Prevalence of confusing code in software projects

Gopstein, Dan; Zhou, Henry H.; Frankl, Phyllis G.; Cappos, Justin

doi:10.1145/3196398.3196432

Cited by 30 publications

(23 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thus we answer RQs 1, 3, and 4 affirmatively, except for models of parentheses removal. We speculate that this may be the result of less consistent style around the usage of parentheses, similar to what Gopstein et al (2017) and Gopstein, Zhou, Frankl, and Cappos (2018) found with bracket usage. We further discuss the influence of confusing code and style guidelines in Section 6.…”

Section: Parentheses Transformationssupporting

confidence: 82%

“…Moreover, as we mentioned, many of our transformations are not covered under explicit style guides and sometimes style guides can be conflicting. We were inspired to try the parentheses transforms based on the work of Gopstein et al (2017Gopstein et al ( , 2018 investigating code patterns that developers find confusing. They found that style guides advocating for minimal curly brace use -for example, when writing if statements with only 1 statement in the body, developers can choose to wrap the body in curly braces or not -sometimes conflicted with what developers found easiest to understand.…”

Section: Threats and Generalizabilitymentioning

confidence: 99%

See 1 more Smart Citation

Do Programmers Prefer Predictable Expressions in Code?

et al. 2020

View full text Add to dashboard Cite

Source code is a form of human communication, albeit one where the information shared between the programmers reading and writing the code is constrained by the requirement that the code executes correctly. Programming languages are more syntactically constrained than natural languages, but they are also very expressive, allowing a great many different ways to express even very simple computations. Still, code written by developers is highly predictable, and many programming tools have taken advantage of this phenomenon, relying on language model surprisal as a guiding mechanism. While surprisal has been validated as a measure of cognitive load in natural language, its relation to human cognitive processes in code is still poorly understood. In this paper, we explore the relationship between surprisal and programmer preference at a small granularity—do programmers prefer more predictable expressions in code? Using meaning‐preserving transformations, we produce equivalent alternatives to developer‐written code expressions and run a corpus study on Java and Python projects. In general, language models rate the code expressions developers choose to write as more predictable than these transformed alternatives. Then, we perform two human subject studies asking participants to choose between two equivalent snippets of Java code with different surprisal scores (one original and transformed). We find that programmers do prefer more predictable variants, and that stronger language models like the transformer align more often and more consistently with these preferences.

show abstract

Section: Parentheses Transformationssupporting

confidence: 82%

Section: Threats and Generalizabilitymentioning

confidence: 99%

Do Programmers Prefer Predictable Expressions in Code?

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Authors describe 19 atoms of confusion, on which this research is based on. To show the real-world relevance of these selected atoms of confusion, follow up research from Gopstein et al [16] shows that the 15 atoms that were proven to be confusing, occur in practice once per 23 lines. Their research is based on the analysis of 14 of the most popular and influential C and C++ software projects.…”

Section: Background and Related Workmentioning

confidence: 99%

Atoms of Confusion in Java

Langhout

Aniche

2021

2021 IEEE/ACM 29th International Conference on Program Comprehension (ICPC)

View full text Add to dashboard Cite

Although writing code seems trivial at times, problems arise when humans misinterpret what the code actually does. One of the potential causes are "atoms of confusion", the smallest possible patterns of misinterpretable source code. Previous research has investigated the impact of atoms of confusion in C code. Results show that developers make significantly more mistakes in code where atoms are present. In this paper, we replicate the work of Gopstein et al. to the Java language. After deriving a set of atoms of confusion for Java, we perform a two-phase experiment with 132 computer science students (i.e., novice developers). Our results show that participants are 2.7 up to 56 times more likely to make mistakes in code snippets affected by 7 out of the 14 studied atoms of confusion, and when faced with both versions of the code snippets, participants perceived the version affected by the atom of confusion to be more confusing and/or less readable in 10 out of the 14 studied atoms of confusion.Index Terms-software engineering, program comprehension, atoms of confusion, empirical software engineering.2 https://gcc.gnu.org/bugzilla/show bug.cgi?id=90885 3 https://gcc.gnu.org/bugzilla/show bug.cgi?id=90885#c6

show abstract

“…Em [9], Gopstein et al apontaram 15 átomos que, quando presentes no código, causam confusão de forma significativa. Em um trabalho complementar, Gopstein et al [10] observou uma forte relação entre linhas de código contendo átomos de confusão e a ocorrência de bugs, mostrando que commits do tipo bug-fix removiam mais átomos quando comparados com outros tipos de commits. Da mesma forma, foi observado que essas linhas causavam mais confusão, pois costumavam ser mais comentadas que outras.…”

Section: áTomos De Confusãounclassified

BOHR - Uma Ferramenta para a Identificação de Átomos de Confusão em Códigos Java

Mendes

Viana

Rocha

2021

Anais Do IX Workshop De Visualização, Evolução E Manutenção De Software (VEM 2021)

View full text Add to dashboard Cite

A atividade de compreensão do código-fonte é fundamental no desenvolvimento e manutenção de software. Nesse contexto, Átomos de Confusão (AC) surgem como a menor porção de código capaz de causar confusão em desenvolvedores nesse processo. Neste artigo, é apresentado o BOHR - The Atoms of Confusion Hunter, uma ferramenta que tem como objetivos: (i) auxiliar a identificação de AC em sistemas Java; (ii) fornecer relatórios sobre a prevalência desses AC; e (iii) fornecer uma API para o desenvolvimento de novos buscadores customizados para a captura de novos AC e, também, melhorias em suas identificações. Nesta primeira versão, o BOHR é capaz de detectar 8 dos 13 tipos de AC apontados por Langhout e Aniche. Uma avaliação preliminar do BOHR foi conduzida em três sistemas reais, de código aberto, largamente adotados pela comunidade de desenvolvimento de softwares Java (Picasso, greenDAO e Socket.IO-client Java). Os resultados iniciais mostraram que o BOHR conseguiu detectar com precisão os AC existentes nesses sistemas analisados, apontando de forma correta os snippets dos AC, assim como o seu tipo, nome da classe que era pertencente e o número da linha de código de sua ocorrência. No total, foram encontrados 105 AC, sendo 37 no Picasso, 25 no greenDAO e 43 no Socket.IO-client Java.

show abstract

Prevalence of confusing code in software projects

Cited by 30 publications

References 20 publications

Do Programmers Prefer Predictable Expressions in Code?

Do Programmers Prefer Predictable Expressions in Code?

Atoms of Confusion in Java

BOHR - Uma Ferramenta para a Identificação de Átomos de Confusão em Códigos Java

Contact Info

Product

Resources

About