On the Nature of Merge Conflicts: A Study of 2,731 Open Source Java Projects Hosted by GitHub
The DOI matches its Crossref record; 47 of the 48 references it cites were found in a registry or an index.
Abstract
When multiple developers change a software system in parallel, these concurrent changes need to be merged to all appear in the software being developed. Numerous merge techniques have been proposed to support this task, but none of them can fully automate the merge process. Indeed, it has been reported that as much as 10 to 20 percent of all merge attempts result in a merge conflict, meaning that a developer has to manually complete the merge. To date, we have little insight into the nature of these merge conflicts. What do they look like, in detail? How do developers resolve them? Do any patterns exist that might suggest new merge techniques that could reduce the manual effort? This paper contributes an in-depth study of the merge conflicts found in the histories of 2,731 open source Java projects. Seeded by the manual analysis of the histories of five projects, our automated analysis of all 2,731 projects: (1) characterizes the merge conflicts in terms of number of chunks, size, and programming language constructs involved, (2) classifies the manual resolution strategies that developers use to address these merge conflicts, and (3) analyzes the relationships between various characteristics of the merge conflicts and the chosen resolution strategies. Our results give rise to three primary recommendations for future merge techniques, that - when implemented - could on one hand help in automatically resolving certain types of conflicts and on the other hand provide the developer with tool-based assistance to more easily resolve other types of conflicts that cannot be automatically resolved.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Traditional benchmarks focus on human-authored conflicts at scale, such as the 2,731 Java-based projects studied by Ghiotto et al. [18], reporting that nearly 20% of merges require manual intervention”
From AgenticFlict: A Large-Scale Dataset of Merge Conflicts in AI Coding Agent Pull Requests on GitHub (2026) · cites this paper as [18]
Citations
Cites
AnO(ND) difference algorithm and its variations
Myers, E. W. (1986) · Algorithmica · doi:10.1007/bf01840446 · Cited by 893
A state-of-the-art survey on software merging
Mens, T. (2002) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2002.1000449 · Cited by 532 · cited 3×
“A traditional approach to do so is to apply a merge tool, which implements some underlying merge technique that aims to automate as much as possible of the task of combining parallel changes [1].”
BackgroundMethodInfluentialContinuous Integration: Improving Software Quality and Reducing Risk
Duvall, P., Matyas, S., & Glover, A. (2007) · Cited by 510
An Algorithm for Differential File Comparison
Hunt, J. W., & McIlroy, M. D. (2008) · Cited by 296
FASTDash: a visual dashboard for fostering awareness in software teams
Biehl, J. T., Czerwinski, M., Smith, G., & Robertson, G. G. (2007) · Proceedings of the SIGCHI Conference on Human Factors in Computing Systems · doi:10.1145/1240624.1240823 · Cited by 260 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundData consistency for P2P collaborative editing
Oster, G., Urso, P., Molli, P., & Imine, A. (2006) · Proceedings of the 2006 20th anniversary conference on Computer supported cooperative work · doi:10.1145/1180875.1180916 · Cited by 215 · cited 2×
“A significant number of merge techniques rely on lines of code as the basis; these techniques are called unstructured merge techniques [3]–[8].”
BackgroundProactive detection of collaboration conflicts
Brun, Y., Holmes, R., Ernst, M. D., & Notkin, D. (2011) · Proceedings of the 19th ACM SIGSOFT symposium and the 13th European conference on Foundations of software engineering · doi:10.1145/2025113.2025139 · Cited by 201 · cited 5×
“, merge, build, or test failure) [22], [23].”
BackgroundInfluentialMiller, W., & Myers, E. W. (1985) · Software: Practice and Experience · doi:10.1002/spe.4380151102 · Cited by 186 · cited 2×
“, [3], [4], [5], [6], [7], [8]) were unstructured, prioritizing performance and generality over precision.”
