A Large-Scale Multi-Dimensional Analysis of Developer Information Needs

Building software systems is a complex task. Teams of developers must gather and exchange large quantities of technical and conceptual information to build reliable systems. This transfer of knowledge is mainly supported by tangible documents that contain information deemed important for different contexts. However, determining what information is important is itself challenging. The importance should be based on the information that developers need. Prior studies have revealed several typical needs, but there is still a considerable variation in the findings of different studies. In this project, we will attempt to reconcile these prior findings to establish a robust, overarching, and actionable framework of developer information needs. We will systematically review the context of prior studies, including the original researchers’ perspective and the context in which developers expressed the needs. We will also attempt to replicate the methodology of prior studies on a larger, heterogeneous data corpus. This investigation will contribute a more holistic understanding of developer needs, to indicate not only what information developer needs, but in which context they need it. The outcome of this research should help improve the quality of software documentation, but also provide a foundation to build a theory of developer needs.

Faculty Supervisor:

Martin Robillard

Student:

Partner:

Universität Hamburg

Discipline:

Computer science

Sector:

Education

University:

McGill University

Program: