Debugging by asking questions about program output
Amy J. Ko
Abstract
Amy J. Ko
Abstract
One reason debugging is the most time-consuming part of software development is because developers struggle to map their questions about a program's behavior onto debugging tools' limited support for analyzing code. Interrogative debugging is a new debugging paradigm that allows developers to ask questions directly about their programs' output, helping them to more efficiently and accurately determine what parts of the system to understand. An interrogative debugging prototype called the Whyline is described, which has been shown to reduce debugging time by a factor of eight. Several extensions and generalizations to it are proposed, including plans for evaluating their effectiveness.
OpenAlex reports 10 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
One reason debugging is the most time-consuming part of software development is because developers struggle to map their questions about a program's behavior onto debugging tools' limited support for analyzing code. Interrogative debugging is a new debugging paradigm that allows developers to ask questions directly about their programs' output, helping them to more efficiently and accurately determine what parts of the system to understand. An interrogative debugging prototype called the Whyline is described, which has been shown to reduce debugging time by a factor of eight. Several extensions and generalizations to it are proposed, including plans for evaluating their effectiveness.
Key concepts: Debugging, Algorithmic program debugging, Computer science, Programming language, Interrogative, Software bug, Software engineering, Software