BackgroundMethodSemantic Diff: a tool for summarizing the effects of modifications
Jackson, & Ladd (1994) · Proceedings International Conference on Software Maintenance ICSM-94 · doi:10.1109/icsm.1994.336770 · Cited by 185 · cited 2×
“Other, more complex techniques have also been developed, some relying on syntax [9]–[15] and others on semantics [16], [17]; these techniques are termed structured merge techniques.”
BackgroundMethodChacon, S. (2009) · Apress eBooks · doi:10.1007/978-1-4302-1834-0 · Cited by 151
Syde: a tool for collaborative software development
Hattori, L., & Lanza, M. (2010) · Proceedings of the 32nd ACM/IEEE International Conference on Software Engineering - Volume 2 · doi:10.1145/1810295.1810339 · Cited by 135 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundProgram integration for languages with procedure calls
Binkley, D., Horwitz, S., & Reps, T. (1995) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/201055.201056 · Cited by 132 · cited 1×
“[10] introduced a technique that considers the behavior of procedure calls.”
MethodSemi-synchronous conflict detection and resolution in asynchronous software development
Dewan, P., & Hegde, R. M. (2007) · Springer eBooks · doi:10.1007/978-1-84800-031-5_9 · Cited by 127 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundJDiff: A differencing technique and tool for object-oriented programs
Apiwattanapong, T., Orso, A., & Harrold, M. J. (2006) · Automated Software Engineering · doi:10.1007/s10515-006-0002-0 · Cited by 126 · cited 2×
“Other, more complex techniques have also been developed, some relying on syntax [9]–[15] and others on semantics [16], [17]; these techniques are termed structured merge techniques.”
BackgroundMethodStructure-oriented merging of revisions of software documents
Westfechtel, B. (1991) · Software Configuration Management Workshop · doi:10.1145/111062.111071 · Cited by 117 · cited 3×
“Other, more complex techniques have also been developed, some relying on syntax [9]–[15] and others on semantics [16], [17]; these techniques are termed structured merge techniques.”
BackgroundMethodSemistructured merge: rethinking merge in revision control systems
Apel, S., Liebig, J., Brandl, B., Lengauer, C., & Kästner, C. (2011) · Proceedings of the 19th ACM SIGSOFT symposium and the 13th European conference on Foundations of software engineering · doi:10.1145/2025113.2025141 · Cited by 113 · cited 3×
“Hybrid approaches that mix aspects of both unstructured and structured techniques have been explored as well [2], [18], [19].”
BackgroundMethodPalantir: Early Detection of Development Conflicts Arising from Parallel Code Changes
Sarma, A., Redmiles, D. F., & van der Hoek, A. (2012) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2011.64 · Cited by 105 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundStructured merge with auto-tuning: balancing precision and performance
Apel, S., Leßenich, O., & Lengauer, C. (2012) · Proceedings of the 27th IEEE/ACM International Conference on Automated Software Engineering · doi:10.1145/2351676.2351694 · Cited by 94 · cited 3×
“Hybrid approaches that mix aspects of both unstructured and structured techniques have been explored as well [2], [18], [19].”
BackgroundImproving early detection of software merge conflicts
Guimarães, M. L., & Silva, A. R. (2012) · doi:10.5555/2337223.2337264 · Cited by 83
Cassandra: Proactive conflict minimization through optimized task scheduling
Kasi, B. K., & Sarma, A. (2013) · 2013 35th International Conference on Software Engineering (ICSE) · doi:10.1109/icse.2013.6606619 · Cited by 81 · cited 4×
“, merge, build, or test failure) [22], [23].”
BackgroundInfluentialA framework for shared applications with a replicated architecture
Berlage, T., & Genau, A. (1993) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/168642.168668 · Cited by 77 · cited 2×
“A significant number of merge techniques rely on lines of code as the basis; these techniques are called unstructured merge techniques [3]–[8].”
BackgroundBuffenbarger, J. (1995) · Lecture notes in computer science · doi:10.1007/3-540-60578-9_14 · Cited by 76
An Empirical Examination of the Relationship between Code Smells and Merge Conflicts
Ahmed, I., Brindescu, C., Mannan, U. A., Jensen, C., & Sarma, A. (2017) · Proceedings of the ... ACM-IEEE International Symposium on Empirical Software Engineering and Measurement/International Symposium on Empirical Software Engineering and Measurement · doi:10.1109/esem.2017.12 · Cited by 69 · cited 1×
“[29] analyzed 163 projects and 6,979 failed merges, finding that smelly code is three times more likely to be involved in merge conflicts and specifically that method-level smells (e.”
