References and Sources

This section lists the sources cited throughout the book. The numbered source index, ordered by source number, is the canonical bibliography: every inline citation of the form (N, author or organisation, year) links to the entry with that number. The index is grouped by category so that readers can locate European legislation, frameworks, incident reports, research on software development, AI-assisted coding, security and work, and practitioner literature quickly.

The bibliography was seeded from the entries of AI Leader (Book 11) and the series entries, with a small number of entries added for the topics of the developer role (Sources 233 to 248). Every seeded entry was opened and checked against the primary text or a bibliographic record in the earlier volume; the entries added for this book were checked against the primary page on 2026-11-11. The categories for software engineering evidence, developer productivity, the secure software supply chain, software law, intellectual property and open source, developer skills and public-sector software are filled by the chapters that need them. Sources added while this book is drafted are written in the reserved range of the chapter that cites them and verified before release, and sources that no chapter cites are removed before the book is merged.

Numbered source index

European legislation and policy

Frameworks, standards and taxonomies

Threat assessments, agency guidance and incident reports

Agents, containment and AI incidents

AI adoption, productivity and automation research

Workforce, skills and automation research

Books and practitioner literature

Leadership, management and organisation research

Decision-making, automation bias and cognition

Work, employment and labour law

Public administration and accountability

The Architecture as Code Series and the author

European legislation and case law

  • Source [172]: European Parliament and Council of the European Union (2019). Directive (EU) 2019/1937 on the protection of persons who report breaches of Union law. Official Journal of the European Union L 305, 26 November 2019. Articles 2(1) (material scope), 5(7) (reporting person), 6(1) (conditions for protection), 8 (internal reporting channels, 50 or more workers), 19 (prohibition of retaliation) and 21 (protection from liability) read in the EUR-Lex text (accessed 2026-11-11). Available at: https://eur-lex.europa.eu/eli/dir/2019/1937/oj Referenced in: Chapter 25: Ethics, Professional Responsibility and the Developer's Judgement, Glossary.
  • Source [173]: Court of Justice of the European Union (2011). Judgment of 1 March 2011 (Grand Chamber) in Case C-236/09, Association belge des Consommateurs Test-Achats ASBL and Others v Conseil des ministres. ECLI:EU:C:2011:100. Holds that Article 5(2) of Directive 2004/113/EC, which allowed sex to be used as an actuarial factor in insurance, is invalid with effect from 21 December 2012. Operative part read on EUR-Lex (CELEX 62009CJ0236) and confirmed against the Fundamental Rights Agency's case-law summary (accessed 2026-11-11). Available at: https://fra.europa.eu/en/caselaw-reference/cjeu-c-23609-judgment Referenced in: Chapter 25: Ethics, Professional Responsibility and the Developer's Judgement.

Company law and director liability

The Architecture as Code Series (Book 11)

Software engineering and AI-assisted development evidence

Developer productivity research

Secure software supply chain

Software law: the Cyber Resilience Act and product liability

Intellectual property and open source

Developer skills and education

  • Source [256]: Ericsson, K. A., Krampe, R. T. & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review 100(3), 363-406. DOI: 10.1037/0033-295X.100.3.363 (accessed 2026-11-11; title, authors, journal and pages confirmed through bibliographic records in search results, the full text was not read and the chapter uses the concept only as vocabulary). Referenced in: Chapter 13: The Junior Developer and the Apprenticeship Problem.

Public-sector software

Professional ethics and codes of conduct

Research papers and studies

  • Source [265]: Yang, C., Shi, Y., Ma, Q., Liu, M. X., Kästner, C. & Wu, T. (2025). What Prompts Don't Say: Understanding and Managing Underspecification in LLM Prompts. arXiv:2505.13360, first submitted 19 May 2025 (preprint; the abstract reports that models infer unspecified requirements by default in 41.1 per cent of cases and that underspecified prompts are twice as likely to regress across model or prompt changes). Available at: https://arxiv.org/abs/2505.13360 (accessed 2026-11-11). Referenced in: Chapter 5: Frame: Problems, Requirements and Intent.
  • Source [266]: Mu, F., Shi, L., Wang, S., Yu, Z., Zhang, B., Wang, C., Liu, S. & Wang, Q. (2023). ClarifyGPT: Empowering LLM-based Code Generation with Intention Clarification. arXiv:2310.10996, submitted 17 October 2023 (preprint; function-level code generation benchmarks, with simulated user answers in the automated evaluation). Available at: https://arxiv.org/abs/2310.10996 (accessed 2026-11-11). Referenced in: Chapter 5: Frame: Problems, Requirements and Intent.

Legacy modernisation and code stewardship

Fairness, bias and accessibility evidence

  • Source [273]: Obermeyer, Z., Powers, B., Vogeli, C. & Mullainathan, S. (2019). Dissecting racial bias in an algorithm used to manage the health of populations. Science 366(6464), 447-453. DOI: 10.1126/science.aax2342. Analysis of a widely used commercial algorithm: at the same risk score Black patients were sicker than White patients, because the algorithm predicted future health-care cost, not illness; the authors estimate that correcting this would raise the share of Black patients receiving extra help from 17.7 to 46.5 per cent. Abstract read through PubMed (accessed 2026-11-11). Referenced in: Chapter 25: Ethics, Professional Responsibility and the Developer's Judgement.
  • Source [274]: WebAIM (2026). The WebAIM Million: An annual accessibility analysis of the top 1,000,000 home pages. Report of February 2026: automated analysis of 1,000,000 home pages found detectable WCAG 2 failures on 95.9 per cent, an average of 56.1 detected errors per page, and low-contrast text (83.9 per cent), missing alternative text (53.1 per cent) and missing form labels (51 per cent) as the most common failures. Automated detection covers only part of the failures a person would find (accessed 2026-11-11). Available at: https://webaim.org/projects/million/ Referenced in: Chapter 25: Ethics, Professional Responsibility and the Developer's Judgement.

AI-assisted coding

Developer experience, identity and wellbeing

Learning, practice and expertise

How to cite

Inline citations give the source number, the author or organisation, and the year, for example (9, MITRE, n.d.), and link to the entry here. Where an author is named in the sentence, the citation carries the year. Each chapter closes with a ## Sources section that lists the works it cites, using the same numbering as this bibliography. New sources are appended to the appropriate category with the next free number; existing numbers are never reused.