ABSTRACT
In software development, bug reports provide crucial information to developers. However,these reports widely differ in their quality. We conducted a survey among developers and users of APACHE, ECLIPSE, and MOZILLA to find out what makes a good bug report. The analysis of the 466 responses revealed an information mismatch between what developers need and what users supply. Most developers consider steps to reproduce, stack traces, and test cases as helpful, which are at the same time most difficult to provide for users. Such insight is helpful to design new bug tracking tools that guide users at collecting and providing more helpful information. Our CUEZILLA prototype is such a tool and measures the quality of new bug reports; it also recommends which elements should be added to improve the quality. We trained CUEZILLA on a sample of 289 bug reports,rated by developers as part of the survey. In our experiments, CUEZILLA was able to predict the quality of 31–48% of bug reports accurately.
INTRODUCTION
Bug reports are vital for any software development. They allow users to inform developers of the problems encountered while using a software. Bug reports typically contain a detailed description of a failure and occasionally hint at the location of the fault in the code (in form of patches or stack traces). However, bug reports vary in their quality of content; they often provide inadequate or incorrect information. Thus, developers sometimes have to face bugs with descriptions such as “Sem Web” (APACHE bug COCOON-1254), “wqqwqw” (ECLIPSE bug #145133), or just “GUI” with comment “The page is too clumsy” (MOZILLA bug #109242). It is no surprise that developers are slowed down by poorly written bug reports
because identifying the problem from such reports takes more time.In this paper, we investigate the quality of bug reports from the perspective of developers. We expected several factors to impact the quality of bug reports such as the length of descriptions, formatting, and presence of stack traces and attachments (such as screenshots). To find out which matter most, we asked 872 developers from the APACHE, ECLIPSE, and MOZILLA projects to:
1. Complete a survey on important information in bug reports and the problems they faced with them. We received a total of 156 responses to our survey (Section 2 and 3).
2. Rate the quality of bug reports from very poor to very good onafive-point Likert scale[22]. We received a total of 1,186 votes for 289 randomly selected bug reports (Section 4).
In addition, we asked 1,354 reporters1 from the same projects to complete a similar survey,out of which 310 responded. The results of both surveys suggest that there is a mismatch between what developers consider most helpful and what users provide. To enable swift fixing of bugs, this mismatch should be bridged, for example with tool support for reporters to furnish information that developers want. We developed a prototype tool called CUEZILLA (seeFigure1),which gauges the quality of bug reports and suggests to reporters what should be added to make a bug report better.
1. CUEZILLA measures the quality of bug reports. We trained and evaluated CUEZILLA on the 289 bug reports rated by the developers (Section 5).
2. CUEZILLA provides incentives to reporters. We automatically mined the bug databases for encouraging facts such as “Bug reports with stack traces are fixed sooner” (Section 6).
ABSTRACT In software development, bug reports provide crucial information to developers. However,these reports widely differ in their quality. We conducted a survey among developers and users of APACHE, ECLIPSE, and MOZILLA to find out what makes a good bug report. The analysis of the 466 responses revealed an information mismatch between what developers need and what users supply. Most developers consider steps to reproduce, stack traces, and test cases as helpful, which are at the same time most difficult to provide for users. Such insight is helpful to design new bug tracking tools that guide users at collecting and providing more helpful information. Our CUEZILLA prototype is such a tool and measures the quality of new bug reports; it also recommends which elements should be added to improve the quality. We trained CUEZILLA on a sample of 289 bug reports,rated by developers as part of the survey. In our experiments, CUEZILLA was able to predict the quality of 31–48% of bug reports accurately.INTRODUCTION Bug reports are vital for any software development. They allow users to inform developers of the problems encountered while using a software. Bug reports typically contain a detailed description of a failure and occasionally hint at the location of the fault in the code (in form of patches or stack traces). However, bug reports vary in their quality of content; they often provide inadequate or incorrect information. Thus, developers sometimes have to face bugs with descriptions such as “Sem Web” (APACHE bug COCOON-1254), “wqqwqw” (ECLIPSE bug #145133), or just “GUI” with comment “The page is too clumsy” (MOZILLA bug #109242). It is no surprise that developers are slowed down by poorly written bug reportsbecause identifying the problem from such reports takes more time.In this paper, we investigate the quality of bug reports from the perspective of developers. We expected several factors to impact the quality of bug reports such as the length of descriptions, formatting, and presence of stack traces and attachments (such as screenshots). To find out which matter most, we asked 872 developers from the APACHE, ECLIPSE, and MOZILLA projects to: 1. Complete a survey on important information in bug reports and the problems they faced with them. We received a total of 156 responses to our survey (Section 2 and 3). 2. Rate the quality of bug reports from very poor to very good onafive-point Likert scale[22]. We received a total of 1,186 votes for 289 randomly selected bug reports (Section 4). In addition, we asked 1,354 reporters1 from the same projects to complete a similar survey,out of which 310 responded. The results of both surveys suggest that there is a mismatch between what developers consider most helpful and what users provide. To enable swift fixing of bugs, this mismatch should be bridged, for example with tool support for reporters to furnish information that developers want. We developed a prototype tool called CUEZILLA (seeFigure1),which gauges the quality of bug reports and suggests to reporters what should be added to make a bug report better.
1. CUEZILLA measures the quality of bug reports. We trained and evaluated CUEZILLA on the 289 bug reports rated by the developers (Section 5).
2. CUEZILLA provides incentives to reporters. We automatically mined the bug databases for encouraging facts such as “Bug reports with stack traces are fixed sooner” (Section 6).
การแปล กรุณารอสักครู่..