BackgroundEvaluating and improving semistructured merge
Cavalcanti, G., Borba, P., & Accioly, P. (2017) · Proceedings of the ACM on Programming Languages · doi:10.1145/3133883 · Cited by 67 · cited 1×
“[44] used the findings from their previous study to improve their semi-structured merge technique to address conflicts involving import declarations and initialization blocks.”
MethodMaking sense of revision-control systems
O'Sullivan, B. (2009) · Communications of the ACM · doi:10.1145/1562164.1562183 · Cited by 61 · cited 1×
“1 Terminology Distributed version control systems such as Git [4] support concurrent development through both implicit and named branches [31].”
BackgroundImproving early detection of software merge conflicts
Guimarães, M. L., & Silva, A. R. (2012) · 2012 34th International Conference on Software Engineering (ICSE) · doi:10.1109/icse.2012.6227180 · Cited by 56 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundSoftware Practitioner Perspectives on Merge Conflicts and Resolutions
McKee, S., Nelson, N., Sarma, A., & Dig, D. (2017) · 2017 IEEE International Conference on Software Maintenance and Evolution (ICSME) · doi:10.1109/icsme.2017.53 · Cited by 53 · cited 1×
“[43] first interviewed 10 developers and then performed a follow-on survey with 162 other developers to build a detailed understanding of developer perceptions regarding merge conflicts.”
BackgroundIndicators for merge conflicts in the wild: survey and empirical study
Leßenich, O., Siegmund, J., Apel, S., Kästner, C., & Hunsen, C. (2017) · Automated Software Engineering · doi:10.1007/s10515-017-0227-0 · Cited by 50 · cited 1×
“[28] proposed seven indicators for identifying merge conflicts, for example based on the number of commits in a branch, the number of commits in a timespan, and the number of files changed in parallel.”
BackgroundBerzins, V. A. (1994) · ACM Transactions on Programming Languages and Systems · doi:10.1145/197320.197403 · Cited by 49 · cited 2×
“Other, more complex techniques have also been developed, some relying on syntax [9]–[15] and others on semantics [16], [17]; these techniques are termed structured merge techniques.”
BackgroundMethodSafe-commit analysis to facilitate team software development
Wloka, J., Ryder, B. G., Tip, F., & Ren, X. (2009) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1109/icse.2009.5070549 · Cited by 49 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundCassandra: proactive conflict minimization through optimized task scheduling
Kasi, B. K., & Sarma, A. (2013) · Lincoln (University of Nebraska) · doi:10.5555/2486788.2486884 · Cited by 48
Mining Workspace Updates in CVS
Zimmermann, T. (2007) · Fourth International Workshop on Mining Software Repositories (MSR'07:ICSE Workshops 2007) · doi:10.1109/msr.2007.22 · Cited by 48 · cited 2×
“Recent studies report that approximately 10% to 20% of all merges fail [22], [23], with some projects experiencing a failure rate of nearly 50% [22], [24].”
BackgroundUnderstanding semi-structured merge conflict characteristics in open-source Java projects
Accioly, P., Borba, P., & Cavalcanti, G. (2018) · Empirical Software Engineering · doi:10.1007/s10664-017-9586-1 · Cited by 48
Extensible language-aware merging
Hunt, J. J., & Tichy, W. F. (2003) · International Conference on Software Maintenance · doi:10.1109/icsm.2002.1167812 · Cited by 45 · cited 1×
“Hunt and Tichy [12] therefore proposed an extensible, language-aware technique that can deal with renaming and non-local conflicts, though it too has limitations in not identifying behavioral differences caused by dynamic binding, as one example.”
