight.com/library/reference/arxiv/2507-08943-choosing-the-right-git-workflow/},
  note = {Permanent ID iil:note:yegmqj02. License: CC-BY-4.0. Summary of https://arxiv.org/abs/2507.08943.},
}

For language models and agents: start at /llms.txt. This page in Markdown: /library/reference/arxiv/2507-08943-choosing-the-right-git-workflow.md.