Balancing Coupling in Software Design • Vlad Khononov & Sheen Brisals • GOTO 2025

YouTube Video

Summary of the Transcript:

This transcript captures a GOTO Book Club interview between Sheen Brisals and Vlad Khononov, author of the new book “Balancing Coupling in Software Design.” The conversation revolves around the core concept of coupling in software design, its historical relevance, and its application in modern architectures like microservices and cloud environments.

Key Discussion Points:

• Introduction of Vlad Khononov: Sheen introduces Vlad, highlighting his previous popular book “Learning Domain-Driven Design” and his extensive experience as a software engineer, architect, speaker, and author. Vlad briefly introduces himself, emphasizing his passion for software engineering and his pride in his two books.
• The Genesis of “Balancing Coupling in Software Design”: The conversation delves into why Vlad chose to write a book on coupling, a seemingly old and fundamental concept. Vlad recounts his first encounter with coupling and cohesion at a Software Craftsmanship meetup in Tel Aviv around 2006-2007. While he grasped coupling, cohesion remained enigmatic.
• Microservices Project Failure and the ’70s Revelation: Vlad shares a pivotal experience: a disastrous microservices project. Despite following best practices from conferences and books, the project failed, leading him to research software design principles. This research led him to a book from the 1970s, “Structured Design,” specifically a chapter on coupling. This chapter profoundly changed his perspective on software design. He realized that the problems addressed in the ’70s are remarkably similar to modern software engineering challenges, despite vastly different technologies.
• Coupling in Modern vs. Older Architectures: Sheen raises the point that coupling was easier to understand in older, homegrown systems compared to today’s complex landscapes with microservices, cloud services, and SaaS providers. Vlad emphasizes that the fundamental principles of coupling remain relevant across all levels of abstraction, from assembly code to distributed systems. He argues that understanding these principles is crucial for navigating the complexity of modern software.
• Decoupling, Loose Coupling, and the Analogy of Planets: Sheen mentions the common terms “decoupling” and “loose coupling” and appreciates Vlad’s analogy of planets connected by gravitational force. This analogy effectively illustrates that coupling is inherent, even in seemingly independent systems, and it’s about managing rather than eliminating it.
• Measurements of Coupling and Practical Application: Sheen inquires about the measurements of coupling introduced in the book and their practical application. Vlad confirms that all principles in the book are battle-tested in his work and have evolved through iterations. He mentions the “integration strength model” and how it was refined based on feedback, including input from Vaughn Vernon. He describes the evolution of the model, including the removal of “logical coupling” and the renaming of “implementation coupling” to “intrusive coupling.”
• Coupling and Estimation: Sheen explores the potential of Vlad’s coupling model for improving software estimations, drawing parallels to function point analysis and story points. Vlad acknowledges that he hadn’t explicitly considered this application but recognizes its potential value in reducing uncertainty and improving estimation accuracy.
• Complexity: Local vs. Global: The discussion shifts to complexity, specifically local and global complexity, concepts also originating from “Structured Design.” Vlad explains how his failed microservices project exemplified optimizing local complexity (simpler individual services) while drastically increasing global complexity (system-wide interactions and dependencies). He emphasizes the importance of considering both types of complexity in software design. He connects these concepts to modern architectures, relating accumulated complexity to monoliths and distributed complexity to cloud-based systems.
• Modularity and Microservices Failure: Sheen brings up modularity in relation to the failed microservices project. Vlad attributes the failure to a flawed decomposition principle focused solely on minimizing local complexity. He explains that changes ended up requiring modifications and deployments across multiple services, negating the supposed benefits of microservices. He highlights the value of modularity in reducing accidental complexity and improving system extendability.
• Modularity and Cloud Architectures: Sheen questions if different patterns are needed for modern cloud architectures. Vlad reiterates that understanding fundamental coupling principles is key, regardless of architecture (monolith or microservices). He mentions the historical trend of adding abstraction layers to manage coupling, suggesting potential future abstractions beyond microservices. He connects event-driven architectures to the need for awareness of these principles in architecting interconnected systems.
• Vlad the Poet and the Book’s Poems: Sheen shifts to Vlad’s “poet” side, noting the unique inclusion of poems at the start of each chapter. Vlad shares the story behind the poems, starting with experimenting with early OpenAI text completion APIs. Initially aiming for funny poems with ChatGPT, he received feedback that they were often “lame.” He then redefined the poems’ purpose to provide chapter overviews, collaborating with ChatGPT and investing significant effort to create meaningful and insightful poems.
• Book Availability and Concluding Message: The conversation concludes with information on where to find “Balancing Coupling in Software Design” (Amazon, informit.com, O’Reilly online learning platform). Vlad’s final message emphasizes the importance of moving away from complexity towards modularity. He defines complexity as uncertainty about action outcomes due to cognitive load from system interactions. Modularity, in contrast, reduces cognitive load, making changes predictable and locatable. He reiterates that managing coupling is the key to achieving modularity and reducing complexity.

