Stop writing software like it's 1999! - Hannes Lowette - NDC Copenhagen 2026

Stop writing software like it's 1999! - Hannes Lowette - NDC Copenhagen 2026

🎙 Hannes Lowette 👥 227K 📅 July 3, 2026 ⏱ 57 min 👁 8K 📄 expert opinion 🧭 2026-08-13
Available in: English (current) Français

Keywords

layered architecturemicroservicesmodular monolithCI/CDsoftware evolution

Summary

Hannes Lowette, a software engineer with 20 years of experience, argues that many common software development practices are outdated and based on constraints that no longer exist. He identifies three main pitfalls: layered architectures, microservices, and lack of CI/CD. Layered architectures, while promoting code reuse and separation of concerns, lead to tightly coupled code and context switching, especially in agile environments where software is continuously modified. He suggests organizing code by feature rather than by layer, which improves developer productivity and facilitates AI-assisted development via better context management. Microservices, though beneficial for large organizations, are often overused, causing operational overhead and context switching; he advocates for modular monoliths as a default, which allow independent modules within a single deployable unit, with the option to split later. Finally, he emphasizes the importance of CI/CD from day one, enabling continuous delivery and feedback, and notes that automation investments pay off quickly. He concludes that while AI is changing the landscape, these principles remain relevant, and developers who excel at understanding business problems will remain valuable.

173 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights into common software architecture pitfalls and offers practical alternatives. The speaker’s argumentation is clear and structured, using relatable analogies and personal anecdotes. He effectively contrasts the historical context of these practices with modern development realities, making a compelling case for change. However, the arguments are largely based on anecdotal evidence and personal experience rather than empirical data, which limits their generalizability. The suggestion to use modular monoliths is well-reasoned and aligns with current industry trends, but the talk could benefit from more concrete examples or case studies.

Scientific Rigor, Source Quality, Title Accuracy

The talk is an expert opinion without formal citations, but it references well-known concepts and practices in software engineering. The speaker mentions Sam Newman’s book on microservices and the concept of service-oriented architecture, but does not provide specific sources. The title accurately reflects the content, and the talk is coherent and well-structured. The lack of external references is a minor weakness, but the speaker’s credibility and practical experience lend weight to his arguments.

180 words

Title / Content Match

The title accurately reflects the content: the speaker argues that many current software practices are outdated and should be modernized.

Quality & Reliability

7/10

The talk is an expert opinion based on two decades of industry experience, with practical examples and references to established concepts (layered architecture, microservices, modular monoliths, CI/CD). It lacks formal citations or empirical data, but the arguments are coherent and align with common software engineering practices.

Key Moments

Cited Sources

  • NDC Conferences — Conference website where the talk was recorded.
  • NDC Copenhagen — Specific conference event page.

Concurring Sources

  • Sam Newman - Building Microservices — Referenced in the talk as a key book on microservices.

Contribution & Novelties

The talk offers a practical, experience-based critique of common software architecture choices, advocating for modular monoliths and feature-based organization. It uniquely ties these recommendations to the age of AI, emphasizing context management for LLMs. The speaker’s perspective as a consultant provides real-world insights.

Pour aller plus loin :

73 words

Radar Profile

The radar profile shows a balanced talk with strong scores in information quantity and quality, moderate technical depth, and good reliability. The speaker provides a comprehensive overview of software architecture pitfalls and solutions, making it accessible to a broad technical audience.

Reliability 7/10

💬 Sur les 0 commentaires analysés, aucune tendance n'est disponible.