Mining Q&A Platforms for Empirical Evidence on Quantum Software Programming

TL;DR

This paper presents an empirical study of quantum software programming by mining 6,935 posts from multiple Stack Exchange sites to uncover topics, frameworks, and practitioner challenges. Using Latent Dirichlet Allocation (LDA) on pre-processed posts and qualitative validation, the authors identify 20 topics, highlight popular and difficult areas, and map common challenges into four thematic domains. They also survey tools and frameworks, finding Qiskit, Forest, and PennyLane as the most discussed, and reveal four broad areas where practitioners struggle: fundamental theories, algorithms/applications, experimental practices, and education/community engagement. The work provides actionable insights for researchers and industry, emphasizing education, tooling, and collaborative efforts to advance the quantum software ecosystem toward real-world adoption. The study also outlines future work, including expanding data sources (e.g., GitHub) and incorporating interviews to deepen understanding of practitioners' needs.

Abstract

The rise of quantum computing has driven the need for quantum software engineering, yet its programming landscape remains largely unexplored in empirical research. As quantum technologies advance toward industrial adoption, understanding programming aspects is crucial to addressing software development challenges. This study analyzes 6,935 quantum software programming discussion posts from Stack Exchange platforms (Quantum Computing, Stack Overflow, Software Engineering, and Code Review). Using topic modeling and qualitative analysis, we identified key discussion topics, trends (popular and difficult), tools/frameworks, and practitioner challenges. Twenty topics were identified, including popular ones such as physical theories and mathematical foundations, as well as security and encryption algorithms, while the most difficult were object-oriented programming and parameter control in quantum algorithms. Additionally, we identified nine frameworks that support quantum programming, with Qiskit emerging as the most widely adopted. Our findings also reveal core challenges in quantum software programming, thematically mapped into four areas: theories and mathematical concepts, algorithms and applications, experimental practices and software development, and education and community engagement. This study provides empirical insights that can inform future research, tool development, and educational efforts, supporting the evolution of the quantum software ecosystem.

Figures (5)

• Figure 1: The search, data extraction and filtration process (with the number of posts returned after the step
• Figure 2: Total number of posts per year along with the proportion of questions whose answers were accepted
• Figure 3: Popularity and difficulty of topics
• Figure 4: Distribution of tools and frameworks in quantum software programming
• Figure 5: Thematic mapping of identified challenges