BackgroundTools to Support Global Software Development Processes: A Survey
Portillo-Rodríguez, J., Vizcaíno, A., Ebert, C., & Piattini, M. G. (2010) · IEEE International Conference on Global Software Engineering (ICGSE) · doi:10.1109/icgse.2010.12 · Cited by 44 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundBalancing precision and performance in structured merge
Leßenich, O., Apel, S., & Lengauer, C. (2014) · Automated Software Engineering · doi:10.1007/s10515-014-0151-5 · Cited by 39 · cited 7×
“In attempts to address precision, generality, and performance together, researchers have begun exploring semistructured techniques, combining aspects of both unstructured and structured merge techniques [2], [18], [19].”
BackgroundInfluentialAssessing Semistructured Merge in Version Control Systems: A Replicated Experiment
Cavalcanti, G., Accioly, P., & Borba, P. H. M. (2015) · Proceedings of the ... ACM-IEEE International Symposium on Empirical Software Engineering and Measurement/International Symposium on Empirical Software Engineering and Measurement · doi:10.1109/esem.2015.7321191 · Cited by 36 · cited 2×
“[25] examined 50 and 60 projects, respectively, to compare semi-structured and unstructured merge techniques in terms of how many conflicts they report.”
BackgroundOdyssey-SCM: An integrated software configuration management infrastructure for UML models
Murta, L., Oliveira, H., Dantas, C., Lopes, L. G., & Werner, C. M. L. (2006) · Science of Computer Programming · doi:10.1016/j.scico.2006.05.011 · Cited by 29 · cited 1×
“Building on lines of code as the unit of comparison and resolution [33] enabled these techniques to merge any textual file at a reasonable speed.”
MethodSelf-citationCoDesign: a highly extensible collaborative software modeling framework
Bang, J. Y., Popescu, D., Edwards, G., Medvidović, N., Kulkarni, N., Rama, G. M., & Padmanabhuni, S. (2010) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1145/1810295.1810341 · Cited by 27 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundUnifying Configuration Management with Merge Conflict Detection and Awareness Systems
Estler, H.-C., Nordio, M., Furia, C. A., & Meyer, B. (2013) · Australasian Software Engineering Conference (ASWEC) · doi:10.1109/aswec.2013.32 · Cited by 20 · cited 1×
“CASI [34], CloudStudio [35], CoDesign [36], CollabVS [37], Crystal [22], FastDash [38], Palantír [39], SafeCommit [40], Syde [41], and WeCode [42] all monitor developer workspaces to detect when two developers begin modifying the same file or different files that exhibit some syntactic or semantic dependency.”
BackgroundHow we resolve conflict: an empirical study of method-level conflict resolution
Yuzuki, R., Hata, H., & Matsumoto, K. (2015) · doi:10.1109/swan.2015.7070484 · Cited by 17 · cited 2×
“Others examine the existence of correlations between certain development and code characteristics and resulting merge failures [26]–[30].”
BackgroundA complete textual merging algorithm for software con .guration management systems
Shen, H., & Sun, C. (2004) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/cmpsac.2004.1342846 · Cited by 14 · cited 2×
“Other, more complex techniques have also been developed, some relying on syntax [9], [10], [11], [12], [13], [14], [15] and others on semantics [16], [17]; these techniques are termed structured merge techniques.”
BackgroundMethodNew Initiative: The Naturalness of Software
Dévanbu, P. (2015) · Cited by 9
New Initiative: The Naturalness of Software
Devanbu, P. (2015) · 2015 IEEE/ACM 37th IEEE International Conference on Software Engineering · doi:10.1109/icse.2015.190 · Cited by 9 · cited 1×
“It might even be possible to push it into the realm of the crowd, by building upon the idea that code is regular [32] and that repetitive patterns of change exist.”
BackgroundEvaluating the Branch Merging Effort in Version Control Systems
Santos, R. D. S., & Murta, L. (2012) · Brazilian Symposium on Software Engineering · doi:10.1109/sbes.2012.16 · Cited by 7 · cited 1×
“Others examine the existence of correlations between certain development and code characteristics and resulting merge failures [26]–[30].”