Accuracy:

The information provided in the transcript is highly accurate in regards to established knowledge within software engineering and design principles.

• Coupling and Cohesion: The transcript accurately portrays coupling and cohesion as fundamental concepts in software design, echoing their long-standing importance in the field.
• Structured Design and Glenford Myers: The reference to “Structured Design” by Glenford Myers (and likely Wayne Stevens and Larry Constantine, though not explicitly mentioned) from the 1970s as a source for coupling principles is correct. This book is indeed foundational in the development of structured programming and design methodologies, and coupling is a central concept within it.
• Microservices Challenges: The described failure of the microservices project due to focusing on local complexity at the expense of global complexity is a well-documented anti-pattern and a common pitfall in microservices adoption. This is a realistic and accurate depiction of challenges often encountered.
• Modularity and Complexity Reduction: The transcript correctly links modularity to reduced complexity, improved maintainability, and better system understanding. This is a core tenet of good software design and is widely accepted.
• Local vs. Global Complexity: The distinction between local and global complexity is accurate and useful for analyzing system architecture, particularly in distributed systems like microservices. This concept helps explain why decomposing a system into smaller parts doesn’t automatically reduce overall complexity.
• Vlad Khononov’s Expertise: Vlad Khononov is indeed a recognized author and speaker in the domain-driven design community, and his expertise on software design principles, including coupling, is well-established.
• Vaughn Vernon’s Feedback: The anecdote about Vaughn Vernon’s feedback on the “integration strength model” is plausible and aligns with Vaughn Vernon’s known expertise in domain-driven design and software architecture.

Overall, the transcript presents a sound and accurate discussion of coupling, complexity, and modularity in software design, grounded in established principles and practical experiences.

Resources:

Here are the top 5 most relevant resources to learn more about the subjects presented in the transcript:

1. “Balancing Coupling in Software Design” by Vlad Khononov (2024): This is the primary resource, being the central topic of the interview. The book likely provides a comprehensive and practical guide to understanding and managing coupling in software systems, incorporating Vlad’s insights and the “integration strength model.” This is the most direct and relevant resource for deepening understanding of the concepts discussed.

   • Relevance: Directly addresses the core topic of the interview, providing in-depth knowledge and practical techniques for balancing coupling.

2. “Structured Design” by Edward Yourdon and Larry L. Constantine (1979) / “Composite/Structured Design” by Glenford J. Myers (1978): These books (or similar works from the era) are foundational texts on structured design methodologies and are repeatedly referenced in the transcript as the source of concepts like coupling, cohesion, local and global complexity. Reading these older texts provides the historical context and theoretical basis for these principles. While potentially dated in terms of technology examples, the core design principles remain highly relevant.

   • Relevance: Provides the historical and theoretical foundation for the concepts discussed, especially coupling and complexity.

3. “Learning Domain-Driven Design” by Vlad Khononov (2021): While focused on Domain-Driven Design, this book is relevant because DDD principles strongly emphasize modularity, bounded contexts, and managing dependencies (coupling) within complex domains. Understanding DDD can provide a valuable framework for applying coupling principles in real-world software projects, particularly when dealing with microservices and complex systems.

   • Relevance: Offers a framework (DDD) that promotes modularity and helps manage coupling within complex software domains, complementing the book on coupling.

4. “Building Microservices: Designing Fine-Grained Systems” by Sam Newman (O’Reilly): Given the discussion about microservices failures and the importance of coupling in distributed systems, a book specifically on microservices architecture is highly relevant. Sam Newman’s book is a widely respected and practical guide to designing and building microservices effectively, addressing topics like service boundaries, communication patterns, and managing dependencies (coupling) in a distributed context.

   • Relevance: Provides practical guidance on applying coupling principles specifically in the context of microservices architecture, addressing the challenges highlighted in the transcript.

5. “Clean Architecture: A Craftsman’s Guide to Software Structure and Design” by Robert C. Martin (“Uncle Bob”) (Prentice Hall): This book offers a broader perspective on software architecture and design principles, emphasizing modularity, separation of concerns, and dependency inversion. It provides a framework for creating maintainable and flexible software systems, which directly relates to the goals of managing coupling and complexity effectively. While not solely focused on coupling, it provides valuable architectural principles that support good coupling practices.

   • Relevance: Offers broader architectural principles that underpin good software design, including aspects of modularity and dependency management, which are crucial for managing coupling effectively in any system.