Growing & Thriving in a Multi-model World

Growing & Thriving in a Multi-model World

🎙 Alberto Brandolini 👥 1.1M 📅 June 17, 2026 ⏱ 48 min 👁 714 📄 expert opinion 🧭 2026-08-02
Available in: English (current) Français

Keywords

bounded contextmodel decompositionsocio-technical stackubiquitous languagedomain-driven design

Summary

Alberto Brandolini, creator of EventStorming, presents a talk on the challenges and strategies for managing multiple models in software development. He begins by illustrating the common pitfalls of a single monolithic model, which often leads to a ‘big ball of mud’ due to lack of boundaries and delayed feedback. He shares his experience with a long-running project that successfully used bounded contexts from the start, highlighting the benefits of stable evolution costs and the importance of discipline. Brandolini emphasizes that discipline is not free and requires conscious effort to avoid model rot. He discusses the need to detect signals for splitting models, such as when a model serves multiple purposes or when discussions become ambiguous. He advocates for visual thinking and mapping to ground conversations about boundaries. The talk covers various strategies for decomposition, including considering the socio-technical stack, and stresses that different approaches fit at different stages of software evolution. He concludes with practical advice on how to start splitting models and the importance of continuous design effort.

169 words

Critical Evaluation

The talk offers a valuable perspective on managing complexity in software design, drawing from the speaker’s extensive experience in Domain-Driven Design. Brandolini’s argumentation is coherent and grounded in practical examples, particularly his long-running project that successfully employed bounded contexts. He effectively communicates the dangers of model rot and the necessity of continuous design effort. The emphasis on the socio-technical stack, including team dynamics and organizational forces, adds depth to the discussion. However, the talk relies heavily on anecdotal evidence and personal experience rather than empirical research, which limits its generalizability. The speaker acknowledges this by mentioning selection bias in his consulting work. The sources cited, such as Eric Evans’ ‘Domain-Driven Design’ and Charles Duhigg’s ‘The Power of Habit’, are reputable and relevant, but the talk does not provide a systematic review of literature. The adéquation between title and content is strong, as the talk indeed focuses on growing and thriving in a multi-model world. The technical level is intermediate, suitable for practitioners familiar with DDD concepts. The talk’s strength lies in its practical insights and actionable advice, such as the ‘discipline isn’t free’ concept and the use of visual mapping. However, it could benefit from more concrete examples of splitting models and addressing potential counterarguments. Overall, the talk is informative and thought-provoking, but its reliance on personal experience and lack of empirical backing prevent it from being a definitive scientific resource.

231 words

Title / Content Match

The title accurately reflects the content, which focuses on managing multiple models in software development.

Quality & Reliability

8/10

The talk is based on the speaker's extensive experience and references established practices like Domain-Driven Design and bounded contexts. The argumentation is coherent and grounded in practical examples, though it relies on anecdotal evidence rather than systematic research.

Chapters

Cited Sources

Concurring Sources

External References

Contribution & Novelties

The talk provides a practical framework for managing multiple models in software development, emphasizing the importance of bounded contexts and continuous design effort. It introduces the concept of ‘discipline isn’t free’ and highlights the need to detect signals for model splitting. The speaker shares insights from a long-running project, offering a real-world example of successful multi-model architecture.

Pour aller plus loin :

127 words

Radar Profile

The radar profile shows high scores in quantity and quality of information, with moderate technical level and reliability. This indicates a talk rich in practical insights but with limited empirical backing, reflecting the speaker's experience-based approach.

Reliability 7/10