BackgroundSelf-citationSyntax-based Reconciliation for Asynchronous Collaborative Writing
Shen, H., & Sun, C. (2006) · International Conference on Collaborative Computing: Networking, Applications and Worksharing (CollaborateCom) · doi:10.1109/colcom.2005.1651243 · Cited by 6 · cited 2×
“Other, more complex techniques have also been developed, some relying on syntax [9], [10], [11], [12], [13], [14], [15] and others on semantics [16], [17]; these techniques are termed structured merge techniques.”
BackgroundMethodParallelism and conflicting changes in Git version control systems
Nguyen, H. L., & Ignat, C. (2017) · HAL · Cited by 5
Research works printed but not linked: 1
- berlin, 2006, Practical Subversion
Cited by
IntelliMerge: a refactoring-aware software merging technique
Shen, B., Zhang, W., Zhao, H., Liang, G., Jin, Z. G., & Wang, Q. (2019) · Proceedings of the ACM on Programming Languages · doi:10.1145/3360596 · Cited by 64
“Although resolving all conflicts automatically without human intervention is suggested to be infeasible by [Menezes et al. 2018] through a large-scale empirical study, reducing the number of false positive merge conflicts does save developers’ effort and time in the manual resolution process.”
BackgroundUsing pre-trained language models to resolve textual and semantic merge conflicts (experience paper)
Zhang, J., Mytkowicz, T., Kaufman, M., Piskac, R., & Lahiri, S. K. (2022) · Proceedings of the 31st ACM SIGSOFT International Symposium on Software Testing and Analysis · doi:10.1145/3533767.3534396 · Cited by 39
“Recent studies [7, 9] have shown that such a bad merge in the code delayed the development cycle or caused damage by simply leaving bugs in the code.”
BackgroundResultInfluentialPredicting Merge Conflicts in Collaborative Software Development
Owhadi-Kareshk, M., Nadi, S., & Rubin, J. (2019) · Proceedings of the ... ACM-IEEE International Symposium on Empirical Software Engineering and Measurement/International Symposium on Empirical Software Engineering and Measurement · doi:10.1109/esem.2019.8870173 · Cited by 67
“[28] also investigated various characteristics of textual merge conflicts, such as their size and resolution types.”
BackgroundBogomolov, E., Kovalenko, V., Rebryk, Y., Bacchelli, A., & Bryksin, T. (2021) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3468264.3468606 · Cited by 66
“Open-source Git repositories, and GitHub in particular, are a uniquely rich source of data for modern software engineering research efforts [8, 17, 32, 42].”
BackgroundDeepMerge: Learning to Merge Programs
Dinella, E., Mytkowicz, T., Svyatkovskiy, A., Bird, C., Naik, M., & Lahiri, S. K. (2022) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2022.3183955 · Cited by 58
Metadata
- Authors
- Ghiotto, G., Murta, L., Barros, M., & van der Hoek, A.
- Affiliations
- Universidade Federal de Juiz de Fora (BR)
- Universidade Federal Fluminense (BR)
- Venue
- IEEE Transactions on Software Engineering
- Published
- 2020-08-01
- Volume
- 46
- Issue
- 8
- Pages
- 892-915
- ISSN
- 0098-5589 (ISSN-L)
- 1939-3520
- 2326-3881
- OpenAlex
- W2890081359
- Cited by
- 67
- References
- 48
- Citations by year
- 2026: 5 · 2025: 7 · 2024: 9 · 2023: 8 · 2022: 16 · 2021: 9 · 2020: 5
- FWCI
- 8.88
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Software Engineering (publishedVersion)
- Funders
- Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
- Conselho Nacional de Desenvolvimento Científico e Tecnológico
- Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro
- National Science Foundation (CCF-1414197)
- Topics
- Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- Physical Sciences › Computer Science › Information Systems › Software Engineering Techniques and Practices
- Physical Sciences › Computer Science › Computer Science Applications › Open Source Software Innovations
- Keywords
- merge conflicts, conflict resolution strategies, open-source Java projects, GitHub
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.